\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Eva Gonzalez Romo v. William Barr\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-14\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: Yes\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nEVA ISABEL GONZALEZ ROMO, Eva No. 16-71559\nAKA Eva Isabel Romo,\n Petitioner, Agency No.\n A047-334-955\n v.\n\nWILLIAM P. BARR, Attorney General, OPINION\n Respondent.\n\n On Petition for Review of an Order of the\n Board of Immigration Appeals\n\n Argued and Submitted July 8, 2019\n Portland, Oregon\n\n Filed August 14, 2019\n\n Before: Ferdinand F. Fernandez, Susan P. Graber,\n and John B. Owens, Circuit Judges.\n\n Opinion by Judge Fernandez;\n Concurrence by Judge Owens;\n Dissent by Judge Graber\n\n2 GONZALEZ ROMO V. BARR\n\n SUMMARY*\n\n\n Immigration\n\n Denying Eva Isabel Gonzalez Romo’s petition for review\nof the Board of Immigration Appeals’ precedential decision\nin Matter of Gonzalez Romo, 26 I. & N. Dec. 743 (BIA 2016),\nthe panel held that Gonzalez was inadmissible under 8 U.S.C.\n§§ 1101(a)(13)(C)(v) and 1182(a)(2)(A)(i)(I), because her\nconviction for solicitation to possess marijuana for sale, in\nviolation of Ariz. Rev. Stat. §§ 13-1002(A), 13-3405(A)(2),\nwas a crime involving moral turpitude.\n\n The panel gave Chevron deference to the BIA’s\ndetermination that a returning lawful permanent resident who\nhas a felony conviction for solicitation to possess marijuana\nfor sale is inadmissible under 8 U.S.C. § 1182(a)(2)(A)(i)(I),\neven though that provision refers only to attempt and\nconspiracy to commit a crime involving moral turpitude, and\nnot solicitation.\n\n The panel noted that there was no doubt that Gonzalez\nwas convicted in Arizona of the crime of solicitation to\npossess over four pounds of marijuana for sale, and that drug\ntrafficking crimes are generally crimes involving moral\nturpitude. The panel further noted that this court held\nin Barragan-Lopez v. Mukasey, 508 F.3d 899 (9th Cir.\n2007), that “solicitation to possess at least four pounds of\nmarijuana for sale constitutes a crime involving moral\nturpitude.” However, Barragan-Lopez involved 8 U.S.C.\n\n *\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\n GONZALEZ ROMO V. BARR 3\n\n§ 1227(a)(2)(A)(i)(I), which applies to deportability\ndeterminations, and does not include the “attempt or\nconspiracy to commit such a crime” phrase that appears in\n§ 1182(a)(2)(A)(i)(I), for inadmissibility purposes.\n\n The panel rejected Gonzalez’s contention that,\nby referencing only “attempt or conspiracy,”\n§ 1182(a)(2)(A)(i)(I) excluded crimes of solicitation. The\npanel stated that it saw no reason to deviate from the holding\nin Barragan-Lopez that solicitation of a crime of the\nmagnitude of possession of at least four pounds of marijuana\nfor sale is as turpitudinous as commission of the crime itself.\nThe panel discussed the cases relied upon by Gonzalez,\nLeyva-Licea v. I.N.S., 187 F.3d 1147 (9th Cir. 1999)\n(addressing drug trafficking aggravated felonies, 8 U.S.C.\n§ 1101(a)(43)(B)) and Coronado-Durazo v. I.N.S., 123 F.3d\n1322 (9th Cir. 1997) (addressing controlled substance\noffenses under 8 U.S.C. § 1227(a)(2)(B)(i)), in which this\ncourt held that other statutes referencing only attempts and\nconspiracies did not cover solicitation offenses. While\nrecognizing that reasoning which excludes solicitation as to\ncertain provisions could be said to apply whenever conspiracy\nor attempt are specifically mentioned in a statute, the panel\nstated that would be unlikely when it comes to crimes\ninvolving moral turpitude, which are a separate concept with\na long history that has been used by Congress in the\nimmigration laws for over a century. The panel explained\nthat it was doubtful that Congress intended to give the phrase\ndifferent scope in different provisions, and this court’s cases\ndo not suggest that it did. The panel further explained that\nthe legislative history indicated that Congress’s inclusion of\nthe terms attempt and conspiracy were meant to be clarifying,\nand that for inchoate crimes that are bad enough to be\nconsidered crimes of moral turpitude, there was little reason\n\n4 GONZALEZ ROMO V. BARR\n\nto think that Congress eliminated some, such as solicitation,\nfrom consideration, simply because it decided to make it clear\nthat others did fall into that category.\n\n Because the panel concluded that Congress’s intent was\nclear, it stopped at step one of the Chevron deference test.\nHowever, to the extent there was some perceived uncertainty\nregarding the meaning and reach of the statute, the panel\nconcluded that it was clear that the BIA’s resolution of the\nuncertainty was permissible and reasonable.\n\n Concurring, Judge Owens wrote that he did not believe\nreliance on Chevron was necessary, but he concurred with the\nmajority opinion because it comports with Supreme Court\nand Ninth Circuit law. Judge Owens continues to believe that\nthe current moral turpitude jurisprudence makes no sense, and\npointed out that he is not a lone wolf in so thinking.\n\n Dissenting, Judge Graber wrote that the majority opinion\nmisapplies both Ninth Circuit and Supreme Court precedent,\nand violates several canons of statutory construction. Judge\nGraber wrote that following this court’s precedents, the text\nof § 1182(a)(2)(A)(i)(I) unambiguously covers only the\ngeneric crimes of “attempt or conspiracy,” and that\nBarragan-Lopez, by its own terms, limited its holding to\n§ 1227(a)(2)(A)(i)(I), which does not contain similar\nwording.\n\n GONZALEZ ROMO V. BARR 5\n\n COUNSEL\n\nSamuel L. Brenner (argued), Matthew L. McGinnis, and\nCourtney M. Cox, Ropes & Gray LLP, Boston,\nMassachusetts; Marta F. Belcher, Ropes & Gray LLP, East\nPalo Alto, California; for Petitioner.\n\nRebecca Hoffberg Phillips (argued), Trial Attorney; John S.\nHogan, Assistant Director; Office of Immigration Litigation,\nCivil Division, United States Department of Justice,\nWashington, D.C.; for Respondent.\n\n\n OPINION\n\nFERNANDEZ, Circuit Judge:\n\n Eva Isabel Gonzalez Romo, a native and citizen of\nMexico and a lawful permanent resident of the United States,\npetitions for review of the Board of Immigration Appeals’\n(“BIA”) determination that she was inadmissible because she\nhad been convicted of a crime involving moral turpitude. See\n8 U.S.C. § 1182(a)(2)(A)(i)(I); 1 see also id.\n§ 1101(a)(13)(C)(v). The crime in question was solicitation\nto possess marijuana for sale, which she committed in the\nState of Arizona. See Ariz. Rev. Stat. §§ 13-1002(A), 13-\n3405(A)(2). We deny the petition.\n\n Gonzalez had been a lawful permanent resident of the\nUnited States since November 18, 1999. On October 11,\n2009, while driving with a suspended license, Gonzalez was\n\n 1\n Hereafter all statutory references are to Title 8 of the United States\nCode, unless otherwise indicated.\n\n6 GONZALEZ ROMO V. BARR\n\npulled over in Arizona for failing to make a complete stop at\na red light. The police officer noticed two large cardboard\nboxes in Gonzalez’s vehicle, and a strong odor of air\nfreshener emanated from that vehicle. Upon securing\nGonzalez’s consent to search the vehicle, officers discovered\n150 pounds of marijuana, which Gonzalez admitted that she\nhad knowingly agreed to transport for $1,000. As a result, on\nMarch 12, 2010, she was convicted of felony solicitation to\npossess marijuana for sale under Arizona Revised Statutes\nsections 13-1002, 13-3405(A)(2) and sentenced to one-and-a-\nhalf years in prison.\n\n After Gonzalez was released from prison, she traveled to\nMexico. Upon returning from her trip on May 3, 2014, she\nwas detained and paroled into the United States for removal\nproceedings. As relevant here, the government charged\nGonzalez with inadmissibility due to her conviction of a\ncrime involving moral turpitude.2 See §§ 1101(a)(13)(C)(v),\n1182(a)(2)(A)(i)(I). Gonzalez disputed that allegation. On\nSeptember 29, 2015, the immigration judge found Gonzalez\ninadmissible as charged. In particular, the immigration judge\nheld that solicitation to possess marijuana for sale is a crime\ninvolving moral turpitude. She appealed that determination\nto the BIA.\n\n In a precedential decision, the BIA dismissed Gonzalez’s\nappeal. The BIA concluded that even though Gonzalez was\na lawful permanent resident, because she had been convicted\n\n\n 2\n “[A] crime involving moral turpitude is generally a crime that (1) is\nvile, base, or depraved and (2) violates accepted moral standards.” Latter-\nSingh v. Holder, 668 F.3d 1156, 1161 (9th Cir. 2012) (internal quotation\nmarks omitted); see also Quintero-Salazar v. Keisler, 506 F.3d 688, 692\n(9th Cir. 2007).\n\n GONZALEZ ROMO V. BARR 7\n\nof an offense identified in § 1182(a)(2)(A)(i)(I), she was\nseeking admission to the United States due to\n§ 1101(a)(13)(C)(v). Matter of Gonzalez Romo, 26 I. & N.\nDec. 743, 748 (BIA 2016). That is, the BIA concluded that\nGonzalez’s conviction was a crime involving moral turpitude.\nId. It rejected Gonzalez’s contention that, by referencing\n“attempt or conspiracy,” § 1182(a)(2)(A)(i)(I) excludes\ncrimes of solicitation even if they otherwise constitute crimes\ninvolving moral turpitude. Id. at 747–48. In doing so, the\nBIA receded from contrary language in a footnote in an\nearlier case,3 which it now characterizes as dicta. Matter of\nGonzalez Romo, 26 I. & N. Dec. at 748. Even in that earlier\nfootnote, however, the BIA had not specifically said that it\nagreed with the Ninth Circuit’s then-supposed position. This\npetition for review followed.\n\n JURISDICTION AND STANDARD OF REVIEW\n\n We have jurisdiction pursuant to 8 U.S.C. § 1252(a)(1).\n“We review the BIA’s determination of purely legal questions\nregarding the Immigration and Nationality Act de novo,” with\nappropriate deference. Kankamalage v. I.N.S., 335 F.3d 858,\n861 (9th Cir. 2003).\n\n\n\n\n 3\n The footnote reads “In effect, the Ninth Circuit has indicated that\n[§ 1227(a)(2)(A)] is broader in its coverage of crimes involving moral\nturpitude than [§ 1182(a)(2)(A)(i)(I)], because it would include inchoate\noffenses, such as solicitation and facilitation, that are not specifically\nenumerated in [§ 1182(a)(2)(A)(i)(I)], which lists only attempts and\nconspiracies.” Matter of Vo, 25 I. & N. Dec. 426, 429 n.4 (BIA 2011).\n\n8 GONZALEZ ROMO V. BARR\n\n Where, as here, the BIA’s decision is published, it may\nwell be entitled to Chevron4 deference. See Chen v. Mukasey,\n524 F.3d 1028, 1031 (9th Cir. 2008); Kankamalage, 335 F.3d\nat 862. Under Chevron’s framework, the first step is to\nascertain “whether Congress has directly spoken to the\nprecise question at issue.” Chevron, 467 U.S. at 842, 104 S.\nCt. at 2781. To so determine, the court employs “traditional\ntools of statutory construction.” Id. at 843 n.9, 104 S. Ct. at\n2782 n.9. If the statute’s meaning is clear, “that is the end of\nthe matter; for the court, as well as the agency, must give\neffect to the unambiguously expressed intent of Congress.”\nId. at 842–43, 104 S. Ct. at 2781. Otherwise, we proceed to\nthe second step, that is: “[I]f the statute is silent or ambiguous\nwith respect to the specific issue, the question for the court is\nwhether the agency’s answer is based on a permissible\nconstruction of the statute.” Id. at 843, 104 S. Ct. at 2782. If\nit is, “a court may not substitute its own construction of a\nstatutory provision for a reasonable interpretation made by\nthe . . . agency.” Id. at 844, 104 S. Ct. at 2782.\n\n DISCUSSION\n\n The BIA determined that even though Gonzalez was a\nlegal permanent resident, she was removable because she was\ninadmissible to the United States when she presented herself\nat the border. See §§ 1101(a)(13)(C)(v), 1182(a)(2)(A)(i)(I);\nsee also Matter of Gonzalez Romo, 26 I. & N. Dec. at 748.\nGonzalez argues that the BIA erred because she was not\n\n\n\n\n 4\n Chevron, U.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837,\n842–44, 104 S. Ct. 2778, 2781–82, 81 L. Ed. 2d 694 (1984).\n\n GONZALEZ ROMO V. BARR 9\n\nsubject to those provisions.5 We disagree. As relevant here,\npursuant to § 1101(a)(13)(C)(v), an alien who is a legal\npermanent resident is treated as one seeking admission if the\nalien “has committed an offense identified in section\n1182(a)(2).” And pursuant to § 1182(a)(2)(A)(i)(I), as\nrelevant here, an “alien convicted of . . . a crime involving\nmoral turpitude . . . or an attempt or conspiracy to commit\nsuch a crime . . . is inadmissible.”\n\n There can be no doubt that Gonzalez was convicted in\nArizona of the crime of solicitation to possess over four\npounds of marijuana for sale. See Ariz. Rev. Stat. §§ 13-\n1002(A), (B)(2), 13-3405(A)(2), B(6). Nor can there be any\ndoubt that drug trafficking crimes are generally crimes\ninvolving moral turpitude. See Barragan-Lopez v. Mukasey,\n508 F.3d 899, 903–04 (9th Cir. 2007) (so stating and\nunderscoring just how turpitudinous drug trafficking is); see\nalso Turijan v. Holder, 744 F.3d 617, 621 (9th Cir. 2014);\nRohit v. Holder, 670 F.3d 1085, 1088–89 (9th Cir. 2012).\nMoreover, we have specifically held that “solicitation to\npossess at least four pounds of marijuana for sale . . .\nconstitutes a crime involving moral turpitude.” Barragan-\nLopez, 508 F.3d at 904; see also Rohit, 670 F.3d at 1089; Atl.\nRichfield Co. v. Guerami, 820 F.2d 280, 282 (9th Cir. 1987).\nIn fact, Barragan-Lopez, 508 F.3d at 901, dealt with a\nconviction under Arizona statutory provisions which are the\nsame as those involved here. All of that would lead\ninexorably to our denial of Gonzalez’s petition were it not for\n\n\n 5\n We note that because her offense was committed in 2009, she is not\nentitled to the benefit of the rules that applied to legal permanent residents\nbefore the statutes were amended in 1996. Cf. Vartelas v. Holder, 566\nU.S. 257, 261, 132 S. Ct. 1479, 1483–84, 182 L. Ed. 2d 473 (2012);\nCenturion v. Sessions, 860 F.3d 69, 77 (2d Cir. 2017).\n\n10 GONZALEZ ROMO V. BARR\n\nthe fact that Barragan-Lopez was construing\n§ 1227(a)(2)(A)(i)(I) which states, in pertinent part, “[a]ny\nalien who . . . is convicted of a crime involving moral\nturpitude . . . is deportable.” As Gonzalez points out, the\nlanguage of § 1227(a)(2)(A)(i)(I), for deportation purposes,\ndoes not include the “attempt or conspiracy to commit such\na crime” phrase that appears in § 1182(a)(2)(A)(i)(I), for\ninadmissibility purposes, and that, she argues, makes all the\ndifference in the world. The argument has some purchase,6\nbut like the BIA,7 we are not convinced.\n\n As Barragan-Lopez stated,8 in neither of the cases that\nGonzalez relies upon were we asked to consider the meaning\nand scope of the phrase “crime involving moral turpitude,”\nand we certainly did not reflect upon whether solicitation to\npossess a large amount of marijuana for sale would be\nturpitudinous enough to come within the reach of that\nphrase.9 Rather, in Leyva-Licea,10 we asked whether a\nconviction in Arizona of solicitation to possess at least two\npounds, but less than four pounds, of marijuana for sale was\nan aggravated felony. We determined that it was not because\nthe definition of aggravated felony, as set forth in\n§ 1101(a)(43)(B) (trafficking in controlled substances) in turn\nreferred to 18 U.S.C. § 924(c), which in turn ultimately relied\n\n\n 6\n See Leyva-Licea v. I.N.S., 187 F.3d 1147, 1150 (9th Cir. 1999);\nCoronado-Durazo v. I.N.S., 123 F.3d 1322, 1325 (9th Cir. 1997).\n 7\n Matter of Gonzalez Romo, 26 I. & N. Dec. at 746–47.\n 8\n 508 F.3d at 904–05.\n 9\n Id.\n 10\n 187 F.3d at 1150.\n\n GONZALEZ ROMO V. BARR 11\n\nupon 21 U.S.C. § 846, which covered only attempts or\nconspiracies.11 Nothing therein treated solicitation as the kind\nof offense that would be an aggravated felony.12 Moreover,\nin Coronado-Durazo,13 we asked whether a conviction in\nArizona of solicitation to possess cocaine was a controlled\nsubstance offense within the meaning of § 1227(a)(2)(B)(i).14\nWe held it was not because that section covered only “‘a\nviolation [by an alien] of (or a conspiracy or attempt to\nviolate) any law or regulation . . . relating to a controlled\nsubstance.’”15 It did not by its terms indicate that the scope\nof the phrase “violation . . . relating to a controlled substance\noffense” included solicitation, or that solicitation was\nencompassed by an attempt or conspiracy. As we have since\nnoted, had solicitation been included within the meaning of\ncontrolled substance offense, the result would have been the\nopposite. See Mielewczyk v. Holder, 575 F.3d 992, 997–98\n(9th Cir. 2009). Again, in Barragan-Lopez,16 we asked\nwhether the scope of “crime involving moral turpitude” did\ninclude solicitation of that crime for amounts of marijuana\nover four pounds and said that it did because solicitation of a\ncrime of that magnitude was as turpitudinous as commission\n\n\n\n 11\n See id.\n 12\n Id.\n 13\n 123 F.3d at 1324.\n 14\n This is the current section containing the provisions which at that\ntime were contained in § 1251(a)(2)(B)(i).\n 15\n Coronado-Durazo, 123 F.3d at 1324; see also id. at 1325–26.\n 16\n 508 F.3d at 903.\n\n12 GONZALEZ ROMO V. BARR\n\nof the crime itself.17 We did not opine on whether the result\nwould be the same for “a very small quantity”18 of the\nsubstance. We agree with that analysis and see no\ncompelling reason to deviate therefrom. On the contrary, we\nsee strong reasons not to do so.\n\n First, we recognize that reasoning which excludes\nsolicitation as to certain provisions could be said to apply\nwhenever conspiracy or attempt are specifically mentioned in\na statute. However, that would be unlikely when it comes to\ncrimes involving moral turpitude. That is a separate concept\nwith a long history19 which has been used by Congress in the\nimmigration laws for over a century.20 It is doubtful that\nCongress intended to give the phrase different scope in\ndifferent provisions,21 and our cases do not suggest that it\ndid.22\n\n\n 17\n See id. at 904.\n 18\n Id.\n 19\n The concept is deeply rooted in our law. See Jordan v. De George,\n341 U.S. 223, 227, 71 S. Ct. 703, 705–06, 95 L. Ed. 886 (1951).\n 20\n See id. at 229–30, 71 S. Ct. at 707.\n 21\n See Merrill Lynch, Pierce, Fenner & Smith Inc. v. Dabit, 547 U.S.\n71, 85–86, 126 S. Ct. 1503, 1513, 164 L. Ed. 2d 179 (2006); cf. Powerex\nCorp. v. Reliant Energy Servs., Inc., 551 U.S. 224, 232, 127 S. Ct. 2411,\n2417, 168 L. Ed. 2d 112 (2007).\n 22\n See, e.g., Islas-Veloz v. Whitaker, 914 F.3d 1249, 1250–51 (9th Cir.\n2019); Gonzalez-Cervantes v. Holder, 709 F.3d 1265, 1269 (9th Cir.\n2013); Morales v. Gonzales, 478 F.3d 972, 978 (9th Cir. 2007), abrogated\non other grounds by Anaya-Ortiz v. Holder, 594 F.3d 673, 677–78 (9th\nCir. 2010).\n\n GONZALEZ ROMO V. BARR 13\n\n Second, we recognize that solicitation is a type of\ninchoate crime because it would not be a crime involving\nmoral turpitude if the underlying offense (here possession for\nsale of over four pounds of marijuana) was not a crime\ninvolving moral turpitude. See Barragan-Lopez, 508 F.3d at\n903. No doubt attempt and conspiracy are also inchoate\ncrimes in that sense. Why then would Congress mention\nthem specifically and not mention other inchoate crimes (like\nsolicitation) when referring to crimes involving moral\nturpitude? Of course, we cannot say for sure, and the\nlegislative history, such as it is, does not make the answer\nperspicuous. It just indicates that the addition of those words\nwas clarifying. See 139 Cong. Rec. 15,275 (1993) (indicating\nthat the language was intended to “make clear” that those\nconcepts were covered); see also Immigration and Nationality\nTechnical Corrections Act of 1994, Pub. L. 103–416, § 203,\n108 Stat. 4305, 4311 (indicating that the language constitutes\na clarification). However, when it comes to inchoate crimes\nthat are bad enough to be considered crimes of moral\nturpitude, we see little reason to think that Congress took this\nsomewhat indirect approach to eliminating some of them\nfrom consideration simply because it decided to make it clear\nthat others did fall into that category. See Ali v. Fed. Bureau\nof Prisons, 552 U.S. 214, 226–27, 128 S. Ct. 831, 840, 169 L.\nEd. 2d 680 (2008). It is more likely that Congress was\nfocused on the two that it mentioned and had no intention of\nexcluding all others. See Marx v. Gen. Revenue Corp.,\n568 U.S. 371, 385, 133 S. Ct. 1166, 1177, 185 L. Ed. 2d 242\n(2013). And, while redundancy and surplusage are not\nalways desirable, they are not anoetic. See Conn. Nat’l Bank\nv. Germain, 503 U.S. 249, 253, 112 S. Ct. 1146, 1149, 117 L.\nEd. 2d 391 (1992). In fact, it is not unusual to see\nredundancy occur in clarifying enactments, which do run the\nrisk of introducing other problems while focusing on the task\n\n14 GONZALEZ ROMO V. BARR\n\nof making sure that a particular problem does not arise. See\nid. As the Court has said: “In any event, our hesitancy to\nconstrue statutes to render language superfluous does not\nrequire us to avoid surplusage at all costs. It is appropriate to\ntolerate a degree of surplusage rather than adopt a textually\ndubious construction that threatens to render the entire\nprovision a nullity.” United States v. Atl. Research Corp.,\n551 U.S. 128, 137, 127 S. Ct. 2331, 2337, 168 L. Ed. 2d 28\n(2007). In the case at hand, the cost would be particularly\nhigh because it would create an anomaly in application of the\nstatutory scheme, which we doubt Congress intended when it\nadopted its clarifying amendments.23 The anomaly can be\nseen if we start with the rather uncontroversial reflection that\n“[o]ur immigration law has generally treated aliens who are\nalready on our soil (and who are therefore deportable) more\nfavorably than aliens who are merely seeking admittance (and\nwho are therefore excludable).” Servin-Espinoza v. Ashcroft,\n309 F.3d 1193, 1198 (9th Cir. 2002). But removing\nsolicitation of possession of over four pounds of marijuana\nfrom the definition of a crime involving moral turpitude in\n§ 1182(a)(2)(A)(i)(I), although solicitation is part of the\ndefinition in § 1227(a)(2)(A)(i)(I), would turn that concept on\nits head. That is because we already know that even a legal\npermanent resident can be removed pursuant to\n§ 1227(a)(2)(A)(i)(I) for soliciting a crime involving moral\nturpitude. See Barragan-Lopez, 508 F.3d at 905. However,\nif solicitation is removed from the definition of crime\ninvolving moral turpitude in § 1182(a)(2)(A)(i)(I), no alien\n(legal permanent resident or not) can be excluded for\nsoliciting a crime involving moral turpitude, regardless of\n\n\n 23\n In general, we do attempt to avoid anomalies and try to view the\nscheme at hand as “symmetrical and coherent.” Perez-Guzman v. Lynch,\n835 F.3d 1066, 1074 (9th Cir. 2016) (internal quotation marks omitted).\n\n GONZALEZ ROMO V. BARR 15\n\nhow turpitudinous that crime may be. Again, we doubt that\nCongress intended any such thing. In fact, in the context of\nthis statutory scheme regarding who can enter and who can be\nremoved, ascribing that intent to Congress would lead to a\nresult so dubious that it would be nothing short of an internal\ninconsistency amounting to an absurdity. See United States\nv. Turkette, 452 U.S. 576, 580, 101 S. Ct. 2524, 2527, 69 L.\nEd. 2d 246 (1981) (“[A]bsurd results are to be avoided and\ninternal inconsistencies in the statute must be dealt with.”);\nsee also McNeill v. United States, 563 U.S. 816, 822, 131 S.\nCt. 2218, 2223, 180 L. Ed. 2d 35 (2011).\n\n The discussion above indicates that we can stop at step\none of the Chevron process24 because Congress’ true intent is\nclear to us. However, to the extent that there is some\nperceived uncertainty25 regarding the meaning and reach of\nthe statute, that ambiguity26 would direct us to step two.27 At\nthat step, it is clear that the BIA’s resolution of the\n\n\n\n\n 24\n 467 U.S. at 842–43, 104 S. Ct. at 2781.\n 25\n See, e.g., Mielewczyk, 575 F.3d at 996–98; Barragan-Lopez,\n508 F.3d at 902–05; Coronado-Durazo, 123 F.3d at 1324–26; see also\nMizrahi v. Gonzales, 492 F.3d 156, 164–67 (2d Cir. 2007); Peters v.\nAshcroft, 383 F.3d 302, 305–09 (5th Cir. 2004).\n 26\n We note that some ambiguity in statutory language does not\npreclude resolution of that ambiguity by the BIA. See Chevron, 467 U.S.\nat 843, 104 S. Ct. at 2782; Perez-Guzman, 835 F.3d at 1076 n.5.\n 27\n Chevron, 467 U.S. at 843–44, 104 S. Ct. at 2781–82.\n\n16 GONZALEZ ROMO V. BARR\n\nuncertainty is permissible and reasonable. Indeed, it appears\nto us that it is the best resolution.28\n\n Thus, we deny the petition.\n\n CONCLUSION\n\n We hold that a conviction in Arizona for solicitation to\npossess at least four pounds of marijuana for sale constitutes\na crime involving moral turpitude for purposes of\n§ 1182(a)(2)(A)(i)(I), and, therefore, Gonzalez was\ninadmissible. See § 1101(a)(13)(C)(v).\n\n Petition DENIED.\n\n\n\n\n 28\n We have not overlooked Gonzalez’s argument that the BIA should\nbe accorded less deference because it changed its position about what it\nthought our case law required. See Matter of Vo, 25 I. & N. Dec. at 429\nn.4. However, as we see it, an agency’s reflections on what it thought we\nhave said, or might say, is quite different from a determination of what the\nlaw is. In any event: “In the circumstances of this case, where the\nagency’s interpretation of a statute is at least as plausible as competing\nones, there is little, if any, reason not to defer to its construction.” Good\nSamaritan Hosp. v. Shalala, 508 U.S. 402, 417, 113 S. Ct. 2151, 2161,\n124 L. Ed. 2d 368 (1993).\n\n GONZALEZ ROMO V. BARR 17\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Eva Gonzalez Romo v. William Barr\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-14\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: Yes\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nEVA ISABEL GONZALEZ ROMO, Eva No. 16-71559\\nAKA Eva Isabel Romo,\\n Petitioner, Agency No.\\n A047-334-955\\n v.\\n\\nWILLIAM P. BARR, Attorney General, OPINION\\n Respondent.\\n\\n On Petition for Review of an Order of the\\n Board of Immigration Appeals\\n\\n Argued and Submitted July 8, 2019\\n Portland, Oregon\\n\\n Filed August 14, 2019\\n\\n Before: Ferdinand F. Fernandez, Susan P. Graber,\\n and John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Fernandez;\\n Concurrence by Judge Owens;\\n Dissent by Judge Graber\\n\\n2 GONZALEZ ROMO V. BARR\\n\\n SUMMARY*\\n\\n\\n Immigration\\n\\n Denying Eva Isabel Gonzalez Romo’s petition for review\\nof the Board of Immigration Appeals’ precedential decision\\nin Matter of Gonzalez Romo, 26 I. & N. Dec. 743 (BIA 2016),\\nthe panel held that Gonzalez was inadmissible under 8 U.S.C.\\n§§ 1101(a)(13)(C)(v) and 1182(a)(2)(A)(i)(I), because her\\nconviction for solicitation to possess marijuana for sale, in\\nviolation of Ariz. Rev. Stat. §§ 13-1002(A), 13-3405(A)(2),\\nwas a crime involving moral turpitude.\\n\\n The panel gave Chevron deference to the BIA’s\\ndetermination that a returning lawful permanent resident who\\nhas a felony conviction for solicitation to possess marijuana\\nfor sale is inadmissible under 8 U.S.C. § 1182(a)(2)(A)(i)(I),\\neven though that provision refers only to attempt and\\nconspiracy to commit a crime involving moral turpitude, and\\nnot solicitation.\\n\\n The panel noted that there was no doubt that Gonzalez\\nwas convicted in Arizona of the crime of solicitation to\\npossess over four pounds of marijuana for sale, and that drug\\ntrafficking crimes are generally crimes involving moral\\nturpitude. The panel further noted that this court held\\nin Barragan-Lopez v. Mukasey, 508 F.3d 899 (9th Cir.\\n2007), that “solicitation to possess at least four pounds of\\nmarijuana for sale constitutes a crime involving moral\\nturpitude.” However, Barragan-Lopez involved 8 U.S.C.\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\n GONZALEZ ROMO V. BARR 3\\n\\n§ 1227(a)(2)(A)(i)(I), which applies to deportability\\ndeterminations, and does not include the “attempt or\\nconspiracy to commit such a crime” phrase that appears in\\n§ 1182(a)(2)(A)(i)(I), for inadmissibility purposes.\\n\\n The panel rejected Gonzalez’s contention that,\\nby referencing only “attempt or conspiracy,”\\n§ 1182(a)(2)(A)(i)(I) excluded crimes of solicitation. The\\npanel stated that it saw no reason to deviate from the holding\\nin Barragan-Lopez that solicitation of a crime of the\\nmagnitude of possession of at least four pounds of marijuana\\nfor sale is as turpitudinous as commission of the crime itself.\\nThe panel discussed the cases relied upon by Gonzalez,\\nLeyva-Licea v. I.N.S., 187 F.3d 1147 (9th Cir. 1999)\\n(addressing drug trafficking aggravated felonies, 8 U.S.C.\\n§ 1101(a)(43)(B)) and Coronado-Durazo v. I.N.S., 123 F.3d\\n1322 (9th Cir. 1997) (addressing controlled substance\\noffenses under 8 U.S.C. § 1227(a)(2)(B)(i)), in which this\\ncourt held that other statutes referencing only attempts and\\nconspiracies did not cover solicitation offenses. While\\nrecognizing that reasoning which excludes solicitation as to\\ncertain provisions could be said to apply whenever conspiracy\\nor attempt are specifically mentioned in a statute, the panel\\nstated that would be unlikely when it comes to crimes\\ninvolving moral turpitude, which are a separate concept with\\na long history that has been used by Congress in the\\nimmigration laws for over a century. The panel explained\\nthat it was doubtful that Congress intended to give the phrase\\ndifferent scope in different provisions, and this court’s cases\\ndo not suggest that it did. The panel further explained that\\nthe legislative history indicated that Congress’s inclusion of\\nthe terms attempt and conspiracy were meant to be clarifying,\\nand that for inchoate crimes that are bad enough to be\\nconsidered crimes of moral turpitude, there was little reason\\n\\n4 GONZALEZ ROMO V. BARR\\n\\nto think that Congress eliminated some, such as solicitation,\\nfrom consideration, simply because it decided to make it clear\\nthat others did fall into that category.\\n\\n Because the panel concluded that Congress’s intent was\\nclear, it stopped at step one of the Chevron deference test.\\nHowever, to the extent there was some perceived uncertainty\\nregarding the meaning and reach of the statute, the panel\\nconcluded that it was clear that the BIA’s resolution of the\\nuncertainty was permissible and reasonable.\\n\\n Concurring, Judge Owens wrote that he did not believe\\nreliance on Chevron was necessary, but he concurred with the\\nmajority opinion because it comports with Supreme Court\\nand Ninth Circuit law. Judge Owens continues to believe that\\nthe current moral turpitude jurisprudence makes no sense, and\\npointed out that he is not a lone wolf in so thinking.\\n\\n Dissenting, Judge Graber wrote that the majority opinion\\nmisapplies both Ninth Circuit and Supreme Court precedent,\\nand violates several canons of statutory construction. Judge\\nGraber wrote that following this court’s precedents, the text\\nof § 1182(a)(2)(A)(i)(I) unambiguously covers only the\\ngeneric crimes of “attempt or conspiracy,” and that\\nBarragan-Lopez, by its own terms, limited its holding to\\n§ 1227(a)(2)(A)(i)(I), which does not contain similar\\nwording.\\n\\n GONZALEZ ROMO V. BARR 5\\n\\n COUNSEL\\n\\nSamuel L. Brenner (argued), Matthew L. McGinnis, and\\nCourtney M. Cox, Ropes & Gray LLP, Boston,\\nMassachusetts; Marta F. Belcher, Ropes & Gray LLP, East\\nPalo Alto, California; for Petitioner.\\n\\nRebecca Hoffberg Phillips (argued), Trial Attorney; John S.\\nHogan, Assistant Director; Office of Immigration Litigation,\\nCivil Division, United States Department of Justice,\\nWashington, D.C.; for Respondent.\\n\\n\\n OPINION\\n\\nFERNANDEZ, Circuit Judge:\\n\\n Eva Isabel Gonzalez Romo, a native and citizen of\\nMexico and a lawful permanent resident of the United States,\\npetitions for review of the Board of Immigration Appeals’\\n(“BIA”) determination that she was inadmissible because she\\nhad been convicted of a crime involving moral turpitude. See\\n8 U.S.C. § 1182(a)(2)(A)(i)(I); 1 see also id.\\n§ 1101(a)(13)(C)(v). The crime in question was solicitation\\nto possess marijuana for sale, which she committed in the\\nState of Arizona. See Ariz. Rev. Stat. §§ 13-1002(A), 13-\\n3405(A)(2). We deny the petition.\\n\\n Gonzalez had been a lawful permanent resident of the\\nUnited States since November 18, 1999. On October 11,\\n2009, while driving with a suspended license, Gonzalez was\\n\\n 1\\n Hereafter all statutory references are to Title 8 of the United States\\nCode, unless otherwise indicated.\\n\\n6 GONZALEZ ROMO V. BARR\\n\\npulled over in Arizona for failing to make a complete stop at\\na red light. The police officer noticed two large cardboard\\nboxes in Gonzalez’s vehicle, and a strong odor of air\\nfreshener emanated from that vehicle. Upon securing\\nGonzalez’s consent to search the vehicle, officers discovered\\n150 pounds of marijuana, which Gonzalez admitted that she\\nhad knowingly agreed to transport for $1,000. As a result, on\\nMarch 12, 2010, she was convicted of felony solicitation to\\npossess marijuana for sale under Arizona Revised Statutes\\nsections 13-1002, 13-3405(A)(2) and sentenced to one-and-a-\\nhalf years in prison.\\n\\n After Gonzalez was released from prison, she traveled to\\nMexico. Upon returning from her trip on May 3, 2014, she\\nwas detained and paroled into the United States for removal\\nproceedings. As relevant here, the government charged\\nGonzalez with inadmissibility due to her conviction of a\\ncrime involving moral turpitude.2 See §§ 1101(a)(13)(C)(v),\\n1182(a)(2)(A)(i)(I). Gonzalez disputed that allegation. On\\nSeptember 29, 2015, the immigration judge found Gonzalez\\ninadmissible as charged. In particular, the immigration judge\\nheld that solicitation to possess marijuana for sale is a crime\\ninvolving moral turpitude. She appealed that determination\\nto the BIA.\\n\\n In a precedential decision, the BIA dismissed Gonzalez’s\\nappeal. The BIA concluded that even though Gonzalez was\\na lawful permanent resident, because she had been convicted\\n\\n\\n 2\\n “[A] crime involving moral turpitude is generally a crime that (1) is\\nvile, base, or depraved and (2) violates accepted moral standards.” Latter-\\nSingh v. Holder, 668 F.3d 1156, 1161 (9th Cir. 2012) (internal quotation\\nmarks omitted); see also Quintero-Salazar v. Keisler, 506 F.3d 688, 692\\n(9th Cir. 2007).\\n\\n GONZALEZ ROMO V. BARR 7\\n\\nof an offense identified in § 1182(a)(2)(A)(i)(I), she was\\nseeking admission to the United States due to\\n§ 1101(a)(13)(C)(v). Matter of Gonzalez Romo, 26 I. & N.\\nDec. 743, 748 (BIA 2016). That is, the BIA concluded that\\nGonzalez’s conviction was a crime involving moral turpitude.\\nId. It rejected Gonzalez’s contention that, by referencing\\n“attempt or conspiracy,” § 1182(a)(2)(A)(i)(I) excludes\\ncrimes of solicitation even if they otherwise constitute crimes\\ninvolving moral turpitude. Id. at 747–48. In doing so, the\\nBIA receded from contrary language in a footnote in an\\nearlier case,3 which it now characterizes as dicta. Matter of\\nGonzalez Romo, 26 I. & N. Dec. at 748. Even in that earlier\\nfootnote, however, the BIA had not specifically said that it\\nagreed with the Ninth Circuit’s then-supposed position. This\\npetition for review followed.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction pursuant to 8 U.S.C. § 1252(a)(1).\\n“We review the BIA’s determination of purely legal questions\\nregarding the Immigration and Nationality Act de novo,” with\\nappropriate deference. Kankamalage v. I.N.S., 335 F.3d 858,\\n861 (9th Cir. 2003).\\n\\n\\n\\n\\n 3\\n The footnote reads “In effect, the Ninth Circuit has indicated that\\n[§ 1227(a)(2)(A)] is broader in its coverage of crimes involving moral\\nturpitude than [§ 1182(a)(2)(A)(i)(I)], because it would include inchoate\\noffenses, such as solicitation and facilitation, that are not specifically\\nenumerated in [§ 1182(a)(2)(A)(i)(I)], which lists only attempts and\\nconspiracies.” Matter of Vo, 25 I. & N. Dec. 426, 429 n.4 (BIA 2011).\\n\\n8 GONZALEZ ROMO V. BARR\\n\\n Where, as here, the BIA’s decision is published, it may\\nwell be entitled to Chevron4 deference. See Chen v. Mukasey,\\n524 F.3d 1028, 1031 (9th Cir. 2008); Kankamalage, 335 F.3d\\nat 862. Under Chevron’s framework, the first step is to\\nascertain “whether Congress has directly spoken to the\\nprecise question at issue.” Chevron, 467 U.S. at 842, 104 S.\\nCt. at 2781. To so determine, the court employs “traditional\\ntools of statutory construction.” Id. at 843 n.9, 104 S. Ct. at\\n2782 n.9. If the statute’s meaning is clear, “that is the end of\\nthe matter; for the court, as well as the agency, must give\\neffect to the unambiguously expressed intent of Congress.”\\nId. at 842–43, 104 S. Ct. at 2781. Otherwise, we proceed to\\nthe second step, that is: “[I]f the statute is silent or ambiguous\\nwith respect to the specific issue, the question for the court is\\nwhether the agency’s answer is based on a permissible\\nconstruction of the statute.” Id. at 843, 104 S. Ct. at 2782. If\\nit is, “a court may not substitute its own construction of a\\nstatutory provision for a reasonable interpretation made by\\nthe . . . agency.” Id. at 844, 104 S. Ct. at 2782.\\n\\n DISCUSSION\\n\\n The BIA determined that even though Gonzalez was a\\nlegal permanent resident, she was removable because she was\\ninadmissible to the United States when she presented herself\\nat the border. See §§ 1101(a)(13)(C)(v), 1182(a)(2)(A)(i)(I);\\nsee also Matter of Gonzalez Romo, 26 I. & N. Dec. at 748.\\nGonzalez argues that the BIA erred because she was not\\n\\n\\n\\n\\n 4\\n Chevron, U.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837,\\n842–44, 104 S. Ct. 2778, 2781–82, 81 L. Ed. 2d 694 (1984).\\n\\n GONZALEZ ROMO V. BARR 9\\n\\nsubject to those provisions.5 We disagree. As relevant here,\\npursuant to § 1101(a)(13)(C)(v), an alien who is a legal\\npermanent resident is treated as one seeking admission if the\\nalien “has committed an offense identified in section\\n1182(a)(2).” And pursuant to § 1182(a)(2)(A)(i)(I), as\\nrelevant here, an “alien convicted of . . . a crime involving\\nmoral turpitude . . . or an attempt or conspiracy to commit\\nsuch a crime . . . is inadmissible.”\\n\\n There can be no doubt that Gonzalez was convicted in\\nArizona of the crime of solicitation to possess over four\\npounds of marijuana for sale. See Ariz. Rev. Stat. §§ 13-\\n1002(A), (B)(2), 13-3405(A)(2), B(6). Nor can there be any\\ndoubt that drug trafficking crimes are generally crimes\\ninvolving moral turpitude. See Barragan-Lopez v. Mukasey,\\n508 F.3d 899, 903–04 (9th Cir. 2007) (so stating and\\nunderscoring just how turpitudinous drug trafficking is); see\\nalso Turijan v. Holder, 744 F.3d 617, 621 (9th Cir. 2014);\\nRohit v. Holder, 670 F.3d 1085, 1088–89 (9th Cir. 2012).\\nMoreover, we have specifically held that “solicitation to\\npossess at least four pounds of marijuana for sale . . .\\nconstitutes a crime involving moral turpitude.” Barragan-\\nLopez, 508 F.3d at 904; see also Rohit, 670 F.3d at 1089; Atl.\\nRichfield Co. v. Guerami, 820 F.2d 280, 282 (9th Cir. 1987).\\nIn fact, Barragan-Lopez, 508 F.3d at 901, dealt with a\\nconviction under Arizona statutory provisions which are the\\nsame as those involved here. All of that would lead\\ninexorably to our denial of Gonzalez’s petition were it not for\\n\\n\\n 5\\n We note that because her offense was committed in 2009, she is not\\nentitled to the benefit of the rules that applied to legal permanent residents\\nbefore the statutes were amended in 1996. Cf. Vartelas v. Holder, 566\\nU.S. 257, 261, 132 S. Ct. 1479, 1483–84, 182 L. Ed. 2d 473 (2012);\\nCenturion v. Sessions, 860 F.3d 69, 77 (2d Cir. 2017).\\n\\n10 GONZALEZ ROMO V. BARR\\n\\nthe fact that Barragan-Lopez was construing\\n§ 1227(a)(2)(A)(i)(I) which states, in pertinent part, “[a]ny\\nalien who . . . is convicted of a crime involving moral\\nturpitude . . . is deportable.” As Gonzalez points out, the\\nlanguage of § 1227(a)(2)(A)(i)(I), for deportation purposes,\\ndoes not include the “attempt or conspiracy to commit such\\na crime” phrase that appears in § 1182(a)(2)(A)(i)(I), for\\ninadmissibility purposes, and that, she argues, makes all the\\ndifference in the world. The argument has some purchase,6\\nbut like the BIA,7 we are not convinced.\\n\\n As Barragan-Lopez stated,8 in neither of the cases that\\nGonzalez relies upon were we asked to consider the meaning\\nand scope of the phrase “crime involving moral turpitude,”\\nand we certainly did not reflect upon whether solicitation to\\npossess a large amount of marijuana for sale would be\\nturpitudinous enough to come within the reach of that\\nphrase.9 Rather, in Leyva-Licea,10 we asked whether a\\nconviction in Arizona of solicitation to possess at least two\\npounds, but less than four pounds, of marijuana for sale was\\nan aggravated felony. We determined that it was not because\\nthe definition of aggravated felony, as set forth in\\n§ 1101(a)(43)(B) (trafficking in controlled substances) in turn\\nreferred to 18 U.S.C. § 924(c), which in turn ultimately relied\\n\\n\\n 6\\n See Leyva-Licea v. I.N.S., 187 F.3d 1147, 1150 (9th Cir. 1999);\\nCoronado-Durazo v. I.N.S., 123 F.3d 1322, 1325 (9th Cir. 1997).\\n 7\\n Matter of Gonzalez Romo, 26 I. & N. Dec. at 746–47.\\n 8\\n 508 F.3d at 904–05.\\n 9\\n Id.\\n 10\\n 187 F.3d at 1150.\\n\\n GONZALEZ ROMO V. BARR 11\\n\\nupon 21 U.S.C. § 846, which covered only attempts or\\nconspiracies.11 Nothing therein treated solicitation as the kind\\nof offense that would be an aggravated felony.12 Moreover,\\nin Coronado-Durazo,13 we asked whether a conviction in\\nArizona of solicitation to possess cocaine was a controlled\\nsubstance offense within the meaning of § 1227(a)(2)(B)(i).14\\nWe held it was not because that section covered only “‘a\\nviolation [by an alien] of (or a conspiracy or attempt to\\nviolate) any law or regulation . . . relating to a controlled\\nsubstance.’”15 It did not by its terms indicate that the scope\\nof the phrase “violation . . . relating to a controlled substance\\noffense” included solicitation, or that solicitation was\\nencompassed by an attempt or conspiracy. As we have since\\nnoted, had solicitation been included within the meaning of\\ncontrolled substance offense, the result would have been the\\nopposite. See Mielewczyk v. Holder, 575 F.3d 992, 997–98\\n(9th Cir. 2009). Again, in Barragan-Lopez,16 we asked\\nwhether the scope of “crime involving moral turpitude” did\\ninclude solicitation of that crime for amounts of marijuana\\nover four pounds and said that it did because solicitation of a\\ncrime of that magnitude was as turpitudinous as commission\\n\\n\\n\\n 11\\n See id.\\n 12\\n Id.\\n 13\\n 123 F.3d at 1324.\\n 14\\n This is the current section containing the provisions which at that\\ntime were contained in § 1251(a)(2)(B)(i).\\n 15\\n Coronado-Durazo, 123 F.3d at 1324; see also id. at 1325–26.\\n 16\\n 508 F.3d at 903.\\n\\n12 GONZALEZ ROMO V. BARR\\n\\nof the crime itself.17 We did not opine on whether the result\\nwould be the same for “a very small quantity”18 of the\\nsubstance. We agree with that analysis and see no\\ncompelling reason to deviate therefrom. On the contrary, we\\nsee strong reasons not to do so.\\n\\n First, we recognize that reasoning which excludes\\nsolicitation as to certain provisions could be said to apply\\nwhenever conspiracy or attempt are specifically mentioned in\\na statute. However, that would be unlikely when it comes to\\ncrimes involving moral turpitude. That is a separate concept\\nwith a long history19 which has been used by Congress in the\\nimmigration laws for over a century.20 It is doubtful that\\nCongress intended to give the phrase different scope in\\ndifferent provisions,21 and our cases do not suggest that it\\ndid.22\\n\\n\\n 17\\n See id. at 904.\\n 18\\n Id.\\n 19\\n The concept is deeply rooted in our law. See Jordan v. De George,\\n341 U.S. 223, 227, 71 S. Ct. 703, 705–06, 95 L. Ed. 886 (1951).\\n 20\\n See id. at 229–30, 71 S. Ct. at 707.\\n 21\\n See Merrill Lynch, Pierce, Fenner & Smith Inc. v. Dabit, 547 U.S.\\n71, 85–86, 126 S. Ct. 1503, 1513, 164 L. Ed. 2d 179 (2006); cf. Powerex\\nCorp. v. Reliant Energy Servs., Inc., 551 U.S. 224, 232, 127 S. Ct. 2411,\\n2417, 168 L. Ed. 2d 112 (2007).\\n 22\\n See, e.g., Islas-Veloz v. Whitaker, 914 F.3d 1249, 1250–51 (9th Cir.\\n2019); Gonzalez-Cervantes v. Holder, 709 F.3d 1265, 1269 (9th Cir.\\n2013); Morales v. Gonzales, 478 F.3d 972, 978 (9th Cir. 2007), abrogated\\non other grounds by Anaya-Ortiz v. Holder, 594 F.3d 673, 677–78 (9th\\nCir. 2010).\\n\\n GONZALEZ ROMO V. BARR 13\\n\\n Second, we recognize that solicitation is a type of\\ninchoate crime because it would not be a crime involving\\nmoral turpitude if the underlying offense (here possession for\\nsale of over four pounds of marijuana) was not a crime\\ninvolving moral turpitude. See Barragan-Lopez, 508 F.3d at\\n903. No doubt attempt and conspiracy are also inchoate\\ncrimes in that sense. Why then would Congress mention\\nthem specifically and not mention other inchoate crimes (like\\nsolicitation) when referring to crimes involving moral\\nturpitude? Of course, we cannot say for sure, and the\\nlegislative history, such as it is, does not make the answer\\nperspicuous. It just indicates that the addition of those words\\nwas clarifying. See 139 Cong. Rec. 15,275 (1993) (indicating\\nthat the language was intended to “make clear” that those\\nconcepts were covered); see also Immigration and Nationality\\nTechnical Corrections Act of 1994, Pub. L. 103–416, § 203,\\n108 Stat. 4305, 4311 (indicating that the language constitutes\\na clarification). However, when it comes to inchoate crimes\\nthat are bad enough to be considered crimes of moral\\nturpitude, we see little reason to think that Congress took this\\nsomewhat indirect approach to eliminating some of them\\nfrom consideration simply because it decided to make it clear\\nthat others did fall into that category. See Ali v. Fed. Bureau\\nof Prisons, 552 U.S. 214, 226–27, 128 S. Ct. 831, 840, 169 L.\\nEd. 2d 680 (2008). It is more likely that Congress was\\nfocused on the two that it mentioned and had no intention of\\nexcluding all others. See Marx v. Gen. Revenue Corp.,\\n568 U.S. 371, 385, 133 S. Ct. 1166, 1177, 185 L. Ed. 2d 242\\n(2013). And, while redundancy and surplusage are not\\nalways desirable, they are not anoetic. See Conn. Nat’l Bank\\nv. Germain, 503 U.S. 249, 253, 112 S. Ct. 1146, 1149, 117 L.\\nEd. 2d 391 (1992). In fact, it is not unusual to see\\nredundancy occur in clarifying enactments, which do run the\\nrisk of introducing other problems while focusing on the task\\n\\n14 GONZALEZ ROMO V. BARR\\n\\nof making sure that a particular problem does not arise. See\\nid. As the Court has said: “In any event, our hesitancy to\\nconstrue statutes to render language superfluous does not\\nrequire us to avoid surplusage at all costs. It is appropriate to\\ntolerate a degree of surplusage rather than adopt a textually\\ndubious construction that threatens to render the entire\\nprovision a nullity.” United States v. Atl. Research Corp.,\\n551 U.S. 128, 137, 127 S. Ct. 2331, 2337, 168 L. Ed. 2d 28\\n(2007). In the case at hand, the cost would be particularly\\nhigh because it would create an anomaly in application of the\\nstatutory scheme, which we doubt Congress intended when it\\nadopted its clarifying amendments.23 The anomaly can be\\nseen if we start with the rather uncontroversial reflection that\\n“[o]ur immigration law has generally treated aliens who are\\nalready on our soil (and who are therefore deportable) more\\nfavorably than aliens who are merely seeking admittance (and\\nwho are therefore excludable).” Servin-Espinoza v. Ashcroft,\\n309 F.3d 1193, 1198 (9th Cir. 2002). But removing\\nsolicitation of possession of over four pounds of marijuana\\nfrom the definition of a crime involving moral turpitude in\\n§ 1182(a)(2)(A)(i)(I), although solicitation is part of the\\ndefinition in § 1227(a)(2)(A)(i)(I), would turn that concept on\\nits head. That is because we already know that even a legal\\npermanent resident can be removed pursuant to\\n§ 1227(a)(2)(A)(i)(I) for soliciting a crime involving moral\\nturpitude. See Barragan-Lopez, 508 F.3d at 905. However,\\nif solicitation is removed from the definition of crime\\ninvolving moral turpitude in § 1182(a)(2)(A)(i)(I), no alien\\n(legal permanent resident or not) can be excluded for\\nsoliciting a crime involving moral turpitude, regardless of\\n\\n\\n 23\\n In general, we do attempt to avoid anomalies and try to view the\\nscheme at hand as “symmetrical and coherent.” Perez-Guzman v. Lynch,\\n835 F.3d 1066, 1074 (9th Cir. 2016) (internal quotation marks omitted).\\n\\n GONZALEZ ROMO V. BARR 15\\n\\nhow turpitudinous that crime may be. Again, we doubt that\\nCongress intended any such thing. In fact, in the context of\\nthis statutory scheme regarding who can enter and who can be\\nremoved, ascribing that intent to Congress would lead to a\\nresult so dubious that it would be nothing short of an internal\\ninconsistency amounting to an absurdity. See United States\\nv. Turkette, 452 U.S. 576, 580, 101 S. Ct. 2524, 2527, 69 L.\\nEd. 2d 246 (1981) (“[A]bsurd results are to be avoided and\\ninternal inconsistencies in the statute must be dealt with.”);\\nsee also McNeill v. United States, 563 U.S. 816, 822, 131 S.\\nCt. 2218, 2223, 180 L. Ed. 2d 35 (2011).\\n\\n The discussion above indicates that we can stop at step\\none of the Chevron process24 because Congress’ true intent is\\nclear to us. However, to the extent that there is some\\nperceived uncertainty25 regarding the meaning and reach of\\nthe statute, that ambiguity26 would direct us to step two.27 At\\nthat step, it is clear that the BIA’s resolution of the\\n\\n\\n\\n\\n 24\\n 467 U.S. at 842–43, 104 S. Ct. at 2781.\\n 25\\n See, e.g., Mielewczyk, 575 F.3d at 996–98; Barragan-Lopez,\\n508 F.3d at 902–05; Coronado-Durazo, 123 F.3d at 1324–26; see also\\nMizrahi v. Gonzales, 492 F.3d 156, 164–67 (2d Cir. 2007); Peters v.\\nAshcroft, 383 F.3d 302, 305–09 (5th Cir. 2004).\\n 26\\n We note that some ambiguity in statutory language does not\\npreclude resolution of that ambiguity by the BIA. See Chevron, 467 U.S.\\nat 843, 104 S. Ct. at 2782; Perez-Guzman, 835 F.3d at 1076 n.5.\\n 27\\n Chevron, 467 U.S. at 843–44, 104 S. Ct. at 2781–82.\\n\\n16 GONZALEZ ROMO V. BARR\\n\\nuncertainty is permissible and reasonable. Indeed, it appears\\nto us that it is the best resolution.28\\n\\n Thus, we deny the petition.\\n\\n CONCLUSION\\n\\n We hold that a conviction in Arizona for solicitation to\\npossess at least four pounds of marijuana for sale constitutes\\na crime involving moral turpitude for purposes of\\n§ 1182(a)(2)(A)(i)(I), and, therefore, Gonzalez was\\ninadmissible. See § 1101(a)(13)(C)(v).\\n\\n Petition DENIED.\\n\\n\\n\\n\\n 28\\n We have not overlooked Gonzalez’s argument that the BIA should\\nbe accorded less deference because it changed its position about what it\\nthought our case law required. See Matter of Vo, 25 I. & N. Dec. at 429\\nn.4. However, as we see it, an agency’s reflections on what it thought we\\nhave said, or might say, is quite different from a determination of what the\\nlaw is. In any event: “In the circumstances of this case, where the\\nagency’s interpretation of a statute is at least as plausible as competing\\nones, there is little, if any, reason not to defer to its construction.” Good\\nSamaritan Hosp. v. Shalala, 508 U.S. 402, 417, 113 S. Ct. 2151, 2161,\\n124 L. Ed. 2d 368 (1993).\\n\\n GONZALEZ ROMO V. BARR 17\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4067,"cells":{"id":{"kind":"number","value":4648244,"string":"4,648,244"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Dennis Claiborne v. Blauser"},"decision_date":{"kind":"string","value":"2019-08-14"},"docket_number":{"kind":"string","value":"16-16077"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Sidney R. Thomas, Chief Judge, Richard A. Paez, Circuit Judge, and Gary Feinerman, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PAEZ, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n DENNIS GERALD CLAIBORNE, No. 16-16077\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:10-cv-02427-VAP\\n\\n BLAUSER, Correctional Officer;\\n S. MARTIN, Correctional ORDER AND\\n Officer, AMENDED\\n Defendants-Appellees. OPINION\\n\\n Appeal from the United States District Court\\n for the Eastern District of California\\n Virginia A. Phillips, Chief District Judge, Presiding\\n\\n Argued and Submitted February 6, 2019\\n San Francisco, California\\n\\n Filed June 28, 2019\\n Amended August 14, 2019\\n\\nBefore: Sidney R. Thomas, Chief Judge, Richard A. Paez,\\n Circuit Judge, and Gary Feinerman, * District Judge.\\n\\n\\n\\n\\n *\\n The Honorable Gary Feinerman, United States District Judge for\\nthe Northern District of Illinois, sitting by designation.\\n\\f2 CLAIBORNE V. BLAUSER\\n\\n Order;\\n Opinion by Judge Paez\\n\\n\\n SUMMARY **\\n\\n\\n Prisoner Civil Rights\\n\\n The panel reversed the district court’s denial of a motion\\nfor a new trial and remanded in an action brought pursuant\\nto 42 U.S.C. § 1983 by a convicted state inmate who alleged\\nthat he was shackled without justification during his three-\\nday trial on his Eighth Amendment excessive force and\\ndeliberate indifference to medical needs claims.\\n\\n The panel first noted that although the inmate did not\\nobject to the shackling during trial, he raised the issue in\\nsupport of his motion for a new trial. The panel applied plain\\nerror review.\\n\\n The panel held that because the inmate’s dangerousness\\nand flight risk were central issues at the trial, the district\\ncourt plainly erred in allowing him to be visibly shackled\\nwithout any showing of a sufficient need for such restraints.\\nThe panel held that on remand, the district court would have\\ndiscretion to impose shackling during the new trial, but it\\ncould only do so after a full hearing at which officers showed\\na compelling need for security and the court considered any\\nless restrictive alternatives.\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f CLAIBORNE V. BLAUSER 3\\n\\n COUNSEL\\n\\nJeremy M. Christiansen (argued), Gibson Dunn & Crutcher\\nLLP, Washington, D.C.; J. Brett Bylund and Blaine H.\\nEvanson, Gibson Dunn & Crutcher LLP, Irvine, California;\\nfor Plaintiff-Appellant.\\n\\nJaime Ganson (argued) and Arthur B. Mark III, Deputy\\nAttorneys General; Neah Huynh, Acting Supervising\\nDeputy Attorney General; Monica N. Anderson, Senior\\nAssistant Attorney General; Xavier Becerra, Attorney\\nGeneral; Office of the Attorney General, Sacramento,\\nCalifornia; for Defendants-Appellees.\\n\\n\\n ORDER\\n\\n The opinion filed in this case on June 28, 2019 is\\namended by replacing the text of footnote 7 on page 28,\\nClaiborne v. Blauser, 928 F.3d 794, 810 n.7 (9th Cir. 2019),\\nwith the following paragraphs:\\n\\n Because we remand for a new trial on the\\n shackling claim, we do not address the merits\\n of Claiborne’s evidentiary arguments that the\\n district court erred in barring his testimony\\n about the ADA, preventing him from\\n introducing or testifying about his medical\\n records, and denying his request for a\\n medical expert. The district court may revisit\\n these issues if Claiborne raises them again at\\n the new trial.\\n\\n We note, however, that the district court\\n appears to misstate the law when it denied\\n\\f4 CLAIBORNE V. BLAUSER\\n\\n Claiborne’s request for a medical expert\\n under Federal Rule of Evidence 706(a).\\n Although the district court correctly\\n recognized that Rule 706(a) provides\\n discretion to appoint a neutral expert witness,\\n see McKinney v. Anderson, 924 F.2d 1500,\\n 1511 (9th Cir. 1991), vacated on other\\n grounds sub nom. Helling v. McKinney,\\n 502 U.S. 903 (1991), judgment reinstated,\\n 959 F.2d 853 (9th Cir. 1992), aff’d, 509 U.S.\\n 25 (1993), the district court seemed to\\n categorically limit the relevance of a medical\\n expert to testifying about a plaintiff’s current\\n condition. Yet courts have regularly\\n considered requests for and appointed experts\\n to review medical records and testify about\\n prior medical needs and treatment in\\n deliberate indifference cases. See Gorton v.\\n Todd, 793 F.Supp.2d 1171, 1179–81 (E.D.\\n Cal. 2011) (collecting cases). Moreover, a\\n medical expert can help with factfinding in\\n excessive force claims because “the extent of\\n injury suffered by an inmate is one factor that\\n may suggest ‘whether the [defendant’s] use\\n of force could plausibly have been thought\\n necessary’ in a particular situation.” Hudson\\n v. McMillian, 503 U.S. 1, 7 (1992). If\\n Claiborne renews his request for appointment\\n of a neutral medical expert on retrial, the\\n district court should weigh these\\n considerations in exercising her discretion.\\n See McKinney, 924 F.2d at 1511; see also\\n Gorton, 793 F.Supp.2d at 1185–86.\\n\\nThe Amended Opinion is filed concurrently with this order.\\n\\f CLAIBORNE V. BLAUSER 5\\n\\n With the filing of the Amended Opinion, the panel has\\nunanimously voted to deny the Defendants-Appellees’\\npetition for panel rehearing. No further petitions for\\nrehearing may be filed.\\n\\n\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n The law has long forbidden the routine use of visible\\nshackling during a criminal defendant’s trial. Deck v.\\nMissouri, 544 U.S. 622, 626 (2005). Visible shackling\\nundermines the presumption of innocence, impedes the\\njury’s factfinding process, hampers presentation of a\\ndefense, and affronts the dignity and decorum of judicial\\nproceedings. Id. at 630–32. In this civil rights case under\\n42 U.S.C. § 1983, we consider whether the unjustified\\nshackling of a convicted state inmate during his three-day\\ntrial on Eighth Amendment excessive force and deliberate\\nindifference claims deprived him of a fair trial in violation\\nof the federal constitution. 1 Although the inmate did not\\nobject to the shackling during trial, he raised the issue in\\nsupport of his motion for a new trial, which the district court\\ndenied.\\n\\n We hold that the district court abused its discretion in\\ndenying a new trial. Because the inmate’s dangerousness\\nand flight risk were central issues at the trial, the district\\ncourt plainly erred in allowing him to be visibly shackled\\nwithout any showing of a sufficient need for such restraints.\\n\\n\\n 1\\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\\n\\f6 CLAIBORNE V. BLAUSER\\n\\nSee Tyars v. Finner, 709 F.2d 1274, 1284–85 (9th Cir. 1983).\\nWe therefore reverse and remand for a new trial.\\n\\n I.\\n\\n This appeal arises out of a lawsuit filed by Dennis Gerald\\nClaiborne who, proceeding pro se, sued Correctional\\nOfficers Jemini Blauser, Greg Martin, and other individual\\nofficials under Section 1983 for the use of excessive force\\nand deliberate indifference to his medical needs.\\n\\n A.\\n\\n Claiborne is a 63-year-old California state prison inmate\\nin the custody of the California Department of Corrections\\nand Rehabilitation (“CDCR”). He is serving a 60-years-to-\\nlife sentence under California’s Three Strikes Law for\\nattempted burglary and receipt of stolen property. People v.\\nClaiborne, No. B260391, 2015 WL 5146746, at *1 (Cal. Ct.\\nApp. Sept. 2, 2015).\\n\\n Claiborne is mobility impaired due to a right knee\\nreplacement in 2007 and ensuing chronic problems with that\\nknee. Given his condition, Claiborne participates in the\\nCDCR Disability Placement Program and receives certain\\naccommodations in prison: he is allowed to use a cane; he is\\nrestricted to housing on the lower level, with no stairs; and,\\nwhen escorting Claiborne within the institution, correctional\\nstaff must use “waist chains” and choose “relatively level\\nterrain and no obstructions in the path of travel.” Waist\\nchains are different from traditional handcuffs; whereas the\\nlatter are typically applied behind the back, the former allow\\na mobility impaired inmate to keep his hands at his sides\\nduring an escort, which lets him use any prescribed\\naccommodation devices like a cane. CDCR documents and\\nconveys information about Claiborne’s accommodations to\\n\\f CLAIBORNE V. BLAUSER 7\\n\\nprison officials through physician orders known as medical\\naccommodation “chronos” in his file. Claiborne also wears\\na green vest to alert officers that he is mobility impaired.\\n\\n The incident between Claiborne and Officers Blauser\\nand Martin took place on May 3, 2010, while Claiborne was\\nhoused at California’s High Desert State Prison, a Level\\nFour security prison. As Claiborne was waiting for his\\nmedication in the morning “pill line,” Correctional Officer\\nDaniel McBride, stationed in an observation tower, believed\\nhe saw Claiborne socializing with other inmates in the line.\\nOfficer McBride called Officer Blauser, who was working\\nin Claiborne’s housing unit along with her partner, Officer\\nMartin, and asked Blauser to counsel Claiborne for\\nunnecessarily lingering in the pill line. After waiting thirty\\nminutes in the pill line, Claiborne received his medication\\nand returned to his housing unit where he was admonished\\nby Officer Blauser. She decided to “put a cap” on\\nClaiborne’s door, meaning he would spend the rest of the\\nday locked up in his cell. Because Claiborne had intended\\nto present at a Bible study group later that day, he asked to\\nspeak with the sergeant, Officer Blauser’s supervisor, to\\ncontest the punishment and explain that he had been properly\\nwaiting in line and not socializing.\\n\\n The account of the facts diverge drastically from there.\\nAccording to Claiborne, he was acting respectfully toward\\nOfficer Blauser when she told him to “cuff up.” He\\ncomplied and, as Officers Blauser and Martin started to\\nescort him, Claiborne informed them that it was difficult for\\nhim to walk and use his cane with his hands cuffed behind\\nhis back. He mentioned his chrono for waist chains but\\nOfficer Blauser told Officer Martin to take Claiborne’s cane.\\nOfficer Martin said he would help hold Claiborne up as the\\nthree of them walked to the program office to speak with the\\n\\f8 CLAIBORNE V. BLAUSER\\n\\nsergeant. Claiborne, however, had difficulty walking with\\nOfficers Blauser and Martin. They escorted him straight\\nacross the yard, rather than along the sidewalk circling the\\nyard, despite Claiborne’s chrono providing for level terrain.\\nBecause the yard was uneven, Claiborne hyperextended his\\nright leg, causing his knee to give out partway across the\\nyard. Claiborne lost balance and shifted rightward, causing\\nOfficer Blauser to order him to stop resisting. Claiborne\\ntried to explain that he was not resisting and that his knee\\nwas bothering him because the officers were pulling him too\\nquickly.\\n\\n When the three were almost at the program office,\\nClaiborne’s right leg hyperextended again as he tried to pick\\nthe leg up three to five inches from the dirt yard onto the\\npavement. Because he had no cane to catch himself,\\nClaiborne leaned to his right again, causing Officer Blauser\\nto shout “he’s resisting” and pull him down to the ground.\\nShe jumped on his right side, including his replaced knee,\\nand pulled his hair and hit him in the face a few times. Other\\nofficers quickly rushed to the scene and Claiborne heard\\nindividuals ask, “Where’s his cane?” and “Why isn’t he in\\nwaist chains?” Claiborne was eventually taken into a\\nholding cell and then interviewed by a sergeant, which was\\nrecorded by a video camera. Before turning on the camera\\nto record the interview, the sergeant warned Claiborne that\\nif he reported excessive force, he would be taken to the\\n“hole,” in other words, administrative segregation, for an\\nunknown amount of time. Worried about whether his\\nmedical needs would be met in the hole, Claiborne\\neventually stated on camera, after extended back-and-forth\\nwith the sergeant, that there was no excessive force used\\nagainst him.\\n\\f CLAIBORNE V. BLAUSER 9\\n\\n Officers Blauser and Martin presented a different\\naccount. They were both aware of Claiborne’s mobility\\nimpairment at the time of the incident. According to Officer\\nBlauser, while she counseled Claiborne for lingering in the\\npill line, he became “really aggravated” and started raising\\nhis voice at her, causing her to feel uncomfortable.\\nClaiborne then walked toward her while holding, not using,\\nhis cane. Because she did not feel safe, Officer Blauser told\\nClaiborne to “cuff up,” and he immediately turned around\\nand complied. She asked her partner, Officer Martin, to\\nassist her with escorting Claiborne to the program office.\\nOfficer Martin took Claiborne’s cane, and they each\\nsupported him by holding onto his bicep or arm on each side.\\nNeither Officer Blauser nor Officer Martin recalled\\nClaiborne saying anything about needing to use waist chains.\\nMoreover, because it would have taken a few extra minutes\\nto obtain waist chains, Officer Blauser decided to use\\nhandcuffs due to Claiborne’s aggression and defiance of her\\norder to return to his cell.\\n\\n According to Officers Blauser and Martin, they escorted\\nClaiborne straight across the yard because it was the quickest\\nand most direct path to the program office and avoided\\nwalking amongst other inmates. They escorted him slowly\\nand did not perceive any problems with his walking. Rather,\\nthey believed that Claiborne tried to break away from them\\ntwice, once halfway across the yard, and a second time close\\nto the program office. At first, Officer Blauser ordered\\nClaiborne not to pull his arm away but he continued to act\\naggressively, yelling and trying to pull away from her.\\nWhen Claiborne pulled his arm away from her a second\\ntime, Officer Blauser decided to pull him down and called a\\n“code one” over the radio. Officers Blauser and Martin used\\ntheir weight to hold Claiborne down on the ground. They\\ndenied that Officer Blauser jumped on Claiborne, pulled his\\n\\f10 CLAIBORNE V. BLAUSER\\n\\nhair or punched him in the face. After other officers escorted\\nClaiborne away, Sergeant Officer Kenneth Gullion followed\\nprotocol and conducted a video-recorded “use-of-force”\\ninterview, asking Claiborne about what took place. Sergeant\\nGullion did not remember what happened prior to\\ninterviewing Claiborne on video. A separate officer wrote\\nClaiborne up for a rule violation, resisting a peace officer.\\n\\n Shortly after the incident, a nurse examined Claiborne\\nand completed a medical report, noting that he had two\\nabrasions, one each on his left knee and left cheek.\\nAfterward, Claiborne experienced more problems with his\\nright knee, and doctors determined that he had significant\\ninjuries, including bursitis. Following a series of\\nevaluations, Claiborne was assessed to have “[f]ailed right\\ntotal knee arthroplasty.” He underwent a revision procedure\\nin 2012, but was told that his knee did not respond properly\\nto the surgery and could not be fixed any more. Claiborne\\nalso had surgery on one of his shoulders in July 2015, when\\nhe was told that he waited too long to fix it.\\n\\n B.\\n\\n After exhausting the prison’s administrative process,\\nClaiborne filed suit in district court. Proceeding pro se, he\\nsued Officers Blauser, Martin and other individual officers\\nunder Section 1983 for various claims, including the use of\\nexcessive force and deliberate indifference to his medical\\nneeds in violation of his Eighth Amendment rights. The\\nother named defendant officers and other claims, including\\nbattery, negligent infliction of emotional distress and\\nintentional infliction of emotional distress, were eventually\\ndismissed from the case. After the district court denied\\ncross-motions for summary judgment, the case proceeded to\\ntrial on the two Eighth Amendment claims against Officers\\nBlauser and Martin.\\n\\f CLAIBORNE V. BLAUSER 11\\n\\n The trial lasted three days, at the start of which the\\ndistrict court noted, outside of the jury’s presence, that\\nClaiborne was shackled. Claiborne testified on his own\\nbehalf, and two fellow inmates also testified on his behalf.\\nThe defense presented testimony from six officers including\\nOfficers Blauser and Martin, the tower guard officer\\n(McBride), and Sergeant Gullion. The jury reached a verdict\\nfor both defendants on both claims.\\n\\n Claiborne filed a timely motion for new trial and relief\\nfrom judgment, raising various arguments. Most relevant to\\nour analysis, Claiborne argued that he was entitled to a new\\ntrial under Federal Rule of Civil Procedure 59(a) because he\\nwas visibly shackled while litigating his case in front of the\\njury. He also argued for relief from judgment under Federal\\nRule of Civil Procedure 60(b) because the district court\\nbarred his testimony about the Americans with Disabilities\\nAct (“ADA”), prevented him from introducing or testifying\\nabout his medical records, and denied his request for a\\nmedical expert. The district court dismissed all the\\narguments and denied the motion.\\n\\n II.\\n\\n When reviewing denial of a motion for new trial under\\nFederal Rule of Civil Procedure 59(a), our default is to\\nreview for abuse of discretion. See Hung Lam v. City of San\\nJose, 869 F.3d 1077, 1084 (9th Cir. 2017). But here,\\nClaiborne did not object to his shackling at trial so we must\\naddress the appropriate standard of review.\\n\\n Errors not objected to at trial are generally subject to\\nwaiver or forfeiture. “Forfeiture is the failure to make a\\ntimely assertion of a right, whereas waiver is the intentional\\nrelinquishment or abandonment of a known right.” United\\nStates v. Perez, 116 F.3d 840, 845 (9th Cir. 1997) (en banc)\\n\\f12 CLAIBORNE V. BLAUSER\\n\\n(internal quotation marks omitted); see also Crowley v.\\nEpiCept Corp., 883 F.3d 739, 748 (9th Cir. 2018) (per\\ncuriam) (applying same standard in civil context).\\n“Forfeited rights are reviewable for plain error, while waived\\nrights are not.” Id.\\n\\n Under plain error review, we may reverse only where:\\n(1) there was an error; (2) the error was obvious; (3) the error\\naffected substantial rights; and, (4) the error seriously affects\\nthe fairness, integrity, or public reputation of judicial\\nproceedings. See C.B. v. City of Sonora, 769 F.3d 1005,\\n1018–19 (9th Cir. 2014) (en banc). In other words, plain\\nerror requires reversal where “review is necessary to prevent\\na miscarriage of justice.” Draper v. Rosario, 836 F.3d 1072,\\n1085 (9th Cir. 2016) (internal quotation marks omitted).\\n\\n Defendants argue that Claiborne forfeited the error\\nbecause he failed to object to his shackling at trial, and so,\\nhis claim should be subject to plain error review. 2 Claiborne\\ncontends that plain error is inapplicable because he timely\\nraised the shackling issue in his motion for a new trial, and\\nthe district court addressed the claim on the merits, holding\\nthat it would have imposed shackling over any objection\\n\\n 2\\n We find no merit to defendants’ alternative argument that\\nClaiborne invited the error. We impose a high standard for applying the\\ninvited error doctrine. See United States v. Lindsey, 634 F.3d 541, 555\\n(9th Cir. 2011) (noting that “in order for the invited error doctrine to\\napply, a [party] must both invite the error and relinquish a known right”\\n(citing Perez, 116 F.3d at 845)). Claiborne’s mention of his shackles in\\nhis opening statement, read in context, suggests that he was trying to\\nintroduce himself to the jury and not hide the fact that he is incarcerated.\\nThere is no indication in the record that Claiborne, who was pro se in the\\ndistrict court, “both invited the error, and relinquished a known right” to\\na fair trial without shackling. Perez, 116 F.3d at 845.\\n\\f CLAIBORNE V. BLAUSER 13\\n\\nbecause Claiborne is a convicted felon serving a lengthy\\nprison sentence.\\n\\n The district court’s denial of a motion for a new trial is\\nreversible if the court made a mistake of law. Molski v. M.J.\\nCable, Inc., 481 F.3d 724, 729 (9th Cir. 2007). As we\\nelaborate below, we agree that the district court disregarded\\nthe law on shackling. However, in prior cases involving\\nlegal claims that were not raised during trial, we still\\nundertook plain error review when reviewing a denial of a\\nmotion for a new trial. 3 See, e.g., Settlegoode v. Portland\\nPub. Sch., 371 F.3d 503, 517 (9th Cir. 2004) (“There is an\\neven ‘high[er] threshold’ for granting a new trial where, as\\nhere, [the moving party] failed to object to the alleged\\nmisconduct during trial.” (first alteration in original)\\n(quoting Kaiser Steel Corp. v. Frank Coluccio Constr. Co.,\\n785 F.2d 656, 658 (9th Cir. 1986))); Hemmings v. Tidyman’s\\nInc., 285 F.3d 1174, 1193 (9th Cir. 2002) (“We will review\\nfor plain or fundamental error, absent a contemporaneous\\nobjection . . .”).\\n\\n The policy behind plain error review is to ensure that the\\nparties and district court are on notice as to any alleged errors\\nso that the court may address the objection, correct any error,\\nand create a proper record for appeal. See Hemmings,\\n285 F.3d at 1193 (observing that “the trial judge is in a\\nsuperior position to evaluate the likely effect of the alleged\\nmisconduct and to fashion an appropriate remedy”); see also\\n 3\\n This is analogous to our review of instructional error claims when\\nthere was no objection at trial. 11 Charles A. Wright et al., Federal\\nPractice and Procedure § 2805 (3d ed. 2012). Under Federal Rule of\\nCivil Procedure 51(d)(2), we review a challenge to civil jury instructions\\nfor plain error in the absence of a timely objection. C.B., 769 F.3d\\nat 1017–19.\\n\\f14 CLAIBORNE V. BLAUSER\\n\\nUnited States v. $11,500.00 in U.S. Currency, 869 F.3d\\n1062, 1075 (9th Cir. 2017).\\n\\n Here, the district court’s holding on the merits—that it\\nwould have imposed shackling over any timely objection—\\ndoes give us some pause as to the standard of review. 4 We\\nhave recognized one limited exception to plain error review\\nin the civil context, the “pointless formality” exception.\\nChess v. Dovey, 790 F.3d 961, 970–72 (9th Cir. 2015); see\\nalso Shorter v. Baca, 895 F.3d 1176, 1182–83 (9th Cir.\\n2018). In Chess, we declined to apply plain error where the\\nplaintiff was pro se and “the district court and opposing party\\nwere fully aware of the potential problem with, and would-\\nbe objection to, [a jury] instruction,” even though the\\nplaintiff did not raise the issue himself. 790 F.3d at 964.\\nBecause the district court and defendants were on notice, we\\nconcluded that “any objection by Chess would have been\\n‘superfluous and futile,’” id. at 972 (quoting Obsidian Fin.\\nGrp. v. Cox, 740 F.3d 1284, 1289 (9th Cir. 2014)), and we\\ndeclined to “punish a pro se litigant with plain error rather\\nthan de novo review simply because he failed to say the\\nwords ‘I object,’” id. at 971.\\n\\n Claiborne, too, was pro se and the district court’s post-\\ntrial ruling suggests that any timely objection would have\\n\\n 4\\n The district court’s ruling distinguishes Claiborne’s situation from\\nthat in Hemmings, in which the defendant moved for a new trial based\\non inappropriate comments made by the plaintiffs’ counsel in closing\\nargument. 285 F.3d at 1192. The district court noted that it remembered\\nthe comment and “would have sustained an objection had one been made\\nbecause it was totally improper.” Id. at 1192–93. Because a timely\\nobjection would have made a difference and changed the district court’s\\nactions, application of plain error review was appropriate under the\\nrationale of giving the trial judge an opportunity to cure any prejudice.\\nId. at 1193. That concern does not apply here.\\n\\f CLAIBORNE V. BLAUSER 15\\n\\nbeen futile. We recognize, however, the policy concerns\\nbehind the plain error doctrine and therefore decline to\\nreview for abuse of discretion. See Hemmings, 285 F.3d at\\n1193 (noting that “allowing a party to wait to raise the error\\nuntil after the negative verdict encourages that party to sit\\nsilent in the face of claimed error”). Accordingly, we review\\nfor plain error.\\n\\n III.\\n\\n The district court must grant a motion for a new trial\\nwhere “the verdict is against the weight of the evidence,”\\n“the damages are excessive” or, “for other reasons, the trial\\nwas not fair to the [moving] party.” Montgomery Ward &\\nCo. v. Duncan, 311 U.S. 243, 251 (1940). Here, we consider\\nan argument based on the latter. Claiborne, now represented\\nby pro bono counsel, argues that he was deprived of a fair\\ntrial because he was visibly shackled in front of the jury\\nthroughout the three-day trial, and that the district court erred\\nby denying his Rule 59(a) motion for a new trial. Even under\\nthe more demanding plain error standard, we agree that the\\ndistrict court erred in denying Claiborne’s motion for a new\\ntrial.\\n\\n A.\\n\\n To understand why the district court plainly erred, we\\nfirst review the case law on visible shackling.\\n\\n It is well established that due process “forbids the use of\\nvisible shackl[ing] . . . during the guilt phase [of a criminal\\ntrial], unless that use is ‘justified by an essential state\\ninterest.’” Deck, 544 U.S. at 624 (citing Holbrook v. Flynn,\\n475 U.S. 560, 568–69 (1986); Illinois v. Allen, 397 U.S. 337,\\n343–44 (1970)). The Supreme Court has identified “three\\nfundamental legal principles” which are “undermined” or\\n\\f16 CLAIBORNE V. BLAUSER\\n\\n“diminished” by the use of visible restraints. Id. at 630–31.\\nFirst, the accused enjoys a presumption of innocence, which\\nvisible shackling undermines by suggesting that the\\ndefendant need be restrained. Id. at 630. This in turn\\n“undermines . . . the related fairness of the factfinding\\nprocess.” Id. Second, the accused has a right to a\\nmeaningful defense from counsel. Id. at 631. Restraints\\nmight physically impair communication between a\\ndefendant and his counsel, or impair the defendant’s\\nparticipation in his own defense. Id. “Third, judges must\\nseek to maintain a judicial process that is a dignified [one].”\\nId. Because “[t]he courtroom’s formal dignity . . . includes\\nthe respectful treatment of defendants,” “the use of shackles\\nat trial ‘affront[s]’ the ‘dignity and decorum of judicial\\nproceedings that the judge is seeking to uphold.’” Id.\\n(quoting Allen, 397 U.S. at 344).\\n\\n In Deck, the Court held that these considerations apply\\nwith equal force to penalty proceedings in capital cases, even\\nthough “the presumption of innocence no longer applies.”\\nId. at 632. Hence, before imposing restraints on criminal\\ndefendants, the trial court must take into account the\\nparticular circumstances of each case, including “special\\nsecurity needs or escape risks[] related to the defendant,” id.\\nat 633. Where the court orders visible shackling without\\nadequate justification, “the defendant need not demonstrate\\nactual prejudice to make out a due process violation,” and\\nthe burden is on the government to “prove beyond a\\nreasonable doubt that the shackling error complained of did\\nnot contribute to the verdict obtained.” Id. at 635 (alterations\\nand internal quotation marks omitted).\\n\\n Prior to Deck, we have long recognized that the\\nprohibition against routine visible shackling applies even\\nwhen the presumption of innocence does not, including in\\n\\f CLAIBORNE V. BLAUSER 17\\n\\nthe civil context. In Duckett v. Godinez, we anticipated Deck\\nby holding that “shackling a defendant during a [capital]\\nsentencing hearing before a jury is an inherently prejudicial\\npractice which comports with due process only when used\\nas a last resort to protect an essential state interest[,] such as\\nmaintaining public safety or assuring the decorum of the\\nproceedings.” 67 F.3d 734, 747 (9th Cir. 1995). We have\\nthus long required trial courts to engage in a two-step process\\nbefore shackling a criminal defendant: “[f]irst, the court\\nmust be persuaded by compelling circumstances” that the\\nmeasure is necessary to “maintain security[;]” and\\n“[s]econd, the court must ‘pursue less restrictive alternatives\\nbefore imposing physical restraints.” Id. at 748 (internal\\nquotation marks omitted). We held that it was error to permit\\nshackling solely on the basis that the defendant was a\\nconvicted felon, and so remanded for the district court to\\ndetermine whether this constitutional error was harmless. Id.\\nat 749–50.\\n\\n Notably, we reached our holding in Duckett “by analogy\\nto the treatment of the shackling issue in civil cases.” Id.\\nat 748 (citing Tyars, 709 F.2d at 1284–85; Lemons v.\\nSkidmore, 985 F.2d 354, 356–58 (7th Cir. 1993); and\\nHolloway v. Alexander, 957 F.2d 529, 530 (8th Cir. 1992)).\\nAs recounted below, this court has recognized the inherent\\nprejudice associated with unjustified shackling in civil\\nproceedings as early as 1983, and other circuits have relied\\non our reasoning to expand the prohibition on routine\\nshackling to other civil contexts.\\n\\n In Tyars, the petitioner challenged the use of restraints to\\nbind him in the presence of the jury during his involuntary\\ncivil commitment proceedings. 709 F.2d at 1276, 1284. We\\nobserved that “[i]t requires no great extension of Supreme\\nCourt precedent to conclude that this may have violated his\\n\\f18 CLAIBORNE V. BLAUSER\\n\\nrights under the due process clause.” Id. at 1284. We\\nreasoned:\\n\\n Although the criminal case precedents do not\\n necessarily apply in a civil proceeding, we\\n find them persuasive. The likelihood of\\n prejudice inherent in exhibiting the subject of\\n a civil commitment hearing to the jury while\\n bound in physical restraints, when the critical\\n question the jury must decide is whether the\\n individual is dangerous to himself or others,\\n is simply too great to be countenanced\\n without at least some prior showing of\\n necessity. In the absence of any such\\n demonstrable or articulable necessity, and in\\n the absence of any showing that less\\n restrictive means not embodying the same\\n potential for prejudice could have maintained\\n order in the courtroom, the circumstances\\n deprived the proceeding of the appearance of\\n evenhanded justice which is at the core of due\\n process.\\n\\nId. at 1285 (emphasis added) (alterations, internal quotation\\nmarks and citation omitted). We instructed the district court\\non remand to “determine whether the State has any\\njustification . . . to support the necessity of physically\\nrestraining Tyars in the presence of the jury.” Id. at 1285–\\n86.\\n\\n Since Tyars, at least four other circuits have adopted a\\nsimilar test when evaluating the use of shackling in civil\\ntrials. In Lemons, the Seventh Circuit held that it was\\nimpermissible to rely solely on the opinion of a state\\ncorrections officer to shackle a plaintiff prison inmate in a\\n\\f CLAIBORNE V. BLAUSER 19\\n\\nSection 1983 action alleging excessive force by corrections\\nofficers. 985 F.2d at 356. The plaintiff claimed that\\ndefendant officers attacked and beat him severely in his cell,\\nwhile defendants asserted that the plaintiff was “ranting and\\nraving,” and had to be subdued. Id. at 355. The Lemons\\ncourt cited with approval Tyars, id. at 356 n.2, and concluded\\nthat, as in the criminal context, the appearance of a civil\\nplaintiff in handcuffs and leg irons “suggested to the jury that\\nthe plaintiff was dangerous and violent, so that whatever\\nforce the guards had used was probably necessary, and not\\nexcessive,” id. at 359. “[S]hackles inevitably prejudiced the\\njury” because “plaintiff’s tendency towards violence was at\\nissue.” Id. at 357. The court therefore remanded for a new\\ntrial and required a hearing to determine what, if any,\\nrestraints were necessary. Id. at 359.\\n\\n The Eighth Circuit in Holloway also recognized that, as\\na general rule, inmate civil plaintiffs should not have to\\nappear in court in shackles unless there was a showing of\\nneed and steps to mitigate any potential prejudice. 957 F.2d\\nat 530. The court found no prejudice in Holloway because\\nthe plaintiff challenged the constitutionality of living\\nconditions in the state prison, which did not raise the\\nplaintiff’s dangerousness as a merits issue at trial. Id.\\n(noting that “[t]he fact that Holloway and his witnesses were\\nsecurity risks inside the courtroom was simply not a factor\\nrelevant to [the jury’s] decision” of whether the “living\\nconditions inside the prison were cruel and inhumane”).\\n\\n Relying on Tyars, Lemons and Holloway, the Second\\nCircuit held that the Supreme Court’s concerns with\\nshackling in criminal proceedings apply to parties in civil\\nsuits, that physical restraints must be justified on the basis of\\nsafety or security concerns, and that the court must take steps\\nto minimize prejudice resulting from the presence of the\\n\\f20 CLAIBORNE V. BLAUSER\\n\\nrestraints. See Davidson v. Riley, 44 F.3d 1118, 1122–23 (2d\\nCir. 1995). Like the plaintiff in Lemons, Davidson was a pro\\nse state prisoner who appeared throughout his six-day trial\\nin his Section 1983 lawsuit in handcuffs and leg irons. Id.\\nat 1119. Even though Davidson’s claim involved\\ncorrectional officers reading his legal mail, which did not\\nbear on “either a propensity toward violence or risk of\\nescape,” the court still found the potential for prejudice to be\\nsignificant because “the verdict apparently was to turn on\\nwhether the jury would believe Davidson and his prisoner-\\nwitnesses or the [prison] witnesses.” Id. at 1126. After\\ndetermining that the unnecessary imposition of restraints\\nwas not harmless, the court vacated the judgment, remanded\\nfor a hearing on the propriety of restraints, and ordered a new\\ntrial. Id. at 1126–27.\\n\\n The Third Circuit most recently joined our court and the\\nSecond, Seventh and Eighth Circuits to hold that “requiring\\na party in a civil trial to appear in shackles ‘may well deprive\\nhim of due process unless the restraints are necessary.’”\\nSides v. Cherry, 609 F.3d 576, 581 (3d Cir. 2010) (quoting\\nDavidson, 44 F.3d at 1122). As in Davidson, “the core issue\\nin Sides’[s] case was credibility.” Id. at 584. The court\\nultimately concluded that Sides’s shackling was harmless\\nbecause the trial court took multiple steps to mitigate any\\nprejudice. Id. at 584–85. Notably, however, the court\\nrejected an alternative argument that the shackling was\\nharmless “simply because [Sides’s] propensity for violence\\nwas not directly at issue in the case.” Id. at 584.\\n\\n Thus, Tyars, Lemons and Holloway held that where a\\nplaintiff’s dangerousness is a merits issue, visible shackling\\nviolates due process unless justified on a case-by-case basis\\nand steps are taken to mitigate prejudice. And, Davidson and\\n\\f CLAIBORNE V. BLAUSER 21\\n\\nSides recognized that prejudice may also arise where a core\\nissue in the civil action is credibility.\\n\\n B.\\n\\n Against this backdrop, we evaluate (1) whether there was\\nerror, (2) whether the error was obvious, (3) whether the\\nerror affected substantial rights, and (4) whether the error\\nseriously affects the fairness, integrity, or public reputation\\nof judicial proceedings. C.B., 769 F.3d at 1018–19. 5\\n\\n 1.\\n\\n We readily conclude that the district court erred in\\nallowing Claiborne to appear before the jury while shackled\\nthroughout his three-day trial without first determining\\nwhether the shackles were necessary.\\n\\n As in Duckett and Tyars, the record does not demonstrate\\nany particular reason why Claiborne had to be visibly\\nrestrained in front of the jury. See Duckett, 67 F.3d at 749;\\nTyars, 709 F.2d at 1284. His status as a convicted felon is\\nnot sufficient. Duckett, 67 F.3d at 749 (noting that murder\\nconviction alone is not sufficient basis for shackling a\\ndefendant at sentencing); cf. Lemons, 985 F.2d at 357\\n(stating that “not all convicted felons are so dangerous and\\nviolent that they must be brought to court and kept in\\n 5\\n “We may also take into consideration the costs of correcting an\\nerror and—in borderline cases—the effect that a verdict may have on\\nnonparties, although these considerations are by no means dispositive as\\nto whether we will exercise our discretion to correct forfeited errors.”\\nHoard v. Hartman, 904 F.3d 780, 787 (9th Cir. 2018) (internal quotation\\nmarks omitted). Here, unlike in Hoard, we are not reviewing civil jury\\ninstructions and this is not a borderline case. Therefore, we do not\\nconsider the costs of correcting the error or the effect a verdict may have\\non nonparties. See id. at 792 n.11.\\n\\f22 CLAIBORNE V. BLAUSER\\n\\nhandcuffs and leg irons”). Claiborne’s criminal record\\nconsisted of nonviolent property and drug offenses.\\nClaiborne, 2015 WL 5146746, at *1. One of the defendant\\nofficers testified that she never had a conflict with Claiborne\\naside from the 2010 incident. It was uncontested at trial that\\nClaiborne was not considered a violent person in jail.\\nFinally, there is nothing in the record to indicate that\\nClaiborne disrupted court proceedings, acted disrespectfully\\nin court, or attempted to escape. See Duckett, 67 F.3d at 749.\\nThus, the district court erred by concluding that it would\\nhave kept Claiborne shackled on the basis that he is a\\nconvicted felon serving a lengthy prison sentence. Such\\nreasoning fails to perform the particularized consideration of\\nnecessity that we described in Duckett. Id. at 748.\\n\\n 2.\\n\\n The second prong of the plain error analysis requires the\\nerror to be plain or obvious. Draper, 836 F.3d at 1085. Plain\\nerror “is error that is so clear-cut, so obvious, a competent\\ndistrict judge should be able to avoid it without benefit of\\nobjection.” United States v. Gonzalez-Aparicio, 663 F.3d\\n419, 428 (9th Cir. 2011). We agree with Claiborne that the\\ndistrict court’s failure to address Claiborne’s shackling, and\\ndenial of a new trial based on the shackling, were plainly\\nerroneous.\\n\\n In 1983, we applied Supreme Court precedent to hold\\nthat there was a due process violation where a civil litigant\\n“was exhibited to the jury in physical restraints when the\\nexpress question it was to decide was whether he was\\ndangerous to himself or others.” Tyars, 709 F.2d at 1284.\\nWe concluded that this posed “an even greater potential for\\njuror bias than trying a criminal defendant in prison garb”\\nbecause the “‘constant reminder of the accused’s condition\\nimplicit in such distinctive, identifiable [restraints]’ surely\\n\\f CLAIBORNE V. BLAUSER 23\\n\\ncould have had a significant, deleterious effect on the jury’s\\njudgment.” Id. at 1284–85 (alteration in original) (quoting\\nEstelle v. Williams, 425 U.S. 501, 504–05 (1976)). Thus, we\\nheld that shackling must be justified by “at least some prior\\nshowing of necessity” where “the critical question the jury\\nmust decide is whether the individual is dangerous to himself\\nor others.” Id. at 1285.\\n\\n Defendants argue that our holding in Tyars is limited to\\ncivil commitment proceedings, and cite to unpublished\\ndispositions characterizing Tyars as such. See Carpenter v.\\nPfeil, 617 Fed. App’x 658, 660 (9th Cir. 2015); Hartman v.\\nMcCarthy, 74 F.3d 1245, at *1 (9th Cir. 1996) (unpublished\\ntable decision). We are unpersuaded by these assertions for\\nthree reasons. 6\\n\\n First, our holding in Tyars hinged on the “express\\nquestion” that the jury was to decide—i.e., the petitioner’s\\ndangerousness—not the type of civil proceeding. 709 F.2d\\nat 1284–85. The commitment proceedings required the jury\\ndetermine whether petitioner was dangerous to himself or\\nothers. Id. That question is also core to Claiborne’s case.\\nSecond, when describing the impact of Tyars in Duckett, we\\nobserved generally that “courts have held that when an\\nindividual’s level of dangerousness is a question the jury\\nmust decide in a civil proceeding, it is a violation of the right\\nto a fair trial to compel that individual to appear before the\\njury bound in physical restraints.” 67 F.3d at 748. Thus, our\\n\\n 6\\n Moreover, we of course are not bound by unpublished dispositions.\\n9th Cir. R. 36-3(a); Pedroza v. BRB, 624 F.3d 926, 931 (9th Cir. 2010)\\n(“[A]n unpublished decision is not precedent for our panel.”).\\nNonetheless, we do not find them persuasive because Carpenter and\\nHartman held in the alternative that even if Tyars and Duckett controlled,\\nthere was no prejudice in those cases so any error was harmless.\\nCarpenter, 617 Fed. App’x at 660; Hartman, 74 F.3d 1245, at *1.\\n\\f24 CLAIBORNE V. BLAUSER\\n\\ndicta in Duckett is consistent with the broad nature of our\\nholding in Tyars. Finally, all other circuits that have\\ndiscussed Tyars in the shackling context treated it as support\\nfor applying in civil cases the Supreme Court’s rationale\\nagainst routine shackling. See Lemons, 985 F.2d at 356–58\\n& n.2; Davidson, 44 F.3d at 1122–23; Sides, 609 F.3d at 581.\\n\\n For these reasons, we conclude that the error was plain.\\n\\n 3.\\n\\n “The third prong of the plain error analysis requires that\\nthe district court’s plain error . . . prejudiced the complaining\\nparty or otherwise affected his or her substantial rights.”\\nHoard, 904 F.3d at 790. “In most cases it means that the\\nerror must have . . . affected the outcome of the district court\\nproceedings. Id. at 791 (quoting United States v. Olano,\\n507 U.S. 725, 734 (1993)). We conclude that Claiborne has\\nalso satisfied this factor.\\n\\n Claiborne’s dangerousness was the key issue at trial. See\\nTyars, 709 F.2d at 1285. The defendant officers justified\\ntheir actions against Claiborne in the 2010 incident on the\\nbasis that he presented a security risk. For instance, Officer\\nBlauser testified that she placed Claiborne in traditional\\nhandcuffs rather than waist chains because she believed he\\nposed an immediate threat. She stated she did not honor\\nClaiborne’s accommodations because “[s]ecurity supersedes\\neverything.” Officer Blauser also explained that she and\\nOfficer Martin escorted Claiborne straight across the yard\\nrather on the paved track area “because of his aggression.”\\nOn cross-examination, when asked why she took him down\\non the yard, Officer Blauser responded it was because\\nClaiborne presented a threat to her and her partner. In total,\\nOfficer Blauser referred to Claiborne as “aggressive” or a\\nthreat at least forty times during her testimony. In closing\\n\\f CLAIBORNE V. BLAUSER 25\\n\\narguments, the defense repeated Blauser’s testimony and\\nemphasized that “she perceived a threat” and “took the most\\nreasonable action under the circumstances.” In short, if the\\njury believed that Claiborne posed a threat to the officers,\\nthen defendants prevailed because their use of force was\\njustified.\\n\\n Not only did Claiborne’s case hinge on his alleged\\ndangerousness, the trial pitted his credibility against the\\ndefendant officers’ credibility, and raised the issue of flight\\nrisk as defendants claimed that he was trying to escape from\\nthem during their escort in 2010. Claiborne’s visible\\nshackling was a “constant reminder” to the jury of his\\ncondition as a potentially violent and unreliable person who\\nneeded to be restrained. Tyars, 709 F.2d at 1284 (quoting\\nEstelle, 425 U.S. at 504); see also Lemons, 985 F.2d at 357.\\nWe have also recognized that physical restraints “may\\nconfuse and embarrass the [individual], thereby impairing\\nhis mental faculties[,] and they may cause him pain.”\\nDuckett, 67 F.3d at 748 (quoting Spain v. Rushen, 883 F.2d\\n712, 720–21 (9th Cir. 1989) (collecting cases) (internal\\nquotation marks omitted)). No doubt, shackles likely\\nimpaired Claiborne as a pro se litigant in the presentation of\\nhis case to a jury over three days. Id. Although the district\\ncourt noted, ten months after the trial, that Claiborne’s hands\\nwere not restrained, this is contradicted by Officer Blauser’s\\ncontemporaneous observation at trial that Claiborne was\\n“wearing [waist chains] right now” and that these waist\\nchains were “big and bulky.”\\n\\n Defendants argue there was no prejudice because the\\njury already knew Claiborne was a convicted felon. But the\\nSupreme Court and other circuit courts have uniformly\\nrejected the idea that a conviction alone could justify the use\\nof shackles because restraint “almost inevitably affects\\n\\f26 CLAIBORNE V. BLAUSER\\n\\nadversely the jury’s perception of the character of the\\ndefendant.” Deck, 544 U.S. at 633; see also Duckett, 67 F.3d\\nat 749 (noting that a defendant’s status as a convicted felon,\\n“[s]tanding alone, . . . is not a sufficient reason to impose\\nphysical restraints”); Lemons, 985 F.2d at 357. We are also\\nnot convinced by defendants’ insistence that the evidence\\npresented at trial overwhelmingly supported the jury’s\\nverdict. Crucially, the district court denied defendants’\\nmotion for judgment as a matter of law at the close of\\nevidence, demonstrating that a reasonable jury had a legally\\nsufficient evidentiary basis to find in Claiborne’s favor. See\\nFed. R. Civ. P. 50(a).\\n\\n Because of the inherent nature of visible shackles and its\\ninterplay with the heart of Claiborne’s excessive force claim,\\nwe conclude that Claiborne was prejudiced by the erroneous\\nuse of shackling in the absence of any showing of a\\ncompelling need for such restraints.\\n\\n 4.\\n\\n The final prong requires showing that “the district\\ncourt’s errors are so grave as to seriously impair the fairness,\\nintegrity, or public reputation of judicial proceedings.”\\nHoard, 904 F.3d at 791 (internal quotation marks and\\nalterations omitted). This is “undoubtedly the hardest\\n[prong] to meet.” Id. Because of the fundamental nature of\\nthe error, however, we find that this last prong has also been\\nmet.\\n\\n Shackling “must be limited to cases urgently demanding\\nthat action,” and because there was no showing of necessity\\nin Claiborne’s case, he was unduly prejudiced in violation of\\nhis due process right to a fair trial. Tyars, 709 F.2d at 1284;\\nsee also Maus v. Baker, 747 F.3d 926, 927 (7th Cir. 2014)\\n(“The sight of a shackled litigant is apt to make jurors think\\n\\f CLAIBORNE V. BLAUSER 27\\n\\nthey’re dealing with a mad dog . . .”). Just as an erroneous\\njury instruction could “place[] a heavy thumb on the scale in\\nfavor of the [d]efendants,” the erroneous use of shackling\\nhere also constituted a “grave injustice” in depriving\\nClaiborne of “a meaningful and fair opportunity to seek\\nredress for alleged violations of his constitutional right to be\\nfree from cruel and unusual punishment.” Hoard, 904 F.3d\\nat 792.\\n\\n We also reiterate that shackling not only prejudices the\\ninmate litigant in a case like this, it presents an affront to the\\ndignity of the courtroom. Deck, 544 U.S. at 631 (citing\\nAllen, 397 U.S. at 344). The Supreme Court has stressed that\\n“[t]he routine use of shackles in the presence of juries would\\nundermine [the] symbolic yet concrete objective[]” of\\n“maintain[ing] a judicial process that is a dignified process.”\\nId. Hence, shackling may only be used when there is an\\n“individualized security determination[]” that “take[s]\\naccount of the circumstances of the particular case.” Id.\\nat 632. That individualized determination did not take place\\nhere and the district court’s blanket assertion that it would\\nhave ordered Claiborne shackled solely based on his status\\nas a convicted felon was plainly erroneous. We therefore\\nexercise our discretion on plain error review to reverse the\\ndistrict court’s denial of the Rule 59(a) motion and remand\\nfor a new trial on Claiborne’s claims against Officers\\nBlauser and Martin.\\n\\n IV.\\n\\n Claiborne was denied a fair trial when he was visibly\\nshackled before a jury that had to decide his dangerousness\\nas it determined whether the jury believed his or defendants’\\nversion of the events that underlie his Section 1983 claims.\\nThere was no individualized determination of the security\\nneed for such restraints, yet this mobility impaired plaintiff\\n\\f28 CLAIBORNE V. BLAUSER\\n\\nwas presented to the jury in shackles, corroborating the\\ndefense’s position that Claiborne was an insubordinate,\\naggressive inmate and thereby undermining his case. On\\nremand, the district court has discretion to impose shackling\\nduring the new trial, but it may only do so after a full hearing\\nat which officers show a compelling need for security and\\nthe court considers any less restrictive alternatives. 7 See\\n\\n\\n\\n 7\\n Because we remand for a new trial on the shackling claim, we do\\nnot address the merits of Claiborne’s evidentiary arguments that the\\ndistrict court erred in barring his testimony about the ADA, preventing\\nhim from introducing or testifying about his medical records, and\\ndenying his request for a medical expert. The district court may revisit\\nthese issues if Claiborne raises them again at the new trial.\\n\\n We note, however, that the district court appears to misstate the law\\nwhen it denied Claiborne’s request for a medical expert under Federal\\nRule of Evidence 706(a). Although the district court correctly\\nrecognized that Rule 706(a) provides discretion to appoint a neutral\\nexpert witness, see McKinney v. Anderson, 924 F.2d 1500, 1511 (9th Cir.\\n1991), vacated on other grounds sub nom. Helling v. McKinney, 502\\nU.S. 903 (1991), judgment reinstated, 959 F.2d 853 (9th Cir. 1992),\\naff’d, 509 U.S. 25 (1993), the district court seemed to categorically limit\\nthe relevance of a medical expert to testifying about a plaintiff’s current\\ncondition. Yet courts have regularly considered requests for and\\nappointed experts to review medical records and testify about prior\\nmedical needs and treatment in deliberate indifference cases. See Gorton\\nv. Todd, 793 F.Supp.2d 1171, 1179–81 (E.D. Cal. 2011) (collecting\\ncases). Moreover, a medical expert can help with factfinding in\\nexcessive force claims because “the extent of injury suffered by an\\ninmate is one factor that may suggest ‘whether the [defendant’s] use of\\nforce could plausibly have been thought necessary’ in a particular\\nsituation.” Hudson v. McMillian, 503 U.S. 1, 7 (1992). If Claiborne\\nrenews his request for appointment of a neutral medical expert on retrial,\\nthe district court should weigh these considerations in exercising her\\ndiscretion. See McKinney, 924 F.2d at 1511; see also Gorton,\\n793 F.Supp.2d at 1185–86.\\n\\f CLAIBORNE V. BLAUSER 29\\n\\nDuckett, 67 F.3d at 748; Tyars, 709 F.2d at 1285; see also\\nDavidson, 44 F.3d at 1126.\\n\\n For the foregoing reasons, we reverse the district court’s\\ndenial of Claiborne’s Rule 59(a) motion and remand for a\\nnew trial.\\n\\n REVERSED AND REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4425497/\", \"author_raw\": \"PAEZ, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"SIDNEY R THOMAS"},"judge2":{"kind":"string","value":"RICHARD A PAEZ"},"judge3":{"kind":"string","value":"GARY FEINERMAN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4648244/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Dennis Claiborne v. Blauser\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-14\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n DENNIS GERALD CLAIBORNE, No. 16-16077\n Plaintiff-Appellant,\n D.C. No.\n v. 2:10-cv-02427-VAP\n\n BLAUSER, Correctional Officer;\n S. MARTIN, Correctional ORDER AND\n Officer, AMENDED\n Defendants-Appellees. OPINION\n\n Appeal from the United States District Court\n for the Eastern District of California\n Virginia A. Phillips, Chief District Judge, Presiding\n\n Argued and Submitted February 6, 2019\n San Francisco, California\n\n Filed June 28, 2019\n Amended August 14, 2019\n\nBefore: Sidney R. Thomas, Chief Judge, Richard A. Paez,\n Circuit Judge, and Gary Feinerman, * District Judge.\n\n\n\n\n *\n The Honorable Gary Feinerman, United States District Judge for\nthe Northern District of Illinois, sitting by designation.\n\f2 CLAIBORNE V. BLAUSER\n\n Order;\n Opinion by Judge Paez\n\n\n SUMMARY **\n\n\n Prisoner Civil Rights\n\n The panel reversed the district court’s denial of a motion\nfor a new trial and remanded in an action brought pursuant\nto 42 U.S.C. § 1983 by a convicted state inmate who alleged\nthat he was shackled without justification during his three-\nday trial on his Eighth Amendment excessive force and\ndeliberate indifference to medical needs claims.\n\n The panel first noted that although the inmate did not\nobject to the shackling during trial, he raised the issue in\nsupport of his motion for a new trial. The panel applied plain\nerror review.\n\n The panel held that because the inmate’s dangerousness\nand flight risk were central issues at the trial, the district\ncourt plainly erred in allowing him to be visibly shackled\nwithout any showing of a sufficient need for such restraints.\nThe panel held that on remand, the district court would have\ndiscretion to impose shackling during the new trial, but it\ncould only do so after a full hearing at which officers showed\na compelling need for security and the court considered any\nless restrictive alternatives.\n\n\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f CLAIBORNE V. BLAUSER 3\n\n COUNSEL\n\nJeremy M. Christiansen (argued), Gibson Dunn & Crutcher\nLLP, Washington, D.C.; J. Brett Bylund and Blaine H.\nEvanson, Gibson Dunn & Crutcher LLP, Irvine, California;\nfor Plaintiff-Appellant.\n\nJaime Ganson (argued) and Arthur B. Mark III, Deputy\nAttorneys General; Neah Huynh, Acting Supervising\nDeputy Attorney General; Monica N. Anderson, Senior\nAssistant Attorney General; Xavier Becerra, Attorney\nGeneral; Office of the Attorney General, Sacramento,\nCalifornia; for Defendants-Appellees.\n\n\n ORDER\n\n The opinion filed in this case on June 28, 2019 is\namended by replacing the text of footnote 7 on page 28,\nClaiborne v. Blauser, 928 F.3d 794, 810 n.7 (9th Cir. 2019),\nwith the following paragraphs:\n\n Because we remand for a new trial on the\n shackling claim, we do not address the merits\n of Claiborne’s evidentiary arguments that the\n district court erred in barring his testimony\n about the ADA, preventing him from\n introducing or testifying about his medical\n records, and denying his request for a\n medical expert. The district court may revisit\n these issues if Claiborne raises them again at\n the new trial.\n\n We note, however, that the district court\n appears to misstate the law when it denied\n\f4 CLAIBORNE V. BLAUSER\n\n Claiborne’s request for a medical expert\n under Federal Rule of Evidence 706(a).\n Although the district court correctly\n recognized that Rule 706(a) provides\n discretion to appoint a neutral expert witness,\n see McKinney v. Anderson, 924 F.2d 1500,\n 1511 (9th Cir. 1991), vacated on other\n grounds sub nom. Helling v. McKinney,\n 502 U.S. 903 (1991), judgment reinstated,\n 959 F.2d 853 (9th Cir. 1992), aff’d, 509 U.S.\n 25 (1993), the district court seemed to\n categorically limit the relevance of a medical\n expert to testifying about a plaintiff’s current\n condition. Yet courts have regularly\n considered requests for and appointed experts\n to review medical records and testify about\n prior medical needs and treatment in\n deliberate indifference cases. See Gorton v.\n Todd, 793 F.Supp.2d 1171, 1179–81 (E.D.\n Cal. 2011) (collecting cases). Moreover, a\n medical expert can help with factfinding in\n excessive force claims because “the extent of\n injury suffered by an inmate is one factor that\n may suggest ‘whether the [defendant’s] use\n of force could plausibly have been thought\n necessary’ in a particular situation.” Hudson\n v. McMillian, 503 U.S. 1, 7 (1992). If\n Claiborne renews his request for appointment\n of a neutral medical expert on retrial, the\n district court should weigh these\n considerations in exercising her discretion.\n See McKinney, 924 F.2d at 1511; see also\n Gorton, 793 F.Supp.2d at 1185–86.\n\nThe Amended Opinion is filed concurrently with this order.\n\f CLAIBORNE V. BLAUSER 5\n\n With the filing of the Amended Opinion, the panel has\nunanimously voted to deny the Defendants-Appellees’\npetition for panel rehearing. No further petitions for\nrehearing may be filed.\n\n\n\n OPINION\n\nPAEZ, Circuit Judge:\n\n The law has long forbidden the routine use of visible\nshackling during a criminal defendant’s trial. Deck v.\nMissouri, 544 U.S. 622, 626 (2005). Visible shackling\nundermines the presumption of innocence, impedes the\njury’s factfinding process, hampers presentation of a\ndefense, and affronts the dignity and decorum of judicial\nproceedings. Id. at 630–32. In this civil rights case under\n42 U.S.C. § 1983, we consider whether the unjustified\nshackling of a convicted state inmate during his three-day\ntrial on Eighth Amendment excessive force and deliberate\nindifference claims deprived him of a fair trial in violation\nof the federal constitution. 1 Although the inmate did not\nobject to the shackling during trial, he raised the issue in\nsupport of his motion for a new trial, which the district court\ndenied.\n\n We hold that the district court abused its discretion in\ndenying a new trial. Because the inmate’s dangerousness\nand flight risk were central issues at the trial, the district\ncourt plainly erred in allowing him to be visibly shackled\nwithout any showing of a sufficient need for such restraints.\n\n\n 1\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\n\f6 CLAIBORNE V. BLAUSER\n\nSee Tyars v. Finner, 709 F.2d 1274, 1284–85 (9th Cir. 1983).\nWe therefore reverse and remand for a new trial.\n\n I.\n\n This appeal arises out of a lawsuit filed by Dennis Gerald\nClaiborne who, proceeding pro se, sued Correctional\nOfficers Jemini Blauser, Greg Martin, and other individual\nofficials under Section 1983 for the use of excessive force\nand deliberate indifference to his medical needs.\n\n A.\n\n Claiborne is a 63-year-old California state prison inmate\nin the custody of the California Department of Corrections\nand Rehabilitation (“CDCR”). He is serving a 60-years-to-\nlife sentence under California’s Three Strikes Law for\nattempted burglary and receipt of stolen property. People v.\nClaiborne, No. B260391, 2015 WL 5146746, at *1 (Cal. Ct.\nApp. Sept. 2, 2015).\n\n Claiborne is mobility impaired due to a right knee\nreplacement in 2007 and ensuing chronic problems with that\nknee. Given his condition, Claiborne participates in the\nCDCR Disability Placement Program and receives certain\naccommodations in prison: he is allowed to use a cane; he is\nrestricted to housing on the lower level, with no stairs; and,\nwhen escorting Claiborne within the institution, correctional\nstaff must use “waist chains” and choose “relatively level\nterrain and no obstructions in the path of travel.” Waist\nchains are different from traditional handcuffs; whereas the\nlatter are typically applied behind the back, the former allow\na mobility impaired inmate to keep his hands at his sides\nduring an escort, which lets him use any prescribed\naccommodation devices like a cane. CDCR documents and\nconveys information about Claiborne’s accommodations to\n\f CLAIBORNE V. BLAUSER 7\n\nprison officials through physician orders known as medical\naccommodation “chronos” in his file. Claiborne also wears\na green vest to alert officers that he is mobility impaired.\n\n The incident between Claiborne and Officers Blauser\nand Martin took place on May 3, 2010, while Claiborne was\nhoused at California’s High Desert State Prison, a Level\nFour security prison. As Claiborne was waiting for his\nmedication in the morning “pill line,” Correctional Officer\nDaniel McBride, stationed in an observation tower, believed\nhe saw Claiborne socializing with other inmates in the line.\nOfficer McBride called Officer Blauser, who was working\nin Claiborne’s housing unit along with her partner, Officer\nMartin, and asked Blauser to counsel Claiborne for\nunnecessarily lingering in the pill line. After waiting thirty\nminutes in the pill line, Claiborne received his medication\nand returned to his housing unit where he was admonished\nby Officer Blauser. She decided to “put a cap” on\nClaiborne’s door, meaning he would spend the rest of the\nday locked up in his cell. Because Claiborne had intended\nto present at a Bible study group later that day, he asked to\nspeak with the sergeant, Officer Blauser’s supervisor, to\ncontest the punishment and explain that he had been properly\nwaiting in line and not socializing.\n\n The account of the facts diverge drastically from there.\nAccording to Claiborne, he was acting respectfully toward\nOfficer Blauser when she told him to “cuff up.” He\ncomplied and, as Officers Blauser and Martin started to\nescort him, Claiborne informed them that it was difficult for\nhim to walk and use his cane with his hands cuffed behind\nhis back. He mentioned his chrono for waist chains but\nOfficer Blauser told Officer Martin to take Claiborne’s cane.\nOfficer Martin said he would help hold Claiborne up as the\nthree of them walked to the program office to speak with the\n\f8 CLAIBORNE V. BLAUSER\n\nsergeant. Claiborne, however, had difficulty walking with\nOfficers Blauser and Martin. They escorted him straight\nacross the yard, rather than along the sidewalk circling the\nyard, despite Claiborne’s chrono providing for level terrain.\nBecause the yard was uneven, Claiborne hyperextended his\nright leg, causing his knee to give out partway across the\nyard. Claiborne lost balance and shifted rightward, causing\nOfficer Blauser to order him to stop resisting. Claiborne\ntried to explain that he was not resisting and that his knee\nwas bothering him because the officers were pulling him too\nquickly.\n\n When the three were almost at the program office,\nClaiborne’s right leg hyperextended again as he tried to pick\nthe leg up three to five inches from the dirt yard onto the\npavement. Because he had no cane to catch himself,\nClaiborne leaned to his right again, causing Officer Blauser\nto shout “he’s resisting” and pull him down to the ground.\nShe jumped on his right side, including his replaced knee,\nand pulled his hair and hit him in the face a few times. Other\nofficers quickly rushed to the scene and Claiborne heard\nindividuals ask, “Where’s his cane?” and “Why isn’t he in\nwaist chains?” Claiborne was eventually taken into a\nholding cell and then interviewed by a sergeant, which was\nrecorded by a video camera. Before turning on the camera\nto record the interview, the sergeant warned Claiborne that\nif he reported excessive force, he would be taken to the\n“hole,” in other words, administrative segregation, for an\nunknown amount of time. Worried about whether his\nmedical needs would be met in the hole, Claiborne\neventually stated on camera, after extended back-and-forth\nwith the sergeant, that there was no excessive force used\nagainst him.\n\f CLAIBORNE V. BLAUSER 9\n\n Officers Blauser and Martin presented a different\naccount. They were both aware of Claiborne’s mobility\nimpairment at the time of the incident. According to Officer\nBlauser, while she counseled Claiborne for lingering in the\npill line, he became “really aggravated” and started raising\nhis voice at her, causing her to feel uncomfortable.\nClaiborne then walked toward her while holding, not using,\nhis cane. Because she did not feel safe, Officer Blauser told\nClaiborne to “cuff up,” and he immediately turned around\nand complied. She asked her partner, Officer Martin, to\nassist her with escorting Claiborne to the program office.\nOfficer Martin took Claiborne’s cane, and they each\nsupported him by holding onto his bicep or arm on each side.\nNeither Officer Blauser nor Officer Martin recalled\nClaiborne saying anything about needing to use waist chains.\nMoreover, because it would have taken a few extra minutes\nto obtain waist chains, Officer Blauser decided to use\nhandcuffs due to Claiborne’s aggression and defiance of her\norder to return to his cell.\n\n According to Officers Blauser and Martin, they escorted\nClaiborne straight across the yard because it was the quickest\nand most direct path to the program office and avoided\nwalking amongst other inmates. They escorted him slowly\nand did not perceive any problems with his walking. Rather,\nthey believed that Claiborne tried to break away from them\ntwice, once halfway across the yard, and a second time close\nto the program office. At first, Officer Blauser ordered\nClaiborne not to pull his arm away but he continued to act\naggressively, yelling and trying to pull away from her.\nWhen Claiborne pulled his arm away from her a second\ntime, Officer Blauser decided to pull him down and called a\n“code one” over the radio. Officers Blauser and Martin used\ntheir weight to hold Claiborne down on the ground. They\ndenied that Officer Blauser jumped on Claiborne, pulled his\n\f10 CLAIBORNE V. BLAUSER\n\nhair or punched him in the face. After other officers escorted\nClaiborne away, Sergeant Officer Kenneth Gullion followed\nprotocol and conducted a video-recorded “use-of-force”\ninterview, asking Claiborne about what took place. Sergeant\nGullion did not remember what happened prior to\ninterviewing Claiborne on video. A separate officer wrote\nClaiborne up for a rule violation, resisting a peace officer.\n\n Shortly after the incident, a nurse examined Claiborne\nand completed a medical report, noting that he had two\nabrasions, one each on his left knee and left cheek.\nAfterward, Claiborne experienced more problems with his\nright knee, and doctors determined that he had significant\ninjuries, including bursitis. Following a series of\nevaluations, Claiborne was assessed to have “[f]ailed right\ntotal knee arthroplasty.” He underwent a revision procedure\nin 2012, but was told that his knee did not respond properly\nto the surgery and could not be fixed any more. Claiborne\nalso had surgery on one of his shoulders in July 2015, when\nhe was told that he waited too long to fix it.\n\n B.\n\n After exhausting the prison’s administrative process,\nClaiborne filed suit in district court. Proceeding pro se, he\nsued Officers Blauser, Martin and other individual officers\nunder Section 1983 for various claims, including the use of\nexcessive force and deliberate indifference to his medical\nneeds in violation of his Eighth Amendment rights. The\nother named defendant officers and other claims, including\nbattery, negligent infliction of emotional distress and\nintentional infliction of emotional distress, were eventually\ndismissed from the case. After the district court denied\ncross-motions for summary judgment, the case proceeded to\ntrial on the two Eighth Amendment claims against Officers\nBlauser and Martin.\n\f CLAIBORNE V. BLAUSER 11\n\n The trial lasted three days, at the start of which the\ndistrict court noted, outside of the jury’s presence, that\nClaiborne was shackled. Claiborne testified on his own\nbehalf, and two fellow inmates also testified on his behalf.\nThe defense presented testimony from six officers including\nOfficers Blauser and Martin, the tower guard officer\n(McBride), and Sergeant Gullion. The jury reached a verdict\nfor both defendants on both claims.\n\n Claiborne filed a timely motion for new trial and relief\nfrom judgment, raising various arguments. Most relevant to\nour analysis, Claiborne argued that he was entitled to a new\ntrial under Federal Rule of Civil Procedure 59(a) because he\nwas visibly shackled while litigating his case in front of the\njury. He also argued for relief from judgment under Federal\nRule of Civil Procedure 60(b) because the district court\nbarred his testimony about the Americans with Disabilities\nAct (“ADA”), prevented him from introducing or testifying\nabout his medical records, and denied his request for a\nmedical expert. The district court dismissed all the\narguments and denied the motion.\n\n II.\n\n When reviewing denial of a motion for new trial under\nFederal Rule of Civil Procedure 59(a), our default is to\nreview for abuse of discretion. See Hung Lam v. City of San\nJose, 869 F.3d 1077, 1084 (9th Cir. 2017). But here,\nClaiborne did not object to his shackling at trial so we must\naddress the appropriate standard of review.\n\n Errors not objected to at trial are generally subject to\nwaiver or forfeiture. “Forfeiture is the failure to make a\ntimely assertion of a right, whereas waiver is the intentional\nrelinquishment or abandonment of a known right.” United\nStates v. Perez, 116 F.3d 840, 845 (9th Cir. 1997) (en banc)\n\f12 CLAIBORNE V. BLAUSER\n\n(internal quotation marks omitted); see also Crowley v.\nEpiCept Corp., 883 F.3d 739, 748 (9th Cir. 2018) (per\ncuriam) (applying same standard in civil context).\n“Forfeited rights are reviewable for plain error, while waived\nrights are not.” Id.\n\n Under plain error review, we may reverse only where:\n(1) there was an error; (2) the error was obvious; (3) the error\naffected substantial rights; and, (4) the error seriously affects\nthe fairness, integrity, or public reputation of judicial\nproceedings. See C.B. v. City of Sonora, 769 F.3d 1005,\n1018–19 (9th Cir. 2014) (en banc). In other words, plain\nerror requires reversal where “review is necessary to prevent\na miscarriage of justice.” Draper v. Rosario, 836 F.3d 1072,\n1085 (9th Cir. 2016) (internal quotation marks omitted).\n\n Defendants argue that Claiborne forfeited the error\nbecause he failed to object to his shackling at trial, and so,\nhis claim should be subject to plain error review. 2 Claiborne\ncontends that plain error is inapplicable because he timely\nraised the shackling issue in his motion for a new trial, and\nthe district court addressed the claim on the merits, holding\nthat it would have imposed shackling over any objection\n\n 2\n We find no merit to defendants’ alternative argument that\nClaiborne invited the error. We impose a high standard for applying the\ninvited error doctrine. See United States v. Lindsey, 634 F.3d 541, 555\n(9th Cir. 2011) (noting that “in order for the invited error doctrine to\napply, a [party] must both invite the error and relinquish a known right”\n(citing Perez, 116 F.3d at 845)). Claiborne’s mention of his shackles in\nhis opening statement, read in context, suggests that he was trying to\nintroduce himself to the jury and not hide the fact that he is incarcerated.\nThere is no indication in the record that Claiborne, who was pro se in the\ndistrict court, “both invited the error, and relinquished a known right” to\na fair trial without shackling. Perez, 116 F.3d at 845.\n\f CLAIBORNE V. BLAUSER 13\n\nbecause Claiborne is a convicted felon serving a lengthy\nprison sentence.\n\n The district court’s denial of a motion for a new trial is\nreversible if the court made a mistake of law. Molski v. M.J.\nCable, Inc., 481 F.3d 724, 729 (9th Cir. 2007). As we\nelaborate below, we agree that the district court disregarded\nthe law on shackling. However, in prior cases involving\nlegal claims that were not raised during trial, we still\nundertook plain error review when reviewing a denial of a\nmotion for a new trial. 3 See, e.g., Settlegoode v. Portland\nPub. Sch., 371 F.3d 503, 517 (9th Cir. 2004) (“There is an\neven ‘high[er] threshold’ for granting a new trial where, as\nhere, [the moving party] failed to object to the alleged\nmisconduct during trial.” (first alteration in original)\n(quoting Kaiser Steel Corp. v. Frank Coluccio Constr. Co.,\n785 F.2d 656, 658 (9th Cir. 1986))); Hemmings v. Tidyman’s\nInc., 285 F.3d 1174, 1193 (9th Cir. 2002) (“We will review\nfor plain or fundamental error, absent a contemporaneous\nobjection . . .”).\n\n The policy behind plain error review is to ensure that the\nparties and district court are on notice as to any alleged errors\nso that the court may address the objection, correct any error,\nand create a proper record for appeal. See Hemmings,\n285 F.3d at 1193 (observing that “the trial judge is in a\nsuperior position to evaluate the likely effect of the alleged\nmisconduct and to fashion an appropriate remedy”); see also\n 3\n This is analogous to our review of instructional error claims when\nthere was no objection at trial. 11 Charles A. Wright et al., Federal\nPractice and Procedure § 2805 (3d ed. 2012). Under Federal Rule of\nCivil Procedure 51(d)(2), we review a challenge to civil jury instructions\nfor plain error in the absence of a timely objection. C.B., 769 F.3d\nat 1017–19.\n\f14 CLAIBORNE V. BLAUSER\n\nUnited States v. $11,500.00 in U.S. Currency, 869 F.3d\n1062, 1075 (9th Cir. 2017).\n\n Here, the district court’s holding on the merits—that it\nwould have imposed shackling over any timely objection—\ndoes give us some pause as to the standard of review. 4 We\nhave recognized one limited exception to plain error review\nin the civil context, the “pointless formality” exception.\nChess v. Dovey, 790 F.3d 961, 970–72 (9th Cir. 2015); see\nalso Shorter v. Baca, 895 F.3d 1176, 1182–83 (9th Cir.\n2018). In Chess, we declined to apply plain error where the\nplaintiff was pro se and “the district court and opposing party\nwere fully aware of the potential problem with, and would-\nbe objection to, [a jury] instruction,” even though the\nplaintiff did not raise the issue himself. 790 F.3d at 964.\nBecause the district court and defendants were on notice, we\nconcluded that “any objection by Chess would have been\n‘superfluous and futile,’” id. at 972 (quoting Obsidian Fin.\nGrp. v. Cox, 740 F.3d 1284, 1289 (9th Cir. 2014)), and we\ndeclined to “punish a pro se litigant with plain error rather\nthan de novo review simply because he failed to say the\nwords ‘I object,’” id. at 971.\n\n Claiborne, too, was pro se and the district court’s post-\ntrial ruling suggests that any timely objection would have\n\n 4\n The district court’s ruling distinguishes Claiborne’s situation from\nthat in Hemmings, in which the defendant moved for a new trial based\non inappropriate comments made by the plaintiffs’ counsel in closing\nargument. 285 F.3d at 1192. The district court noted that it remembered\nthe comment and “would have sustained an objection had one been made\nbecause it was totally improper.” Id. at 1192–93. Because a timely\nobjection would have made a difference and changed the district court’s\nactions, application of plain error review was appropriate under the\nrationale of giving the trial judge an opportunity to cure any prejudice.\nId. at 1193. That concern does not apply here.\n\f CLAIBORNE V. BLAUSER 15\n\nbeen futile. We recognize, however, the policy concerns\nbehind the plain error doctrine and therefore decline to\nreview for abuse of discretion. See Hemmings, 285 F.3d at\n1193 (noting that “allowing a party to wait to raise the error\nuntil after the negative verdict encourages that party to sit\nsilent in the face of claimed error”). Accordingly, we review\nfor plain error.\n\n III.\n\n The district court must grant a motion for a new trial\nwhere “the verdict is against the weight of the evidence,”\n“the damages are excessive” or, “for other reasons, the trial\nwas not fair to the [moving] party.” Montgomery Ward &\nCo. v. Duncan, 311 U.S. 243, 251 (1940). Here, we consider\nan argument based on the latter. Claiborne, now represented\nby pro bono counsel, argues that he was deprived of a fair\ntrial because he was visibly shackled in front of the jury\nthroughout the three-day trial, and that the district court erred\nby denying his Rule 59(a) motion for a new trial. Even under\nthe more demanding plain error standard, we agree that the\ndistrict court erred in denying Claiborne’s motion for a new\ntrial.\n\n A.\n\n To understand why the district court plainly erred, we\nfirst review the case law on visible shackling.\n\n It is well established that due process “forbids the use of\nvisible shackl[ing] . . . during the guilt phase [of a criminal\ntrial], unless that use is ‘justified by an essential state\ninterest.’” Deck, 544 U.S. at 624 (citing Holbrook v. Flynn,\n475 U.S. 560, 568–69 (1986); Illinois v. Allen, 397 U.S. 337,\n343–44 (1970)). The Supreme Court has identified “three\nfundamental legal principles” which are “undermined” or\n\f16 CLAIBORNE V. BLAUSER\n\n“diminished” by the use of visible restraints. Id. at 630–31.\nFirst, the accused enjoys a presumption of innocence, which\nvisible shackling undermines by suggesting that the\ndefendant need be restrained. Id. at 630. This in turn\n“undermines . . . the related fairness of the factfinding\nprocess.” Id. Second, the accused has a right to a\nmeaningful defense from counsel. Id. at 631. Restraints\nmight physically impair communication between a\ndefendant and his counsel, or impair the defendant’s\nparticipation in his own defense. Id. “Third, judges must\nseek to maintain a judicial process that is a dignified [one].”\nId. Because “[t]he courtroom’s formal dignity . . . includes\nthe respectful treatment of defendants,” “the use of shackles\nat trial ‘affront[s]’ the ‘dignity and decorum of judicial\nproceedings that the judge is seeking to uphold.’” Id.\n(quoting Allen, 397 U.S. at 344).\n\n In Deck, the Court held that these considerations apply\nwith equal force to penalty proceedings in capital cases, even\nthough “the presumption of innocence no longer applies.”\nId. at 632. Hence, before imposing restraints on criminal\ndefendants, the trial court must take into account the\nparticular circumstances of each case, including “special\nsecurity needs or escape risks[] related to the defendant,” id.\nat 633. Where the court orders visible shackling without\nadequate justification, “the defendant need not demonstrate\nactual prejudice to make out a due process violation,” and\nthe burden is on the government to “prove beyond a\nreasonable doubt that the shackling error complained of did\nnot contribute to the verdict obtained.” Id. at 635 (alterations\nand internal quotation marks omitted).\n\n Prior to Deck, we have long recognized that the\nprohibition against routine visible shackling applies even\nwhen the presumption of innocence does not, including in\n\f CLAIBORNE V. BLAUSER 17\n\nthe civil context. In Duckett v. Godinez, we anticipated Deck\nby holding that “shackling a defendant during a [capital]\nsentencing hearing before a jury is an inherently prejudicial\npractice which comports with due process only when used\nas a last resort to protect an essential state interest[,] such as\nmaintaining public safety or assuring the decorum of the\nproceedings.” 67 F.3d 734, 747 (9th Cir. 1995). We have\nthus long required trial courts to engage in a two-step process\nbefore shackling a criminal defendant: “[f]irst, the court\nmust be persuaded by compelling circumstances” that the\nmeasure is necessary to “maintain security[;]” and\n“[s]econd, the court must ‘pursue less restrictive alternatives\nbefore imposing physical restraints.” Id. at 748 (internal\nquotation marks omitted). We held that it was error to permit\nshackling solely on the basis that the defendant was a\nconvicted felon, and so remanded for the district court to\ndetermine whether this constitutional error was harmless. Id.\nat 749–50.\n\n Notably, we reached our holding in Duckett “by analogy\nto the treatment of the shackling issue in civil cases.” Id.\nat 748 (citing Tyars, 709 F.2d at 1284–85; Lemons v.\nSkidmore, 985 F.2d 354, 356–58 (7th Cir. 1993); and\nHolloway v. Alexander, 957 F.2d 529, 530 (8th Cir. 1992)).\nAs recounted below, this court has recognized the inherent\nprejudice associated with unjustified shackling in civil\nproceedings as early as 1983, and other circuits have relied\non our reasoning to expand the prohibition on routine\nshackling to other civil contexts.\n\n In Tyars, the petitioner challenged the use of restraints to\nbind him in the presence of the jury during his involuntary\ncivil commitment proceedings. 709 F.2d at 1276, 1284. We\nobserved that “[i]t requires no great extension of Supreme\nCourt precedent to conclude that this may have violated his\n\f18 CLAIBORNE V. BLAUSER\n\nrights under the due process clause.” Id. at 1284. We\nreasoned:\n\n Although the criminal case precedents do not\n necessarily apply in a civil proceeding, we\n find them persuasive. The likelihood of\n prejudice inherent in exhibiting the subject of\n a civil commitment hearing to the jury while\n bound in physical restraints, when the critical\n question the jury must decide is whether the\n individual is dangerous to himself or others,\n is simply too great to be countenanced\n without at least some prior showing of\n necessity. In the absence of any such\n demonstrable or articulable necessity, and in\n the absence of any showing that less\n restrictive means not embodying the same\n potential for prejudice could have maintained\n order in the courtroom, the circumstances\n deprived the proceeding of the appearance of\n evenhanded justice which is at the core of due\n process.\n\nId. at 1285 (emphasis added) (alterations, internal quotation\nmarks and citation omitted). We instructed the district court\non remand to “determine whether the State has any\njustification . . . to support the necessity of physically\nrestraining Tyars in the presence of the jury.” Id. at 1285–\n86.\n\n Since Tyars, at least four other circuits have adopted a\nsimilar test when evaluating the use of shackling in civil\ntrials. In Lemons, the Seventh Circuit held that it was\nimpermissible to rely solely on the opinion of a state\ncorrections officer to shackle a plaintiff prison inmate in a\n\f CLAIBORNE V. BLAUSER 19\n\nSection 1983 action alleging excessive force by corrections\nofficers. 985 F.2d at 356. The plaintiff claimed that\ndefendant officers attacked and beat him severely in his cell,\nwhile defendants asserted that the plaintiff was “ranting and\nraving,” and had to be subdued. Id. at 355. The Lemons\ncourt cited with approval Tyars, id. at 356 n.2, and concluded\nthat, as in the criminal context, the appearance of a civil\nplaintiff in handcuffs and leg irons “suggested to the jury that\nthe plaintiff was dangerous and violent, so that whatever\nforce the guards had used was probably necessary, and not\nexcessive,” id. at 359. “[S]hackles inevitably prejudiced the\njury” because “plaintiff’s tendency towards violence was at\nissue.” Id. at 357. The court therefore remanded for a new\ntrial and required a hearing to determine what, if any,\nrestraints were necessary. Id. at 359.\n\n The Eighth Circuit in Holloway also recognized that, as\na general rule, inmate civil plaintiffs should not have to\nappear in court in shackles unless there was a showing of\nneed and steps to mitigate any potential prejudice. 957 F.2d\nat 530. The court found no prejudice in Holloway because\nthe plaintiff challenged the constitutionality of living\nconditions in the state prison, which did not raise the\nplaintiff’s dangerousness as a merits issue at trial. Id.\n(noting that “[t]he fact that Holloway and his witnesses were\nsecurity risks inside the courtroom was simply not a factor\nrelevant to [the jury’s] decision” of whether the “living\nconditions inside the prison were cruel and inhumane”).\n\n Relying on Tyars, Lemons and Holloway, the Second\nCircuit held that the Supreme Court’s concerns with\nshackling in criminal proceedings apply to parties in civil\nsuits, that physical restraints must be justified on the basis of\nsafety or security concerns, and that the court must take steps\nto minimize prejudice resulting from the presence of the\n\f20 CLAIBORNE V. BLAUSER\n\nrestraints. See Davidson v. Riley, 44 F.3d 1118, 1122–23 (2d\nCir. 1995). Like the plaintiff in Lemons, Davidson was a pro\nse state prisoner who appeared throughout his six-day trial\nin his Section 1983 lawsuit in handcuffs and leg irons. Id.\nat 1119. Even though Davidson’s claim involved\ncorrectional officers reading his legal mail, which did not\nbear on “either a propensity toward violence or risk of\nescape,” the court still found the potential for prejudice to be\nsignificant because “the verdict apparently was to turn on\nwhether the jury would believe Davidson and his prisoner-\nwitnesses or the [prison] witnesses.” Id. at 1126. After\ndetermining that the unnecessary imposition of restraints\nwas not harmless, the court vacated the judgment, remanded\nfor a hearing on the propriety of restraints, and ordered a new\ntrial. Id. at 1126–27.\n\n The Third Circuit most recently joined our court and the\nSecond, Seventh and Eighth Circuits to hold that “requiring\na party in a civil trial to appear in shackles ‘may well deprive\nhim of due process unless the restraints are necessary.’”\nSides v. Cherry, 609 F.3d 576, 581 (3d Cir. 2010) (quoting\nDavidson, 44 F.3d at 1122). As in Davidson, “the core issue\nin Sides’[s] case was credibility.” Id. at 584. The court\nultimately concluded that Sides’s shackling was harmless\nbecause the trial court took multiple steps to mitigate any\nprejudice. Id. at 584–85. Notably, however, the court\nrejected an alternative argument that the shackling was\nharmless “simply because [Sides’s] propensity for violence\nwas not directly at issue in the case.” Id. at 584.\n\n Thus, Tyars, Lemons and Holloway held that where a\nplaintiff’s dangerousness is a merits issue, visible shackling\nviolates due process unless justified on a case-by-case basis\nand steps are taken to mitigate prejudice. And, Davidson and\n\f CLAIBORNE V. BLAUSER 21\n\nSides recognized that prejudice may also arise where a core\nissue in the civil action is credibility.\n\n B.\n\n Against this backdrop, we evaluate (1) whether there was\nerror, (2) whether the error was obvious, (3) whether the\nerror affected substantial rights, and (4) whether the error\nseriously affects the fairness, integrity, or public reputation\nof judicial proceedings. C.B., 769 F.3d at 1018–19. 5\n\n 1.\n\n We readily conclude that the district court erred in\nallowing Claiborne to appear before the jury while shackled\nthroughout his three-day trial without first determining\nwhether the shackles were necessary.\n\n As in Duckett and Tyars, the record does not demonstrate\nany particular reason why Claiborne had to be visibly\nrestrained in front of the jury. See Duckett, 67 F.3d at 749;\nTyars, 709 F.2d at 1284. His status as a convicted felon is\nnot sufficient. Duckett, 67 F.3d at 749 (noting that murder\nconviction alone is not sufficient basis for shackling a\ndefendant at sentencing); cf. Lemons, 985 F.2d at 357\n(stating that “not all convicted felons are so dangerous and\nviolent that they must be brought to court and kept in\n 5\n “We may also take into consideration the costs of correcting an\nerror and—in borderline cases—the effect that a verdict may have on\nnonparties, although these considerations are by no means dispositive as\nto whether we will exercise our discretion to correct forfeited errors.”\nHoard v. Hartman, 904 F.3d 780, 787 (9th Cir. 2018) (internal quotation\nmarks omitted). Here, unlike in Hoard, we are not reviewing civil jury\ninstructions and this is not a borderline case. Therefore, we do not\nconsider the costs of correcting the error or the effect a verdict may have\non nonparties. See id. at 792 n.11.\n\f22 CLAIBORNE V. BLAUSER\n\nhandcuffs and leg irons”). Claiborne’s criminal record\nconsisted of nonviolent property and drug offenses.\nClaiborne, 2015 WL 5146746, at *1. One of the defendant\nofficers testified that she never had a conflict with Claiborne\naside from the 2010 incident. It was uncontested at trial that\nClaiborne was not considered a violent person in jail.\nFinally, there is nothing in the record to indicate that\nClaiborne disrupted court proceedings, acted disrespectfully\nin court, or attempted to escape. See Duckett, 67 F.3d at 749.\nThus, the district court erred by concluding that it would\nhave kept Claiborne shackled on the basis that he is a\nconvicted felon serving a lengthy prison sentence. Such\nreasoning fails to perform the particularized consideration of\nnecessity that we described in Duckett. Id. at 748.\n\n 2.\n\n The second prong of the plain error analysis requires the\nerror to be plain or obvious. Draper, 836 F.3d at 1085. Plain\nerror “is error that is so clear-cut, so obvious, a competent\ndistrict judge should be able to avoid it without benefit of\nobjection.” United States v. Gonzalez-Aparicio, 663 F.3d\n419, 428 (9th Cir. 2011). We agree with Claiborne that the\ndistrict court’s failure to address Claiborne’s shackling, and\ndenial of a new trial based on the shackling, were plainly\nerroneous.\n\n In 1983, we applied Supreme Court precedent to hold\nthat there was a due process violation where a civil litigant\n“was exhibited to the jury in physical restraints when the\nexpress question it was to decide was whether he was\ndangerous to himself or others.” Tyars, 709 F.2d at 1284.\nWe concluded that this posed “an even greater potential for\njuror bias than trying a criminal defendant in prison garb”\nbecause the “‘constant reminder of the accused’s condition\nimplicit in such distinctive, identifiable [restraints]’ surely\n\f CLAIBORNE V. BLAUSER 23\n\ncould have had a significant, deleterious effect on the jury’s\njudgment.” Id. at 1284–85 (alteration in original) (quoting\nEstelle v. Williams, 425 U.S. 501, 504–05 (1976)). Thus, we\nheld that shackling must be justified by “at least some prior\nshowing of necessity” where “the critical question the jury\nmust decide is whether the individual is dangerous to himself\nor others.” Id. at 1285.\n\n Defendants argue that our holding in Tyars is limited to\ncivil commitment proceedings, and cite to unpublished\ndispositions characterizing Tyars as such. See Carpenter v.\nPfeil, 617 Fed. App’x 658, 660 (9th Cir. 2015); Hartman v.\nMcCarthy, 74 F.3d 1245, at *1 (9th Cir. 1996) (unpublished\ntable decision). We are unpersuaded by these assertions for\nthree reasons. 6\n\n First, our holding in Tyars hinged on the “express\nquestion” that the jury was to decide—i.e., the petitioner’s\ndangerousness—not the type of civil proceeding. 709 F.2d\nat 1284–85. The commitment proceedings required the jury\ndetermine whether petitioner was dangerous to himself or\nothers. Id. That question is also core to Claiborne’s case.\nSecond, when describing the impact of Tyars in Duckett, we\nobserved generally that “courts have held that when an\nindividual’s level of dangerousness is a question the jury\nmust decide in a civil proceeding, it is a violation of the right\nto a fair trial to compel that individual to appear before the\njury bound in physical restraints.” 67 F.3d at 748. Thus, our\n\n 6\n Moreover, we of course are not bound by unpublished dispositions.\n9th Cir. R. 36-3(a); Pedroza v. BRB, 624 F.3d 926, 931 (9th Cir. 2010)\n(“[A]n unpublished decision is not precedent for our panel.”).\nNonetheless, we do not find them persuasive because Carpenter and\nHartman held in the alternative that even if Tyars and Duckett controlled,\nthere was no prejudice in those cases so any error was harmless.\nCarpenter, 617 Fed. App’x at 660; Hartman, 74 F.3d 1245, at *1.\n\f24 CLAIBORNE V. BLAUSER\n\ndicta in Duckett is consistent with the broad nature of our\nholding in Tyars. Finally, all other circuits that have\ndiscussed Tyars in the shackling context treated it as support\nfor applying in civil cases the Supreme Court’s rationale\nagainst routine shackling. See Lemons, 985 F.2d at 356–58\n& n.2; Davidson, 44 F.3d at 1122–23; Sides, 609 F.3d at 581.\n\n For these reasons, we conclude that the error was plain.\n\n 3.\n\n “The third prong of the plain error analysis requires that\nthe district court’s plain error . . . prejudiced the complaining\nparty or otherwise affected his or her substantial rights.”\nHoard, 904 F.3d at 790. “In most cases it means that the\nerror must have . . . affected the outcome of the district court\nproceedings. Id. at 791 (quoting United States v. Olano,\n507 U.S. 725, 734 (1993)). We conclude that Claiborne has\nalso satisfied this factor.\n\n Claiborne’s dangerousness was the key issue at trial. See\nTyars, 709 F.2d at 1285. The defendant officers justified\ntheir actions against Claiborne in the 2010 incident on the\nbasis that he presented a security risk. For instance, Officer\nBlauser testified that she placed Claiborne in traditional\nhandcuffs rather than waist chains because she believed he\nposed an immediate threat. She stated she did not honor\nClaiborne’s accommodations because “[s]ecurity supersedes\neverything.” Officer Blauser also explained that she and\nOfficer Martin escorted Claiborne straight across the yard\nrather on the paved track area “because of his aggression.”\nOn cross-examination, when asked why she took him down\non the yard, Officer Blauser responded it was because\nClaiborne presented a threat to her and her partner. In total,\nOfficer Blauser referred to Claiborne as “aggressive” or a\nthreat at least forty times during her testimony. In closing\n\f CLAIBORNE V. BLAUSER 25\n\narguments, the defense repeated Blauser’s testimony and\nemphasized that “she perceived a threat” and “took the most\nreasonable action under the circumstances.” In short, if the\njury believed that Claiborne posed a threat to the officers,\nthen defendants prevailed because their use of force was\njustified.\n\n Not only did Claiborne’s case hinge on his alleged\ndangerousness, the trial pitted his credibility against the\ndefendant officers’ credibility, and raised the issue of flight\nrisk as defendants claimed that he was trying to escape from\nthem during their escort in 2010. Claiborne’s visible\nshackling was a “constant reminder” to the jury of his\ncondition as a potentially violent and unreliable person who\nneeded to be restrained. Tyars, 709 F.2d at 1284 (quoting\nEstelle, 425 U.S. at 504); see also Lemons, 985 F.2d at 357.\nWe have also recognized that physical restraints “may\nconfuse and embarrass the [individual], thereby impairing\nhis mental faculties[,] and they may cause him pain.”\nDuckett, 67 F.3d at 748 (quoting Spain v. Rushen, 883 F.2d\n712, 720–21 (9th Cir. 1989) (collecting cases) (internal\nquotation marks omitted)). No doubt, shackles likely\nimpaired Claiborne as a pro se litigant in the presentation of\nhis case to a jury over three days. Id. Although the district\ncourt noted, ten months after the trial, that Claiborne’s hands\nwere not restrained, this is contradicted by Officer Blauser’s\ncontemporaneous observation at trial that Claiborne was\n“wearing [waist chains] right now” and that these waist\nchains were “big and bulky.”\n\n Defendants argue there was no prejudice because the\njury already knew Claiborne was a convicted felon. But the\nSupreme Court and other circuit courts have uniformly\nrejected the idea that a conviction alone could justify the use\nof shackles because restraint “almost inevitably affects\n\f26 CLAIBORNE V. BLAUSER\n\nadversely the jury’s perception of the character of the\ndefendant.” Deck, 544 U.S. at 633; see also Duckett, 67 F.3d\nat 749 (noting that a defendant’s status as a convicted felon,\n“[s]tanding alone, . . . is not a sufficient reason to impose\nphysical restraints”); Lemons, 985 F.2d at 357. We are also\nnot convinced by defendants’ insistence that the evidence\npresented at trial overwhelmingly supported the jury’s\nverdict. Crucially, the district court denied defendants’\nmotion for judgment as a matter of law at the close of\nevidence, demonstrating that a reasonable jury had a legally\nsufficient evidentiary basis to find in Claiborne’s favor. See\nFed. R. Civ. P. 50(a).\n\n Because of the inherent nature of visible shackles and its\ninterplay with the heart of Claiborne’s excessive force claim,\nwe conclude that Claiborne was prejudiced by the erroneous\nuse of shackling in the absence of any showing of a\ncompelling need for such restraints.\n\n 4.\n\n The final prong requires showing that “the district\ncourt’s errors are so grave as to seriously impair the fairness,\nintegrity, or public reputation of judicial proceedings.”\nHoard, 904 F.3d at 791 (internal quotation marks and\nalterations omitted). This is “undoubtedly the hardest\n[prong] to meet.” Id. Because of the fundamental nature of\nthe error, however, we find that this last prong has also been\nmet.\n\n Shackling “must be limited to cases urgently demanding\nthat action,” and because there was no showing of necessity\nin Claiborne’s case, he was unduly prejudiced in violation of\nhis due process right to a fair trial. Tyars, 709 F.2d at 1284;\nsee also Maus v. Baker, 747 F.3d 926, 927 (7th Cir. 2014)\n(“The sight of a shackled litigant is apt to make jurors think\n\f CLAIBORNE V. BLAUSER 27\n\nthey’re dealing with a mad dog . . .”). Just as an erroneous\njury instruction could “place[] a heavy thumb on the scale in\nfavor of the [d]efendants,” the erroneous use of shackling\nhere also constituted a “grave injustice” in depriving\nClaiborne of “a meaningful and fair opportunity to seek\nredress for alleged violations of his constitutional right to be\nfree from cruel and unusual punishment.” Hoard, 904 F.3d\nat 792.\n\n We also reiterate that shackling not only prejudices the\ninmate litigant in a case like this, it presents an affront to the\ndignity of the courtroom. Deck, 544 U.S. at 631 (citing\nAllen, 397 U.S. at 344). The Supreme Court has stressed that\n“[t]he routine use of shackles in the presence of juries would\nundermine [the] symbolic yet concrete objective[]” of\n“maintain[ing] a judicial process that is a dignified process.”\nId. Hence, shackling may only be used when there is an\n“individualized security determination[]” that “take[s]\naccount of the circumstances of the particular case.” Id.\nat 632. That individualized determination did not take place\nhere and the district court’s blanket assertion that it would\nhave ordered Claiborne shackled solely based on his status\nas a convicted felon was plainly erroneous. We therefore\nexercise our discretion on plain error review to reverse the\ndistrict court’s denial of the Rule 59(a) motion and remand\nfor a new trial on Claiborne’s claims against Officers\nBlauser and Martin.\n\n IV.\n\n Claiborne was denied a fair trial when he was visibly\nshackled before a jury that had to decide his dangerousness\nas it determined whether the jury believed his or defendants’\nversion of the events that underlie his Section 1983 claims.\nThere was no individualized determination of the security\nneed for such restraints, yet this mobility impaired plaintiff\n\f28 CLAIBORNE V. BLAUSER\n\nwas presented to the jury in shackles, corroborating the\ndefense’s position that Claiborne was an insubordinate,\naggressive inmate and thereby undermining his case. On\nremand, the district court has discretion to impose shackling\nduring the new trial, but it may only do so after a full hearing\nat which officers show a compelling need for security and\nthe court considers any less restrictive alternatives. 7 See\n\n\n\n 7\n Because we remand for a new trial on the shackling claim, we do\nnot address the merits of Claiborne’s evidentiary arguments that the\ndistrict court erred in barring his testimony about the ADA, preventing\nhim from introducing or testifying about his medical records, and\ndenying his request for a medical expert. The district court may revisit\nthese issues if Claiborne raises them again at the new trial.\n\n We note, however, that the district court appears to misstate the law\nwhen it denied Claiborne’s request for a medical expert under Federal\nRule of Evidence 706(a). Although the district court correctly\nrecognized that Rule 706(a) provides discretion to appoint a neutral\nexpert witness, see McKinney v. Anderson, 924 F.2d 1500, 1511 (9th Cir.\n1991), vacated on other grounds sub nom. Helling v. McKinney, 502\nU.S. 903 (1991), judgment reinstated, 959 F.2d 853 (9th Cir. 1992),\naff’d, 509 U.S. 25 (1993), the district court seemed to categorically limit\nthe relevance of a medical expert to testifying about a plaintiff’s current\ncondition. Yet courts have regularly considered requests for and\nappointed experts to review medical records and testify about prior\nmedical needs and treatment in deliberate indifference cases. See Gorton\nv. Todd, 793 F.Supp.2d 1171, 1179–81 (E.D. Cal. 2011) (collecting\ncases). Moreover, a medical expert can help with factfinding in\nexcessive force claims because “the extent of injury suffered by an\ninmate is one factor that may suggest ‘whether the [defendant’s] use of\nforce could plausibly have been thought necessary’ in a particular\nsituation.” Hudson v. McMillian, 503 U.S. 1, 7 (1992). If Claiborne\nrenews his request for appointment of a neutral medical expert on retrial,\nthe district court should weigh these considerations in exercising her\ndiscretion. See McKinney, 924 F.2d at 1511; see also Gorton,\n793 F.Supp.2d at 1185–86.\n\f CLAIBORNE V. BLAUSER 29\n\nDuckett, 67 F.3d at 748; Tyars, 709 F.2d at 1285; see also\nDavidson, 44 F.3d at 1126.\n\n For the foregoing reasons, we reverse the district court’s\ndenial of Claiborne’s Rule 59(a) motion and remand for a\nnew trial.\n\n REVERSED AND REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Dennis Claiborne v. Blauser\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-14\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n DENNIS GERALD CLAIBORNE, No. 16-16077\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:10-cv-02427-VAP\\n\\n BLAUSER, Correctional Officer;\\n S. MARTIN, Correctional ORDER AND\\n Officer, AMENDED\\n Defendants-Appellees. OPINION\\n\\n Appeal from the United States District Court\\n for the Eastern District of California\\n Virginia A. Phillips, Chief District Judge, Presiding\\n\\n Argued and Submitted February 6, 2019\\n San Francisco, California\\n\\n Filed June 28, 2019\\n Amended August 14, 2019\\n\\nBefore: Sidney R. Thomas, Chief Judge, Richard A. Paez,\\n Circuit Judge, and Gary Feinerman, * District Judge.\\n\\n\\n\\n\\n *\\n The Honorable Gary Feinerman, United States District Judge for\\nthe Northern District of Illinois, sitting by designation.\\n\\f2 CLAIBORNE V. BLAUSER\\n\\n Order;\\n Opinion by Judge Paez\\n\\n\\n SUMMARY **\\n\\n\\n Prisoner Civil Rights\\n\\n The panel reversed the district court’s denial of a motion\\nfor a new trial and remanded in an action brought pursuant\\nto 42 U.S.C. § 1983 by a convicted state inmate who alleged\\nthat he was shackled without justification during his three-\\nday trial on his Eighth Amendment excessive force and\\ndeliberate indifference to medical needs claims.\\n\\n The panel first noted that although the inmate did not\\nobject to the shackling during trial, he raised the issue in\\nsupport of his motion for a new trial. The panel applied plain\\nerror review.\\n\\n The panel held that because the inmate’s dangerousness\\nand flight risk were central issues at the trial, the district\\ncourt plainly erred in allowing him to be visibly shackled\\nwithout any showing of a sufficient need for such restraints.\\nThe panel held that on remand, the district court would have\\ndiscretion to impose shackling during the new trial, but it\\ncould only do so after a full hearing at which officers showed\\na compelling need for security and the court considered any\\nless restrictive alternatives.\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f CLAIBORNE V. BLAUSER 3\\n\\n COUNSEL\\n\\nJeremy M. Christiansen (argued), Gibson Dunn & Crutcher\\nLLP, Washington, D.C.; J. Brett Bylund and Blaine H.\\nEvanson, Gibson Dunn & Crutcher LLP, Irvine, California;\\nfor Plaintiff-Appellant.\\n\\nJaime Ganson (argued) and Arthur B. Mark III, Deputy\\nAttorneys General; Neah Huynh, Acting Supervising\\nDeputy Attorney General; Monica N. Anderson, Senior\\nAssistant Attorney General; Xavier Becerra, Attorney\\nGeneral; Office of the Attorney General, Sacramento,\\nCalifornia; for Defendants-Appellees.\\n\\n\\n ORDER\\n\\n The opinion filed in this case on June 28, 2019 is\\namended by replacing the text of footnote 7 on page 28,\\nClaiborne v. Blauser, 928 F.3d 794, 810 n.7 (9th Cir. 2019),\\nwith the following paragraphs:\\n\\n Because we remand for a new trial on the\\n shackling claim, we do not address the merits\\n of Claiborne’s evidentiary arguments that the\\n district court erred in barring his testimony\\n about the ADA, preventing him from\\n introducing or testifying about his medical\\n records, and denying his request for a\\n medical expert. The district court may revisit\\n these issues if Claiborne raises them again at\\n the new trial.\\n\\n We note, however, that the district court\\n appears to misstate the law when it denied\\n\\f4 CLAIBORNE V. BLAUSER\\n\\n Claiborne’s request for a medical expert\\n under Federal Rule of Evidence 706(a).\\n Although the district court correctly\\n recognized that Rule 706(a) provides\\n discretion to appoint a neutral expert witness,\\n see McKinney v. Anderson, 924 F.2d 1500,\\n 1511 (9th Cir. 1991), vacated on other\\n grounds sub nom. Helling v. McKinney,\\n 502 U.S. 903 (1991), judgment reinstated,\\n 959 F.2d 853 (9th Cir. 1992), aff’d, 509 U.S.\\n 25 (1993), the district court seemed to\\n categorically limit the relevance of a medical\\n expert to testifying about a plaintiff’s current\\n condition. Yet courts have regularly\\n considered requests for and appointed experts\\n to review medical records and testify about\\n prior medical needs and treatment in\\n deliberate indifference cases. See Gorton v.\\n Todd, 793 F.Supp.2d 1171, 1179–81 (E.D.\\n Cal. 2011) (collecting cases). Moreover, a\\n medical expert can help with factfinding in\\n excessive force claims because “the extent of\\n injury suffered by an inmate is one factor that\\n may suggest ‘whether the [defendant’s] use\\n of force could plausibly have been thought\\n necessary’ in a particular situation.” Hudson\\n v. McMillian, 503 U.S. 1, 7 (1992). If\\n Claiborne renews his request for appointment\\n of a neutral medical expert on retrial, the\\n district court should weigh these\\n considerations in exercising her discretion.\\n See McKinney, 924 F.2d at 1511; see also\\n Gorton, 793 F.Supp.2d at 1185–86.\\n\\nThe Amended Opinion is filed concurrently with this order.\\n\\f CLAIBORNE V. BLAUSER 5\\n\\n With the filing of the Amended Opinion, the panel has\\nunanimously voted to deny the Defendants-Appellees’\\npetition for panel rehearing. No further petitions for\\nrehearing may be filed.\\n\\n\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n The law has long forbidden the routine use of visible\\nshackling during a criminal defendant’s trial. Deck v.\\nMissouri, 544 U.S. 622, 626 (2005). Visible shackling\\nundermines the presumption of innocence, impedes the\\njury’s factfinding process, hampers presentation of a\\ndefense, and affronts the dignity and decorum of judicial\\nproceedings. Id. at 630–32. In this civil rights case under\\n42 U.S.C. § 1983, we consider whether the unjustified\\nshackling of a convicted state inmate during his three-day\\ntrial on Eighth Amendment excessive force and deliberate\\nindifference claims deprived him of a fair trial in violation\\nof the federal constitution. 1 Although the inmate did not\\nobject to the shackling during trial, he raised the issue in\\nsupport of his motion for a new trial, which the district court\\ndenied.\\n\\n We hold that the district court abused its discretion in\\ndenying a new trial. Because the inmate’s dangerousness\\nand flight risk were central issues at the trial, the district\\ncourt plainly erred in allowing him to be visibly shackled\\nwithout any showing of a sufficient need for such restraints.\\n\\n\\n 1\\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\\n\\f6 CLAIBORNE V. BLAUSER\\n\\nSee Tyars v. Finner, 709 F.2d 1274, 1284–85 (9th Cir. 1983).\\nWe therefore reverse and remand for a new trial.\\n\\n I.\\n\\n This appeal arises out of a lawsuit filed by Dennis Gerald\\nClaiborne who, proceeding pro se, sued Correctional\\nOfficers Jemini Blauser, Greg Martin, and other individual\\nofficials under Section 1983 for the use of excessive force\\nand deliberate indifference to his medical needs.\\n\\n A.\\n\\n Claiborne is a 63-year-old California state prison inmate\\nin the custody of the California Department of Corrections\\nand Rehabilitation (“CDCR”). He is serving a 60-years-to-\\nlife sentence under California’s Three Strikes Law for\\nattempted burglary and receipt of stolen property. People v.\\nClaiborne, No. B260391, 2015 WL 5146746, at *1 (Cal. Ct.\\nApp. Sept. 2, 2015).\\n\\n Claiborne is mobility impaired due to a right knee\\nreplacement in 2007 and ensuing chronic problems with that\\nknee. Given his condition, Claiborne participates in the\\nCDCR Disability Placement Program and receives certain\\naccommodations in prison: he is allowed to use a cane; he is\\nrestricted to housing on the lower level, with no stairs; and,\\nwhen escorting Claiborne within the institution, correctional\\nstaff must use “waist chains” and choose “relatively level\\nterrain and no obstructions in the path of travel.” Waist\\nchains are different from traditional handcuffs; whereas the\\nlatter are typically applied behind the back, the former allow\\na mobility impaired inmate to keep his hands at his sides\\nduring an escort, which lets him use any prescribed\\naccommodation devices like a cane. CDCR documents and\\nconveys information about Claiborne’s accommodations to\\n\\f CLAIBORNE V. BLAUSER 7\\n\\nprison officials through physician orders known as medical\\naccommodation “chronos” in his file. Claiborne also wears\\na green vest to alert officers that he is mobility impaired.\\n\\n The incident between Claiborne and Officers Blauser\\nand Martin took place on May 3, 2010, while Claiborne was\\nhoused at California’s High Desert State Prison, a Level\\nFour security prison. As Claiborne was waiting for his\\nmedication in the morning “pill line,” Correctional Officer\\nDaniel McBride, stationed in an observation tower, believed\\nhe saw Claiborne socializing with other inmates in the line.\\nOfficer McBride called Officer Blauser, who was working\\nin Claiborne’s housing unit along with her partner, Officer\\nMartin, and asked Blauser to counsel Claiborne for\\nunnecessarily lingering in the pill line. After waiting thirty\\nminutes in the pill line, Claiborne received his medication\\nand returned to his housing unit where he was admonished\\nby Officer Blauser. She decided to “put a cap” on\\nClaiborne’s door, meaning he would spend the rest of the\\nday locked up in his cell. Because Claiborne had intended\\nto present at a Bible study group later that day, he asked to\\nspeak with the sergeant, Officer Blauser’s supervisor, to\\ncontest the punishment and explain that he had been properly\\nwaiting in line and not socializing.\\n\\n The account of the facts diverge drastically from there.\\nAccording to Claiborne, he was acting respectfully toward\\nOfficer Blauser when she told him to “cuff up.” He\\ncomplied and, as Officers Blauser and Martin started to\\nescort him, Claiborne informed them that it was difficult for\\nhim to walk and use his cane with his hands cuffed behind\\nhis back. He mentioned his chrono for waist chains but\\nOfficer Blauser told Officer Martin to take Claiborne’s cane.\\nOfficer Martin said he would help hold Claiborne up as the\\nthree of them walked to the program office to speak with the\\n\\f8 CLAIBORNE V. BLAUSER\\n\\nsergeant. Claiborne, however, had difficulty walking with\\nOfficers Blauser and Martin. They escorted him straight\\nacross the yard, rather than along the sidewalk circling the\\nyard, despite Claiborne’s chrono providing for level terrain.\\nBecause the yard was uneven, Claiborne hyperextended his\\nright leg, causing his knee to give out partway across the\\nyard. Claiborne lost balance and shifted rightward, causing\\nOfficer Blauser to order him to stop resisting. Claiborne\\ntried to explain that he was not resisting and that his knee\\nwas bothering him because the officers were pulling him too\\nquickly.\\n\\n When the three were almost at the program office,\\nClaiborne’s right leg hyperextended again as he tried to pick\\nthe leg up three to five inches from the dirt yard onto the\\npavement. Because he had no cane to catch himself,\\nClaiborne leaned to his right again, causing Officer Blauser\\nto shout “he’s resisting” and pull him down to the ground.\\nShe jumped on his right side, including his replaced knee,\\nand pulled his hair and hit him in the face a few times. Other\\nofficers quickly rushed to the scene and Claiborne heard\\nindividuals ask, “Where’s his cane?” and “Why isn’t he in\\nwaist chains?” Claiborne was eventually taken into a\\nholding cell and then interviewed by a sergeant, which was\\nrecorded by a video camera. Before turning on the camera\\nto record the interview, the sergeant warned Claiborne that\\nif he reported excessive force, he would be taken to the\\n“hole,” in other words, administrative segregation, for an\\nunknown amount of time. Worried about whether his\\nmedical needs would be met in the hole, Claiborne\\neventually stated on camera, after extended back-and-forth\\nwith the sergeant, that there was no excessive force used\\nagainst him.\\n\\f CLAIBORNE V. BLAUSER 9\\n\\n Officers Blauser and Martin presented a different\\naccount. They were both aware of Claiborne’s mobility\\nimpairment at the time of the incident. According to Officer\\nBlauser, while she counseled Claiborne for lingering in the\\npill line, he became “really aggravated” and started raising\\nhis voice at her, causing her to feel uncomfortable.\\nClaiborne then walked toward her while holding, not using,\\nhis cane. Because she did not feel safe, Officer Blauser told\\nClaiborne to “cuff up,” and he immediately turned around\\nand complied. She asked her partner, Officer Martin, to\\nassist her with escorting Claiborne to the program office.\\nOfficer Martin took Claiborne’s cane, and they each\\nsupported him by holding onto his bicep or arm on each side.\\nNeither Officer Blauser nor Officer Martin recalled\\nClaiborne saying anything about needing to use waist chains.\\nMoreover, because it would have taken a few extra minutes\\nto obtain waist chains, Officer Blauser decided to use\\nhandcuffs due to Claiborne’s aggression and defiance of her\\norder to return to his cell.\\n\\n According to Officers Blauser and Martin, they escorted\\nClaiborne straight across the yard because it was the quickest\\nand most direct path to the program office and avoided\\nwalking amongst other inmates. They escorted him slowly\\nand did not perceive any problems with his walking. Rather,\\nthey believed that Claiborne tried to break away from them\\ntwice, once halfway across the yard, and a second time close\\nto the program office. At first, Officer Blauser ordered\\nClaiborne not to pull his arm away but he continued to act\\naggressively, yelling and trying to pull away from her.\\nWhen Claiborne pulled his arm away from her a second\\ntime, Officer Blauser decided to pull him down and called a\\n“code one” over the radio. Officers Blauser and Martin used\\ntheir weight to hold Claiborne down on the ground. They\\ndenied that Officer Blauser jumped on Claiborne, pulled his\\n\\f10 CLAIBORNE V. BLAUSER\\n\\nhair or punched him in the face. After other officers escorted\\nClaiborne away, Sergeant Officer Kenneth Gullion followed\\nprotocol and conducted a video-recorded “use-of-force”\\ninterview, asking Claiborne about what took place. Sergeant\\nGullion did not remember what happened prior to\\ninterviewing Claiborne on video. A separate officer wrote\\nClaiborne up for a rule violation, resisting a peace officer.\\n\\n Shortly after the incident, a nurse examined Claiborne\\nand completed a medical report, noting that he had two\\nabrasions, one each on his left knee and left cheek.\\nAfterward, Claiborne experienced more problems with his\\nright knee, and doctors determined that he had significant\\ninjuries, including bursitis. Following a series of\\nevaluations, Claiborne was assessed to have “[f]ailed right\\ntotal knee arthroplasty.” He underwent a revision procedure\\nin 2012, but was told that his knee did not respond properly\\nto the surgery and could not be fixed any more. Claiborne\\nalso had surgery on one of his shoulders in July 2015, when\\nhe was told that he waited too long to fix it.\\n\\n B.\\n\\n After exhausting the prison’s administrative process,\\nClaiborne filed suit in district court. Proceeding pro se, he\\nsued Officers Blauser, Martin and other individual officers\\nunder Section 1983 for various claims, including the use of\\nexcessive force and deliberate indifference to his medical\\nneeds in violation of his Eighth Amendment rights. The\\nother named defendant officers and other claims, including\\nbattery, negligent infliction of emotional distress and\\nintentional infliction of emotional distress, were eventually\\ndismissed from the case. After the district court denied\\ncross-motions for summary judgment, the case proceeded to\\ntrial on the two Eighth Amendment claims against Officers\\nBlauser and Martin.\\n\\f CLAIBORNE V. BLAUSER 11\\n\\n The trial lasted three days, at the start of which the\\ndistrict court noted, outside of the jury’s presence, that\\nClaiborne was shackled. Claiborne testified on his own\\nbehalf, and two fellow inmates also testified on his behalf.\\nThe defense presented testimony from six officers including\\nOfficers Blauser and Martin, the tower guard officer\\n(McBride), and Sergeant Gullion. The jury reached a verdict\\nfor both defendants on both claims.\\n\\n Claiborne filed a timely motion for new trial and relief\\nfrom judgment, raising various arguments. Most relevant to\\nour analysis, Claiborne argued that he was entitled to a new\\ntrial under Federal Rule of Civil Procedure 59(a) because he\\nwas visibly shackled while litigating his case in front of the\\njury. He also argued for relief from judgment under Federal\\nRule of Civil Procedure 60(b) because the district court\\nbarred his testimony about the Americans with Disabilities\\nAct (“ADA”), prevented him from introducing or testifying\\nabout his medical records, and denied his request for a\\nmedical expert. The district court dismissed all the\\narguments and denied the motion.\\n\\n II.\\n\\n When reviewing denial of a motion for new trial under\\nFederal Rule of Civil Procedure 59(a), our default is to\\nreview for abuse of discretion. See Hung Lam v. City of San\\nJose, 869 F.3d 1077, 1084 (9th Cir. 2017). But here,\\nClaiborne did not object to his shackling at trial so we must\\naddress the appropriate standard of review.\\n\\n Errors not objected to at trial are generally subject to\\nwaiver or forfeiture. “Forfeiture is the failure to make a\\ntimely assertion of a right, whereas waiver is the intentional\\nrelinquishment or abandonment of a known right.” United\\nStates v. Perez, 116 F.3d 840, 845 (9th Cir. 1997) (en banc)\\n\\f12 CLAIBORNE V. BLAUSER\\n\\n(internal quotation marks omitted); see also Crowley v.\\nEpiCept Corp., 883 F.3d 739, 748 (9th Cir. 2018) (per\\ncuriam) (applying same standard in civil context).\\n“Forfeited rights are reviewable for plain error, while waived\\nrights are not.” Id.\\n\\n Under plain error review, we may reverse only where:\\n(1) there was an error; (2) the error was obvious; (3) the error\\naffected substantial rights; and, (4) the error seriously affects\\nthe fairness, integrity, or public reputation of judicial\\nproceedings. See C.B. v. City of Sonora, 769 F.3d 1005,\\n1018–19 (9th Cir. 2014) (en banc). In other words, plain\\nerror requires reversal where “review is necessary to prevent\\na miscarriage of justice.” Draper v. Rosario, 836 F.3d 1072,\\n1085 (9th Cir. 2016) (internal quotation marks omitted).\\n\\n Defendants argue that Claiborne forfeited the error\\nbecause he failed to object to his shackling at trial, and so,\\nhis claim should be subject to plain error review. 2 Claiborne\\ncontends that plain error is inapplicable because he timely\\nraised the shackling issue in his motion for a new trial, and\\nthe district court addressed the claim on the merits, holding\\nthat it would have imposed shackling over any objection\\n\\n 2\\n We find no merit to defendants’ alternative argument that\\nClaiborne invited the error. We impose a high standard for applying the\\ninvited error doctrine. See United States v. Lindsey, 634 F.3d 541, 555\\n(9th Cir. 2011) (noting that “in order for the invited error doctrine to\\napply, a [party] must both invite the error and relinquish a known right”\\n(citing Perez, 116 F.3d at 845)). Claiborne’s mention of his shackles in\\nhis opening statement, read in context, suggests that he was trying to\\nintroduce himself to the jury and not hide the fact that he is incarcerated.\\nThere is no indication in the record that Claiborne, who was pro se in the\\ndistrict court, “both invited the error, and relinquished a known right” to\\na fair trial without shackling. Perez, 116 F.3d at 845.\\n\\f CLAIBORNE V. BLAUSER 13\\n\\nbecause Claiborne is a convicted felon serving a lengthy\\nprison sentence.\\n\\n The district court’s denial of a motion for a new trial is\\nreversible if the court made a mistake of law. Molski v. M.J.\\nCable, Inc., 481 F.3d 724, 729 (9th Cir. 2007). As we\\nelaborate below, we agree that the district court disregarded\\nthe law on shackling. However, in prior cases involving\\nlegal claims that were not raised during trial, we still\\nundertook plain error review when reviewing a denial of a\\nmotion for a new trial. 3 See, e.g., Settlegoode v. Portland\\nPub. Sch., 371 F.3d 503, 517 (9th Cir. 2004) (“There is an\\neven ‘high[er] threshold’ for granting a new trial where, as\\nhere, [the moving party] failed to object to the alleged\\nmisconduct during trial.” (first alteration in original)\\n(quoting Kaiser Steel Corp. v. Frank Coluccio Constr. Co.,\\n785 F.2d 656, 658 (9th Cir. 1986))); Hemmings v. Tidyman’s\\nInc., 285 F.3d 1174, 1193 (9th Cir. 2002) (“We will review\\nfor plain or fundamental error, absent a contemporaneous\\nobjection . . .”).\\n\\n The policy behind plain error review is to ensure that the\\nparties and district court are on notice as to any alleged errors\\nso that the court may address the objection, correct any error,\\nand create a proper record for appeal. See Hemmings,\\n285 F.3d at 1193 (observing that “the trial judge is in a\\nsuperior position to evaluate the likely effect of the alleged\\nmisconduct and to fashion an appropriate remedy”); see also\\n 3\\n This is analogous to our review of instructional error claims when\\nthere was no objection at trial. 11 Charles A. Wright et al., Federal\\nPractice and Procedure § 2805 (3d ed. 2012). Under Federal Rule of\\nCivil Procedure 51(d)(2), we review a challenge to civil jury instructions\\nfor plain error in the absence of a timely objection. C.B., 769 F.3d\\nat 1017–19.\\n\\f14 CLAIBORNE V. BLAUSER\\n\\nUnited States v. $11,500.00 in U.S. Currency, 869 F.3d\\n1062, 1075 (9th Cir. 2017).\\n\\n Here, the district court’s holding on the merits—that it\\nwould have imposed shackling over any timely objection—\\ndoes give us some pause as to the standard of review. 4 We\\nhave recognized one limited exception to plain error review\\nin the civil context, the “pointless formality” exception.\\nChess v. Dovey, 790 F.3d 961, 970–72 (9th Cir. 2015); see\\nalso Shorter v. Baca, 895 F.3d 1176, 1182–83 (9th Cir.\\n2018). In Chess, we declined to apply plain error where the\\nplaintiff was pro se and “the district court and opposing party\\nwere fully aware of the potential problem with, and would-\\nbe objection to, [a jury] instruction,” even though the\\nplaintiff did not raise the issue himself. 790 F.3d at 964.\\nBecause the district court and defendants were on notice, we\\nconcluded that “any objection by Chess would have been\\n‘superfluous and futile,’” id. at 972 (quoting Obsidian Fin.\\nGrp. v. Cox, 740 F.3d 1284, 1289 (9th Cir. 2014)), and we\\ndeclined to “punish a pro se litigant with plain error rather\\nthan de novo review simply because he failed to say the\\nwords ‘I object,’” id. at 971.\\n\\n Claiborne, too, was pro se and the district court’s post-\\ntrial ruling suggests that any timely objection would have\\n\\n 4\\n The district court’s ruling distinguishes Claiborne’s situation from\\nthat in Hemmings, in which the defendant moved for a new trial based\\non inappropriate comments made by the plaintiffs’ counsel in closing\\nargument. 285 F.3d at 1192. The district court noted that it remembered\\nthe comment and “would have sustained an objection had one been made\\nbecause it was totally improper.” Id. at 1192–93. Because a timely\\nobjection would have made a difference and changed the district court’s\\nactions, application of plain error review was appropriate under the\\nrationale of giving the trial judge an opportunity to cure any prejudice.\\nId. at 1193. That concern does not apply here.\\n\\f CLAIBORNE V. BLAUSER 15\\n\\nbeen futile. We recognize, however, the policy concerns\\nbehind the plain error doctrine and therefore decline to\\nreview for abuse of discretion. See Hemmings, 285 F.3d at\\n1193 (noting that “allowing a party to wait to raise the error\\nuntil after the negative verdict encourages that party to sit\\nsilent in the face of claimed error”). Accordingly, we review\\nfor plain error.\\n\\n III.\\n\\n The district court must grant a motion for a new trial\\nwhere “the verdict is against the weight of the evidence,”\\n“the damages are excessive” or, “for other reasons, the trial\\nwas not fair to the [moving] party.” Montgomery Ward &\\nCo. v. Duncan, 311 U.S. 243, 251 (1940). Here, we consider\\nan argument based on the latter. Claiborne, now represented\\nby pro bono counsel, argues that he was deprived of a fair\\ntrial because he was visibly shackled in front of the jury\\nthroughout the three-day trial, and that the district court erred\\nby denying his Rule 59(a) motion for a new trial. Even under\\nthe more demanding plain error standard, we agree that the\\ndistrict court erred in denying Claiborne’s motion for a new\\ntrial.\\n\\n A.\\n\\n To understand why the district court plainly erred, we\\nfirst review the case law on visible shackling.\\n\\n It is well established that due process “forbids the use of\\nvisible shackl[ing] . . . during the guilt phase [of a criminal\\ntrial], unless that use is ‘justified by an essential state\\ninterest.’” Deck, 544 U.S. at 624 (citing Holbrook v. Flynn,\\n475 U.S. 560, 568–69 (1986); Illinois v. Allen, 397 U.S. 337,\\n343–44 (1970)). The Supreme Court has identified “three\\nfundamental legal principles” which are “undermined” or\\n\\f16 CLAIBORNE V. BLAUSER\\n\\n“diminished” by the use of visible restraints. Id. at 630–31.\\nFirst, the accused enjoys a presumption of innocence, which\\nvisible shackling undermines by suggesting that the\\ndefendant need be restrained. Id. at 630. This in turn\\n“undermines . . . the related fairness of the factfinding\\nprocess.” Id. Second, the accused has a right to a\\nmeaningful defense from counsel. Id. at 631. Restraints\\nmight physically impair communication between a\\ndefendant and his counsel, or impair the defendant’s\\nparticipation in his own defense. Id. “Third, judges must\\nseek to maintain a judicial process that is a dignified [one].”\\nId. Because “[t]he courtroom’s formal dignity . . . includes\\nthe respectful treatment of defendants,” “the use of shackles\\nat trial ‘affront[s]’ the ‘dignity and decorum of judicial\\nproceedings that the judge is seeking to uphold.’” Id.\\n(quoting Allen, 397 U.S. at 344).\\n\\n In Deck, the Court held that these considerations apply\\nwith equal force to penalty proceedings in capital cases, even\\nthough “the presumption of innocence no longer applies.”\\nId. at 632. Hence, before imposing restraints on criminal\\ndefendants, the trial court must take into account the\\nparticular circumstances of each case, including “special\\nsecurity needs or escape risks[] related to the defendant,” id.\\nat 633. Where the court orders visible shackling without\\nadequate justification, “the defendant need not demonstrate\\nactual prejudice to make out a due process violation,” and\\nthe burden is on the government to “prove beyond a\\nreasonable doubt that the shackling error complained of did\\nnot contribute to the verdict obtained.” Id. at 635 (alterations\\nand internal quotation marks omitted).\\n\\n Prior to Deck, we have long recognized that the\\nprohibition against routine visible shackling applies even\\nwhen the presumption of innocence does not, including in\\n\\f CLAIBORNE V. BLAUSER 17\\n\\nthe civil context. In Duckett v. Godinez, we anticipated Deck\\nby holding that “shackling a defendant during a [capital]\\nsentencing hearing before a jury is an inherently prejudicial\\npractice which comports with due process only when used\\nas a last resort to protect an essential state interest[,] such as\\nmaintaining public safety or assuring the decorum of the\\nproceedings.” 67 F.3d 734, 747 (9th Cir. 1995). We have\\nthus long required trial courts to engage in a two-step process\\nbefore shackling a criminal defendant: “[f]irst, the court\\nmust be persuaded by compelling circumstances” that the\\nmeasure is necessary to “maintain security[;]” and\\n“[s]econd, the court must ‘pursue less restrictive alternatives\\nbefore imposing physical restraints.” Id. at 748 (internal\\nquotation marks omitted). We held that it was error to permit\\nshackling solely on the basis that the defendant was a\\nconvicted felon, and so remanded for the district court to\\ndetermine whether this constitutional error was harmless. Id.\\nat 749–50.\\n\\n Notably, we reached our holding in Duckett “by analogy\\nto the treatment of the shackling issue in civil cases.” Id.\\nat 748 (citing Tyars, 709 F.2d at 1284–85; Lemons v.\\nSkidmore, 985 F.2d 354, 356–58 (7th Cir. 1993); and\\nHolloway v. Alexander, 957 F.2d 529, 530 (8th Cir. 1992)).\\nAs recounted below, this court has recognized the inherent\\nprejudice associated with unjustified shackling in civil\\nproceedings as early as 1983, and other circuits have relied\\non our reasoning to expand the prohibition on routine\\nshackling to other civil contexts.\\n\\n In Tyars, the petitioner challenged the use of restraints to\\nbind him in the presence of the jury during his involuntary\\ncivil commitment proceedings. 709 F.2d at 1276, 1284. We\\nobserved that “[i]t requires no great extension of Supreme\\nCourt precedent to conclude that this may have violated his\\n\\f18 CLAIBORNE V. BLAUSER\\n\\nrights under the due process clause.” Id. at 1284. We\\nreasoned:\\n\\n Although the criminal case precedents do not\\n necessarily apply in a civil proceeding, we\\n find them persuasive. The likelihood of\\n prejudice inherent in exhibiting the subject of\\n a civil commitment hearing to the jury while\\n bound in physical restraints, when the critical\\n question the jury must decide is whether the\\n individual is dangerous to himself or others,\\n is simply too great to be countenanced\\n without at least some prior showing of\\n necessity. In the absence of any such\\n demonstrable or articulable necessity, and in\\n the absence of any showing that less\\n restrictive means not embodying the same\\n potential for prejudice could have maintained\\n order in the courtroom, the circumstances\\n deprived the proceeding of the appearance of\\n evenhanded justice which is at the core of due\\n process.\\n\\nId. at 1285 (emphasis added) (alterations, internal quotation\\nmarks and citation omitted). We instructed the district court\\non remand to “determine whether the State has any\\njustification . . . to support the necessity of physically\\nrestraining Tyars in the presence of the jury.” Id. at 1285–\\n86.\\n\\n Since Tyars, at least four other circuits have adopted a\\nsimilar test when evaluating the use of shackling in civil\\ntrials. In Lemons, the Seventh Circuit held that it was\\nimpermissible to rely solely on the opinion of a state\\ncorrections officer to shackle a plaintiff prison inmate in a\\n\\f CLAIBORNE V. BLAUSER 19\\n\\nSection 1983 action alleging excessive force by corrections\\nofficers. 985 F.2d at 356. The plaintiff claimed that\\ndefendant officers attacked and beat him severely in his cell,\\nwhile defendants asserted that the plaintiff was “ranting and\\nraving,” and had to be subdued. Id. at 355. The Lemons\\ncourt cited with approval Tyars, id. at 356 n.2, and concluded\\nthat, as in the criminal context, the appearance of a civil\\nplaintiff in handcuffs and leg irons “suggested to the jury that\\nthe plaintiff was dangerous and violent, so that whatever\\nforce the guards had used was probably necessary, and not\\nexcessive,” id. at 359. “[S]hackles inevitably prejudiced the\\njury” because “plaintiff’s tendency towards violence was at\\nissue.” Id. at 357. The court therefore remanded for a new\\ntrial and required a hearing to determine what, if any,\\nrestraints were necessary. Id. at 359.\\n\\n The Eighth Circuit in Holloway also recognized that, as\\na general rule, inmate civil plaintiffs should not have to\\nappear in court in shackles unless there was a showing of\\nneed and steps to mitigate any potential prejudice. 957 F.2d\\nat 530. The court found no prejudice in Holloway because\\nthe plaintiff challenged the constitutionality of living\\nconditions in the state prison, which did not raise the\\nplaintiff’s dangerousness as a merits issue at trial. Id.\\n(noting that “[t]he fact that Holloway and his witnesses were\\nsecurity risks inside the courtroom was simply not a factor\\nrelevant to [the jury’s] decision” of whether the “living\\nconditions inside the prison were cruel and inhumane”).\\n\\n Relying on Tyars, Lemons and Holloway, the Second\\nCircuit held that the Supreme Court’s concerns with\\nshackling in criminal proceedings apply to parties in civil\\nsuits, that physical restraints must be justified on the basis of\\nsafety or security concerns, and that the court must take steps\\nto minimize prejudice resulting from the presence of the\\n\\f20 CLAIBORNE V. BLAUSER\\n\\nrestraints. See Davidson v. Riley, 44 F.3d 1118, 1122–23 (2d\\nCir. 1995). Like the plaintiff in Lemons, Davidson was a pro\\nse state prisoner who appeared throughout his six-day trial\\nin his Section 1983 lawsuit in handcuffs and leg irons. Id.\\nat 1119. Even though Davidson’s claim involved\\ncorrectional officers reading his legal mail, which did not\\nbear on “either a propensity toward violence or risk of\\nescape,” the court still found the potential for prejudice to be\\nsignificant because “the verdict apparently was to turn on\\nwhether the jury would believe Davidson and his prisoner-\\nwitnesses or the [prison] witnesses.” Id. at 1126. After\\ndetermining that the unnecessary imposition of restraints\\nwas not harmless, the court vacated the judgment, remanded\\nfor a hearing on the propriety of restraints, and ordered a new\\ntrial. Id. at 1126–27.\\n\\n The Third Circuit most recently joined our court and the\\nSecond, Seventh and Eighth Circuits to hold that “requiring\\na party in a civil trial to appear in shackles ‘may well deprive\\nhim of due process unless the restraints are necessary.’”\\nSides v. Cherry, 609 F.3d 576, 581 (3d Cir. 2010) (quoting\\nDavidson, 44 F.3d at 1122). As in Davidson, “the core issue\\nin Sides’[s] case was credibility.” Id. at 584. The court\\nultimately concluded that Sides’s shackling was harmless\\nbecause the trial court took multiple steps to mitigate any\\nprejudice. Id. at 584–85. Notably, however, the court\\nrejected an alternative argument that the shackling was\\nharmless “simply because [Sides’s] propensity for violence\\nwas not directly at issue in the case.” Id. at 584.\\n\\n Thus, Tyars, Lemons and Holloway held that where a\\nplaintiff’s dangerousness is a merits issue, visible shackling\\nviolates due process unless justified on a case-by-case basis\\nand steps are taken to mitigate prejudice. And, Davidson and\\n\\f CLAIBORNE V. BLAUSER 21\\n\\nSides recognized that prejudice may also arise where a core\\nissue in the civil action is credibility.\\n\\n B.\\n\\n Against this backdrop, we evaluate (1) whether there was\\nerror, (2) whether the error was obvious, (3) whether the\\nerror affected substantial rights, and (4) whether the error\\nseriously affects the fairness, integrity, or public reputation\\nof judicial proceedings. C.B., 769 F.3d at 1018–19. 5\\n\\n 1.\\n\\n We readily conclude that the district court erred in\\nallowing Claiborne to appear before the jury while shackled\\nthroughout his three-day trial without first determining\\nwhether the shackles were necessary.\\n\\n As in Duckett and Tyars, the record does not demonstrate\\nany particular reason why Claiborne had to be visibly\\nrestrained in front of the jury. See Duckett, 67 F.3d at 749;\\nTyars, 709 F.2d at 1284. His status as a convicted felon is\\nnot sufficient. Duckett, 67 F.3d at 749 (noting that murder\\nconviction alone is not sufficient basis for shackling a\\ndefendant at sentencing); cf. Lemons, 985 F.2d at 357\\n(stating that “not all convicted felons are so dangerous and\\nviolent that they must be brought to court and kept in\\n 5\\n “We may also take into consideration the costs of correcting an\\nerror and—in borderline cases—the effect that a verdict may have on\\nnonparties, although these considerations are by no means dispositive as\\nto whether we will exercise our discretion to correct forfeited errors.”\\nHoard v. Hartman, 904 F.3d 780, 787 (9th Cir. 2018) (internal quotation\\nmarks omitted). Here, unlike in Hoard, we are not reviewing civil jury\\ninstructions and this is not a borderline case. Therefore, we do not\\nconsider the costs of correcting the error or the effect a verdict may have\\non nonparties. See id. at 792 n.11.\\n\\f22 CLAIBORNE V. BLAUSER\\n\\nhandcuffs and leg irons”). Claiborne’s criminal record\\nconsisted of nonviolent property and drug offenses.\\nClaiborne, 2015 WL 5146746, at *1. One of the defendant\\nofficers testified that she never had a conflict with Claiborne\\naside from the 2010 incident. It was uncontested at trial that\\nClaiborne was not considered a violent person in jail.\\nFinally, there is nothing in the record to indicate that\\nClaiborne disrupted court proceedings, acted disrespectfully\\nin court, or attempted to escape. See Duckett, 67 F.3d at 749.\\nThus, the district court erred by concluding that it would\\nhave kept Claiborne shackled on the basis that he is a\\nconvicted felon serving a lengthy prison sentence. Such\\nreasoning fails to perform the particularized consideration of\\nnecessity that we described in Duckett. Id. at 748.\\n\\n 2.\\n\\n The second prong of the plain error analysis requires the\\nerror to be plain or obvious. Draper, 836 F.3d at 1085. Plain\\nerror “is error that is so clear-cut, so obvious, a competent\\ndistrict judge should be able to avoid it without benefit of\\nobjection.” United States v. Gonzalez-Aparicio, 663 F.3d\\n419, 428 (9th Cir. 2011). We agree with Claiborne that the\\ndistrict court’s failure to address Claiborne’s shackling, and\\ndenial of a new trial based on the shackling, were plainly\\nerroneous.\\n\\n In 1983, we applied Supreme Court precedent to hold\\nthat there was a due process violation where a civil litigant\\n“was exhibited to the jury in physical restraints when the\\nexpress question it was to decide was whether he was\\ndangerous to himself or others.” Tyars, 709 F.2d at 1284.\\nWe concluded that this posed “an even greater potential for\\njuror bias than trying a criminal defendant in prison garb”\\nbecause the “‘constant reminder of the accused’s condition\\nimplicit in such distinctive, identifiable [restraints]’ surely\\n\\f CLAIBORNE V. BLAUSER 23\\n\\ncould have had a significant, deleterious effect on the jury’s\\njudgment.” Id. at 1284–85 (alteration in original) (quoting\\nEstelle v. Williams, 425 U.S. 501, 504–05 (1976)). Thus, we\\nheld that shackling must be justified by “at least some prior\\nshowing of necessity” where “the critical question the jury\\nmust decide is whether the individual is dangerous to himself\\nor others.” Id. at 1285.\\n\\n Defendants argue that our holding in Tyars is limited to\\ncivil commitment proceedings, and cite to unpublished\\ndispositions characterizing Tyars as such. See Carpenter v.\\nPfeil, 617 Fed. App’x 658, 660 (9th Cir. 2015); Hartman v.\\nMcCarthy, 74 F.3d 1245, at *1 (9th Cir. 1996) (unpublished\\ntable decision). We are unpersuaded by these assertions for\\nthree reasons. 6\\n\\n First, our holding in Tyars hinged on the “express\\nquestion” that the jury was to decide—i.e., the petitioner’s\\ndangerousness—not the type of civil proceeding. 709 F.2d\\nat 1284–85. The commitment proceedings required the jury\\ndetermine whether petitioner was dangerous to himself or\\nothers. Id. That question is also core to Claiborne’s case.\\nSecond, when describing the impact of Tyars in Duckett, we\\nobserved generally that “courts have held that when an\\nindividual’s level of dangerousness is a question the jury\\nmust decide in a civil proceeding, it is a violation of the right\\nto a fair trial to compel that individual to appear before the\\njury bound in physical restraints.” 67 F.3d at 748. Thus, our\\n\\n 6\\n Moreover, we of course are not bound by unpublished dispositions.\\n9th Cir. R. 36-3(a); Pedroza v. BRB, 624 F.3d 926, 931 (9th Cir. 2010)\\n(“[A]n unpublished decision is not precedent for our panel.”).\\nNonetheless, we do not find them persuasive because Carpenter and\\nHartman held in the alternative that even if Tyars and Duckett controlled,\\nthere was no prejudice in those cases so any error was harmless.\\nCarpenter, 617 Fed. App’x at 660; Hartman, 74 F.3d 1245, at *1.\\n\\f24 CLAIBORNE V. BLAUSER\\n\\ndicta in Duckett is consistent with the broad nature of our\\nholding in Tyars. Finally, all other circuits that have\\ndiscussed Tyars in the shackling context treated it as support\\nfor applying in civil cases the Supreme Court’s rationale\\nagainst routine shackling. See Lemons, 985 F.2d at 356–58\\n& n.2; Davidson, 44 F.3d at 1122–23; Sides, 609 F.3d at 581.\\n\\n For these reasons, we conclude that the error was plain.\\n\\n 3.\\n\\n “The third prong of the plain error analysis requires that\\nthe district court’s plain error . . . prejudiced the complaining\\nparty or otherwise affected his or her substantial rights.”\\nHoard, 904 F.3d at 790. “In most cases it means that the\\nerror must have . . . affected the outcome of the district court\\nproceedings. Id. at 791 (quoting United States v. Olano,\\n507 U.S. 725, 734 (1993)). We conclude that Claiborne has\\nalso satisfied this factor.\\n\\n Claiborne’s dangerousness was the key issue at trial. See\\nTyars, 709 F.2d at 1285. The defendant officers justified\\ntheir actions against Claiborne in the 2010 incident on the\\nbasis that he presented a security risk. For instance, Officer\\nBlauser testified that she placed Claiborne in traditional\\nhandcuffs rather than waist chains because she believed he\\nposed an immediate threat. She stated she did not honor\\nClaiborne’s accommodations because “[s]ecurity supersedes\\neverything.” Officer Blauser also explained that she and\\nOfficer Martin escorted Claiborne straight across the yard\\nrather on the paved track area “because of his aggression.”\\nOn cross-examination, when asked why she took him down\\non the yard, Officer Blauser responded it was because\\nClaiborne presented a threat to her and her partner. In total,\\nOfficer Blauser referred to Claiborne as “aggressive” or a\\nthreat at least forty times during her testimony. In closing\\n\\f CLAIBORNE V. BLAUSER 25\\n\\narguments, the defense repeated Blauser’s testimony and\\nemphasized that “she perceived a threat” and “took the most\\nreasonable action under the circumstances.” In short, if the\\njury believed that Claiborne posed a threat to the officers,\\nthen defendants prevailed because their use of force was\\njustified.\\n\\n Not only did Claiborne’s case hinge on his alleged\\ndangerousness, the trial pitted his credibility against the\\ndefendant officers’ credibility, and raised the issue of flight\\nrisk as defendants claimed that he was trying to escape from\\nthem during their escort in 2010. Claiborne’s visible\\nshackling was a “constant reminder” to the jury of his\\ncondition as a potentially violent and unreliable person who\\nneeded to be restrained. Tyars, 709 F.2d at 1284 (quoting\\nEstelle, 425 U.S. at 504); see also Lemons, 985 F.2d at 357.\\nWe have also recognized that physical restraints “may\\nconfuse and embarrass the [individual], thereby impairing\\nhis mental faculties[,] and they may cause him pain.”\\nDuckett, 67 F.3d at 748 (quoting Spain v. Rushen, 883 F.2d\\n712, 720–21 (9th Cir. 1989) (collecting cases) (internal\\nquotation marks omitted)). No doubt, shackles likely\\nimpaired Claiborne as a pro se litigant in the presentation of\\nhis case to a jury over three days. Id. Although the district\\ncourt noted, ten months after the trial, that Claiborne’s hands\\nwere not restrained, this is contradicted by Officer Blauser’s\\ncontemporaneous observation at trial that Claiborne was\\n“wearing [waist chains] right now” and that these waist\\nchains were “big and bulky.”\\n\\n Defendants argue there was no prejudice because the\\njury already knew Claiborne was a convicted felon. But the\\nSupreme Court and other circuit courts have uniformly\\nrejected the idea that a conviction alone could justify the use\\nof shackles because restraint “almost inevitably affects\\n\\f26 CLAIBORNE V. BLAUSER\\n\\nadversely the jury’s perception of the character of the\\ndefendant.” Deck, 544 U.S. at 633; see also Duckett, 67 F.3d\\nat 749 (noting that a defendant’s status as a convicted felon,\\n“[s]tanding alone, . . . is not a sufficient reason to impose\\nphysical restraints”); Lemons, 985 F.2d at 357. We are also\\nnot convinced by defendants’ insistence that the evidence\\npresented at trial overwhelmingly supported the jury’s\\nverdict. Crucially, the district court denied defendants’\\nmotion for judgment as a matter of law at the close of\\nevidence, demonstrating that a reasonable jury had a legally\\nsufficient evidentiary basis to find in Claiborne’s favor. See\\nFed. R. Civ. P. 50(a).\\n\\n Because of the inherent nature of visible shackles and its\\ninterplay with the heart of Claiborne’s excessive force claim,\\nwe conclude that Claiborne was prejudiced by the erroneous\\nuse of shackling in the absence of any showing of a\\ncompelling need for such restraints.\\n\\n 4.\\n\\n The final prong requires showing that “the district\\ncourt’s errors are so grave as to seriously impair the fairness,\\nintegrity, or public reputation of judicial proceedings.”\\nHoard, 904 F.3d at 791 (internal quotation marks and\\nalterations omitted). This is “undoubtedly the hardest\\n[prong] to meet.” Id. Because of the fundamental nature of\\nthe error, however, we find that this last prong has also been\\nmet.\\n\\n Shackling “must be limited to cases urgently demanding\\nthat action,” and because there was no showing of necessity\\nin Claiborne’s case, he was unduly prejudiced in violation of\\nhis due process right to a fair trial. Tyars, 709 F.2d at 1284;\\nsee also Maus v. Baker, 747 F.3d 926, 927 (7th Cir. 2014)\\n(“The sight of a shackled litigant is apt to make jurors think\\n\\f CLAIBORNE V. BLAUSER 27\\n\\nthey’re dealing with a mad dog . . .”). Just as an erroneous\\njury instruction could “place[] a heavy thumb on the scale in\\nfavor of the [d]efendants,” the erroneous use of shackling\\nhere also constituted a “grave injustice” in depriving\\nClaiborne of “a meaningful and fair opportunity to seek\\nredress for alleged violations of his constitutional right to be\\nfree from cruel and unusual punishment.” Hoard, 904 F.3d\\nat 792.\\n\\n We also reiterate that shackling not only prejudices the\\ninmate litigant in a case like this, it presents an affront to the\\ndignity of the courtroom. Deck, 544 U.S. at 631 (citing\\nAllen, 397 U.S. at 344). The Supreme Court has stressed that\\n“[t]he routine use of shackles in the presence of juries would\\nundermine [the] symbolic yet concrete objective[]” of\\n“maintain[ing] a judicial process that is a dignified process.”\\nId. Hence, shackling may only be used when there is an\\n“individualized security determination[]” that “take[s]\\naccount of the circumstances of the particular case.” Id.\\nat 632. That individualized determination did not take place\\nhere and the district court’s blanket assertion that it would\\nhave ordered Claiborne shackled solely based on his status\\nas a convicted felon was plainly erroneous. We therefore\\nexercise our discretion on plain error review to reverse the\\ndistrict court’s denial of the Rule 59(a) motion and remand\\nfor a new trial on Claiborne’s claims against Officers\\nBlauser and Martin.\\n\\n IV.\\n\\n Claiborne was denied a fair trial when he was visibly\\nshackled before a jury that had to decide his dangerousness\\nas it determined whether the jury believed his or defendants’\\nversion of the events that underlie his Section 1983 claims.\\nThere was no individualized determination of the security\\nneed for such restraints, yet this mobility impaired plaintiff\\n\\f28 CLAIBORNE V. BLAUSER\\n\\nwas presented to the jury in shackles, corroborating the\\ndefense’s position that Claiborne was an insubordinate,\\naggressive inmate and thereby undermining his case. On\\nremand, the district court has discretion to impose shackling\\nduring the new trial, but it may only do so after a full hearing\\nat which officers show a compelling need for security and\\nthe court considers any less restrictive alternatives. 7 See\\n\\n\\n\\n 7\\n Because we remand for a new trial on the shackling claim, we do\\nnot address the merits of Claiborne’s evidentiary arguments that the\\ndistrict court erred in barring his testimony about the ADA, preventing\\nhim from introducing or testifying about his medical records, and\\ndenying his request for a medical expert. The district court may revisit\\nthese issues if Claiborne raises them again at the new trial.\\n\\n We note, however, that the district court appears to misstate the law\\nwhen it denied Claiborne’s request for a medical expert under Federal\\nRule of Evidence 706(a). Although the district court correctly\\nrecognized that Rule 706(a) provides discretion to appoint a neutral\\nexpert witness, see McKinney v. Anderson, 924 F.2d 1500, 1511 (9th Cir.\\n1991), vacated on other grounds sub nom. Helling v. McKinney, 502\\nU.S. 903 (1991), judgment reinstated, 959 F.2d 853 (9th Cir. 1992),\\naff’d, 509 U.S. 25 (1993), the district court seemed to categorically limit\\nthe relevance of a medical expert to testifying about a plaintiff’s current\\ncondition. Yet courts have regularly considered requests for and\\nappointed experts to review medical records and testify about prior\\nmedical needs and treatment in deliberate indifference cases. See Gorton\\nv. Todd, 793 F.Supp.2d 1171, 1179–81 (E.D. Cal. 2011) (collecting\\ncases). Moreover, a medical expert can help with factfinding in\\nexcessive force claims because “the extent of injury suffered by an\\ninmate is one factor that may suggest ‘whether the [defendant’s] use of\\nforce could plausibly have been thought necessary’ in a particular\\nsituation.” Hudson v. McMillian, 503 U.S. 1, 7 (1992). If Claiborne\\nrenews his request for appointment of a neutral medical expert on retrial,\\nthe district court should weigh these considerations in exercising her\\ndiscretion. See McKinney, 924 F.2d at 1511; see also Gorton,\\n793 F.Supp.2d at 1185–86.\\n\\f CLAIBORNE V. BLAUSER 29\\n\\nDuckett, 67 F.3d at 748; Tyars, 709 F.2d at 1285; see also\\nDavidson, 44 F.3d at 1126.\\n\\n For the foregoing reasons, we reverse the district court’s\\ndenial of Claiborne’s Rule 59(a) motion and remand for a\\nnew trial.\\n\\n REVERSED AND REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4068,"cells":{"id":{"kind":"number","value":9024987,"string":"9,024,987"},"name":{"kind":"string","value":"Louis TAYLOR, a single man v. COUNTY OF PIMA, a body politic City of Tucson, a body politic"},"name_abbreviation":{"kind":"string","value":"Taylor v. Cnty. of Pima"},"decision_date":{"kind":"string","value":"2019-08-14"},"docket_number":{"kind":"string","value":"No. 17-16980"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Deliberations, Hurwitz, Vote\", \"parties\": \"\", \"opinions\": [{\"author\": \"\", \"type\": \"020lead\", \"text\": \"\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/9018241/\", \"author_raw\": \"\"}]}"},"judge1":{"kind":"string","value":"HURWITZ"},"judge2":{"kind":"string","value":""},"judge3":{"kind":"string","value":"HURWITZ"},"exactly_three_caps":{"kind":"number","value":0,"string":"0"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/9024987/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":0,"string":"0"},"020lead_count":{"kind":"number","value":1,"string":"1"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Louis TAYLOR, a single man v. COUNTY OF PIMA, a body politic City of Tucson, a body politic\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-14\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\n\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Louis TAYLOR, a single man v. COUNTY OF PIMA, a body politic City of Tucson, a body politic\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-14\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\n\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4069,"cells":{"id":{"kind":"number","value":4648689,"string":"4,648,689"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Jenny Flores v. William Barr"},"decision_date":{"kind":"string","value":"2019-08-15"},"docket_number":{"kind":"string","value":"17-56297"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: A. Wallace Tashima, William A. Fletcher, and Marsha S. Berzon, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"BERZON, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nJENNY LISETTE FLORES, et al., No. 17-56297\\n Plaintiffs-Appellees,\\n D.C. No.\\n v. 2:85-cv-04544-\\n DMG-AGR\\nWILLIAM P. BARR, Attorney General;\\nKEVIN K. MCALEENAN, Acting\\nSecretary of Homeland Security; OPINION\\nU.S. DEPARTMENT OF HOMELAND\\nSECURITY; U.S. IMMIGRATION AND\\nCUSTOMS ENFORCEMENT; U.S.\\nCUSTOMS AND BORDER PROTECTION,\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Dolly M. Gee, District Judge, Presiding\\n\\n Argued and Submitted June 18, 2019\\n San Francisco, California\\n\\n Filed August 15, 2019\\n\\n Before: A. Wallace Tashima, William A. Fletcher,\\n and Marsha S. Berzon, Circuit Judges.\\n\\n Opinion by Judge Berzon\\n\\f2 FLORES V. BARR\\n\\n SUMMARY *\\n\\n\\n Immigration / Juvenile Detention\\n\\n The panel dismissed for lack of jurisdiction an appeal\\nbrought by the Department of Homeland Security and its\\ncomponent agencies of the district court’s June 2017 order\\ngranting in part the motion of a plaintiff class to enforce a\\n1997 Settlement Agreement with the government which set\\na nationwide policy for the detention, release, and treatment\\nof minors detained in Immigration and Naturalization\\nService custody.\\n\\n In 1997, the United States entered into a settlement with\\na class of minors subject to detention by U.S. immigration\\nauthorities. The Settlement Agreement, incorporated into a\\nconsent decree, requires immigration agencies to hold such\\nminors in their custody “in facilities that are safe and\\nsanitary.” The Agreement also requires the government to\\ntreat these “minors in its custody with dignity, respect, and\\nspecial concern for their particular vulnerability as minors.”\\n\\n Plaintiffs filed a motion in district court to enforce the\\nAgreement. The district court found that the government was\\nviolating the Agreement by detaining minors in unsanitary\\nand unsafe conditions at Border Patrol stations. These\\nfindings were based on evidence that minors in U.S.\\nCustoms and Border Protection custody were held in\\nconditions that deprived them of sleep and did not provide\\nadequate access to food, clean water, and basic hygiene\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f FLORES V. BARR 3\\n\\nitems. The court also found that the government was\\nviolating the Agreement by failing to consider minors for\\nrelease as specified in the Agreement and by detaining\\nminors in detention facilities not licensed for the care of\\nminors. The district court ordered “enforced” various\\nparagraphs of the Agreement and also directed the\\ngovernment to appoint an internal “Juvenile Coordinator,”\\nas contemplated by the Agreement, to monitor the\\ngovernment’s compliance with the Agreement and report to\\nthe court.\\n\\n The parties agreed that this court has jurisdiction over the\\nappeal of this post-judgment order only if it modified the\\nAgreement.\\n\\n The government argued that the district court’s order did\\nmodify the Agreement by requiring the government to\\nprovide specific hygiene items and adequate sleeping\\naccommodations not explicitly listed in the text of the\\nAgreement. The panel held that the district court’s order did\\nnot modify the Agreement, but instead interpreted the\\nAgreement’s requirement that minors be held in “safe and\\nsanitary” conditions “consistent with the [government’s]\\nconcern for the particular vulnerability of minors.”\\n\\n The government also argued that the district court\\nmodified the Agreement by concluding that it requires the\\ngovernment to consider releasing class members subject to\\nexpedited removal. The government contended that this\\ninterpretation of the Agreement was inconsistent with the\\nImmigration and Nationality Act and related regulations—\\nprimarily with the expedited removal provisions, which\\nprovide that noncitizens in expedited removal proceedings\\n“shall be detained for further consideration of the[ir]\\napplication[s] for asylum.” 8 U.S.C. § 1225(b)(1)(B)(ii).\\n\\f4 FLORES V. BARR\\n\\nThe panel held that, rather than modifying the Agreement,\\nthe district court appropriately interpreted it as consistent\\nwith both the INA and this court’s prior interpretation of the\\nAgreement.\\n\\n Regarding the government’s argument that the district\\ncourt erred in concluding that the Agreement prohibits the\\ngovernment from detaining minors in secure, unlicensed\\nfamily detention centers, the panel noted that the district\\ncourt addressed this issue directly in its earlier July 2015\\norder, and that although the government appealed that order,\\nit did not on appeal challenge the district court’s holding on\\nthis issue. The panel concluded that this issue belatedly\\nraised in this appeal was not properly before the court.\\n\\n Because the panel concluded that the district court’s\\norder did not modify the Agreement, it dismissed the appeal.\\n\\n\\n COUNSEL\\n\\nSarah Fabian (argued), Senior Litigation Counsel; William\\nC. Silvis, Assistant Director, District Court Section; William\\nC. Peachey, Director, District Court Section; Office of\\nImmigration Litigation, Civil Division, United States\\nDepartment of Justice, Washington, D.C.; for Defendants-\\nAppellants.\\n\\nPeter Anthony Schey (argued) and Carlos R. Holguin,\\nCenter for Human Rights and Constitutional Law, Los\\nAngeles, California; Elena Garcia, Orrick Herrington &\\nSutcliffe LLP, Los Angeles, California; Michael S. Sorgen,\\nLa Raza Centro Legal Inc., San Francisco, California;\\nJennifer Kelleher Cloyd, Katherine H. Manning, and\\nAnnette Kirkham, The Law Foundation of Silicon Valley\\n\\f FLORES V. BARR 5\\n\\nLegal Advocates for Children and Youth Public Interest Law\\nFirm, San Jose, California; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nBERZON, Circuit Judge:\\n\\n In 1997, the United States entered into a settlement with\\na class of minors subject to detention by U.S. immigration\\nauthorities (“Plaintiffs”). The settlement agreement,\\nincorporated into a consent decree, requires immigration\\nagencies to hold such minors in their custody “in facilities\\nthat are safe and sanitary.” Flores Agreement (“the\\nAgreement”) at ¶ 12A. The Agreement also requires the\\ngovernment to treat these “minors in its custody with dignity,\\nrespect, and special concern for their particular vulnerability\\nas minors.” Id. at ¶ 11.\\n\\n This appeal began as a motion by the Plaintiffs to enforce\\nthe Agreement. The district court found that the government\\nwas violating the Agreement by detaining minors in\\nunsanitary and unsafe conditions at Border Patrol stations.\\nThese findings were based on evidence that minors in U.S.\\nCustoms and Border Protection custody were held in\\nconditions that deprived them of sleep and did not provide\\nadequate access to food, clean water, and basic hygiene\\nitems. The court also found that the government was\\nviolating the Agreement by failing to consider minors for\\nrelease as specified in the Agreement and by detaining\\nminors in detention facilities not licensed for the care of\\nminors. The district court ordered “enforced” various\\nparagraphs of the Agreement and also directed the\\ngovernment to appoint an internal “Juvenile Coordinator,”\\nas contemplated by the Agreement, to monitor the\\n\\f6 FLORES V. BARR\\n\\ngovernment’s compliance with the Agreement and report to\\nthe court.\\n\\n The parties agree that this court has jurisdiction over the\\nappeal of this post-judgment order only if it modified the\\nAgreement. The government argues that the district court’s\\norder did modify the Agreement by requiring the\\ngovernment to provide specific hygiene items and adequate\\nsleeping accommodations not explicitly listed in the text of\\nthe Agreement. We disagree. The district court’s order does\\nnot modify the Agreement. Instead, it interprets the\\nAgreement’s requirement that minors be held in “safe and\\nsanitary” conditions “consistent with the [government’s]\\nconcern for the particular vulnerability of minors.” See\\nAgreement at ¶ 12A. The government also argues that the\\ndistrict court’s order modifies the Agreement in other\\nrespects, but those arguments likewise lack merit. As the\\ndistrict court’s order did not modify the Agreement we\\ndismiss the appeal.\\n\\n I\\n\\n A\\n\\n This case stems from a 1985 lawsuit filed on behalf of a\\nclass of minors detained by U.S. immigration authorities.\\nAfter considerable litigation, the parties negotiated the\\nAgreement; it was entered by the district court as a consent\\ndecree in January 1997. The Agreement remains in effect\\ntoday. 1 The Agreement “sets out nationwide policy for the\\n 1\\n The Agreement included a specified termination date, but in 2001\\nthe parties stipulated to extend the Agreement until “45 days following\\ndefendants’ publication of final regulations implementing this\\nAgreement.” The government has issued proposed regulations but the\\nregulations have not been made final. See Apprehension, Processing,\\n\\f FLORES V. BARR 7\\n\\ndetention, release, and treatment of minors in the custody of\\nthe INS.” Id. at ¶ 9. 2 It requires the government to “place\\neach detained minor in the least restrictive setting\\nappropriate to the minor’s age and special needs, provided\\nthat such setting is consistent with its interests to ensure the\\nminor’s timely appearance before the INS and the\\nimmigration courts and to protect the minor’s well-being and\\nthat of others.” Id. at ¶ 11. The Agreement’s provisions\\n“create[] a presumption in favor of releasing minors and\\nrequire[] placement of those not released in licensed, non-\\nsecure facilities that meet certain standards.” Flores v.\\nLynch, 828 F.3d 898, 901 (9th Cir. 2016).\\n\\n Under paragraph 12A of the Agreement, “[f]ollowing\\narrest, the INS shall hold minors in facilities that are safe and\\nsanitary and that are consistent with the INS’s concern for\\nthe particular vulnerability of minors.” The Agreement\\ncontinues: “Facilities will provide access to toilets and sinks,\\ndrinking water and food as appropriate, medical assistance if\\nthe minor is in need of emergency services, adequate\\ntemperature control and ventilation, adequate supervision to\\n\\n\\n\\n\\nCare, and Custody of Alien Minors and Unaccompanied Alien Children,\\n83 Fed. Reg. 45,486 (proposed Sept. 7, 2018) (to be codified at 8 C.F.R.\\n§§ 212, 236; 45 C.F.R. § 410); see also Apprehension, Processing, Care,\\nand Custody of Alien Minors, Regulations.gov, https://www.regulations\\n.gov/docket?D=ICEB-2018-0002 (last visited Aug. 6, 2019).\\n\\n 2\\n Although the Agreement’s terms refer to “INS,” the Immigration\\nand Naturalization Service’s obligations under the Agreement now apply\\nto the Department of Homeland Security and the Department of Health\\nand Human Services. See Flores v. Sessions, 862 F.3d 863, 870 (9th Cir.\\n2017).\\n\\f8 FLORES V. BARR\\n\\nprotect minors from others, and contact with family\\nmembers who were arrested with the minor.”\\n\\n Within a few days of initial detention—three days if a\\nsuitable detention facility “is located and has space\\navailable” or five days “in all other cases”—the government\\nordinarily must choose between two options for placement\\nof minors. 3 The first, and preferable, option, discussed in\\nparagraph 14 of the Agreement, is releasing the minor to a\\nparent, legal guardian, adult relative, or another “capable and\\nwilling” designated adult. Alternatively, under paragraph 19,\\nthe minor may be placed in a facility “licensed by an\\nappropriate State agency to provide residential, group, or\\nfoster care services for dependent children.” Licensed\\nfacilities must be “non-secure as required under state law.”\\n\\n Finally, paragraph 28A provides that “[a]n INS Juvenile\\nCoordinator . . . shall monitor compliance with the terms of\\nthis Agreement and shall maintain an up-to-date record of all\\nminors who are placed in proceedings and remain in INS\\ncustody for longer than 72 hours.”\\n\\n\\n\\n\\n 3\\n Paragraph 21 of the Agreement provides a third option—\\nplacement in a secure juvenile detention facility—in limited\\ncircumstances, such as where the minor has been charged with a crime.\\nAlthough the Agreement contemplates secure detention for minors that\\nare “an escape-risk,” the William Wilberforce Trafficking Victims\\nProtection Reauthorization Act of 2008 (“TVPRA”) limits the use of\\nsecure custody, stating that “[a] child shall not be placed in a secure\\nfacility absent a determination that the child poses a danger to self or\\nothers or has been charged with having committed a criminal offense.”\\n8 U.S.C. § 1232(c)(2)(A); see also Flores v. Sessions, 862 F.3d at 881\\nn.19.\\n\\f FLORES V. BARR 9\\n\\n B\\n\\n In May 2016, the plaintiffs filed a motion to enforce the\\nAgreement. The plaintiffs alleged that despite earlier such\\nmotions and resulting orders enforcing the Agreement, the\\ngovernment continued to violate it by detaining class\\nmembers in unsafe and unsanitary conditions at Border\\nPatrol stations and by detaining minors in secure, unlicensed\\nfacilities. The plaintiffs requested that the district court\\naddress these violations by appointing an independent\\nmonitor.\\n\\n In June 2017, after an evidentiary hearing, the district\\ncourt granted in part plaintiffs’ motion to enforce. In its order\\n(“the Order”), the court found that the government was\\nviolating the Agreement’s express requirements to provide\\nadequate access to appropriate food and water and “adequate\\ntemperature controls at a reasonable and comfortable range.”\\nThe court further found that although the Agreement “makes\\nno mention of the words ‘soap,’ ‘towels,’ ‘showers,’ ‘dry\\nclothing,’ or ‘toothbrushes,’ . . . these hygiene products fall\\nwithin the rubric of the Agreement’s language requiring\\n‘safe and sanitary’ conditions.” Certain Border Patrol\\nstations, the district court found, were violating paragraph\\n12A of the Agreement by failing to provide such sanitary\\nnecessities.\\n\\n The district court also determined that although “the\\nword ‘sleep’ does not appear in the Agreement, . . . whether\\nDefendants have set up conditions that allow class members\\nto sleep in the [Border Patrol] facilities is relevant to the\\nissue of whether they have acted in a manner that is\\nconsistent with ‘the INS’s concern for the particular\\nvulnerability of minors’ as well as the Agreement’s ‘safe and\\nsanitary’ requirement.” Citing evidence that many minors in\\nBorder Patrol custody are forced to sleep on concrete floors,\\n\\f10 FLORES V. BARR\\n\\nwith no bedding aside from pieces of thin polyester foil, and\\nare subjected to cold temperatures, serious overcrowding,\\nand constant lighting, the district court found that the\\ngovernment was violating the Agreement at certain Border\\nPatrol stations by holding children in facilities that deprived\\nthem of adequate sleep.\\n\\n The court also found that the government was failing to\\nmake and record ongoing efforts aimed at releasing or\\nplacing class members in violation of paragraph 14 of the\\nAgreement, and was detaining class members in secure,\\nunlicensed facilities in violation of paragraph 19.\\n\\n The district court declined, however, to appoint an\\nindependent monitor as the plaintiffs had requested. Instead,\\nit directed the government to identify an internal “Juvenile\\nCoordinator” in accordance with paragraph 28A of the\\nAgreement. The court instructed that the Juvenile\\nCoordinator, once appointed, “will monitor compliance with\\nthose terms of the Flores Agreement, which this Court has\\nfound must be enforced and shall report directly to the Court\\nregarding the status of Defendants’ compliance.” 4\\n\\n II\\n\\n A\\n\\n This court has appellate jurisdiction over interlocutory\\ndistrict court orders “granting, continuing, modifying,\\nrefusing or dissolving injunctions, or refusing to dissolve or\\n\\n 4\\n A Special Master has since been appointed. See Order Appointing\\nSpecial Master/Independent Monitor, Flores v. Barr, No. 2:85-cv-4544\\n(Order) (C.D. Cal. Oct. 5, 2018) (ECF No. 494). Neither the appointment\\nof the Special Master nor the Special Master’s enforcement activities is\\nthe subject of this appeal.\\n\\f FLORES V. BARR 11\\n\\nmodify injunctions.” 28 U.S.C. § 1292(a)(1). The parties\\nagree that the district court’s Order did not grant, continue,\\nrefuse, dissolve, or refuse to dissolve an injunction. The\\nonly possible basis for appellate jurisdiction therefore\\ndepends on whether the Order “modif[ied]” the Agreement.\\nSee Thompson v. Enomoto, 815 F.2d 1323, 1327 (9th Cir.\\n1987) (appointment of a special master cannot be appealed\\nunder § 1292(a)(1) unless “the interlocutory order\\nappointing a special master ‘modifies’ the consent decree”);\\nsee also Bogard v. Wright, 159 F.3d 1060, 1064–65 (7th Cir.\\n1998) (dismissing for lack of jurisdiction an appeal of an\\norder that enforced but did not modify a consent decree, and\\ndiscussing the “appellate remedy” available to each party\\nunder section 1292(a)(1) with respect to orders after a final\\njudgement).\\n\\n The text of the district court’s opinion in this case\\ninterpreted the existing Agreement; its operative provisions,\\nwhich appear at the end of the Order, granted enforcement\\nof various provisions of the Agreement. The conclusion of\\nthe Order reads, for example, “Plaintiffs’ motion to enforce\\nParagraph 12A of the Agreement on the issue of unsanitary\\nconditions is GRANTED . . .” The Order’s operative\\nprovisions do not require the government to take any specific\\naction other than to propose a Juvenile Coordinator for\\nappointment. 5\\n\\n The government does not challenge the appointment of\\na Juvenile Coordinator, as that appointment is expressly\\n 5\\n In contrast, the district court’s July 2015 order, over which this\\ncourt exercised appellate jurisdiction, see Flores v. Lynch, 828 F.3d\\nat 905, did direct the government to take specific actions, see id. at 908\\n(“To comply with Paragraph 14A of the Agreement and as contemplated\\nin Paragraph 15, a class member’s accompanying parent shall be released\\nwith the class member . . . .”).\\n\\f12 FLORES V. BARR\\n\\nprovided for in the Agreement. Rather, the government\\nargues that we have jurisdiction to review other parts of the\\ndistrict court’s Order because they modified the Flores\\nAgreement.\\n\\n Specifically, the government contends that, by\\ninterpreting paragraph 12A in the body of its opinion to\\nrequire that Border Patrol stations provide the most basic\\nhuman necessities—accommodations that allow for\\nadequate sleep, essential hygiene items, and adequate, clean\\nfood and water—the district court modified the Agreement’s\\nrequirement that minors be held in “safe and sanitary”\\nconditions that comport with the “special concern for the\\nparticular vulnerability of minors.” We emphatically\\ndisagree.\\n\\n The government first suggests that the key phrases in\\nparagraph 12A—“safe and sanitary” and “special concern\\nfor the particular vulnerability of minors”—add nothing to\\nthe enumerated specific requirements found in the next\\nsentence of the Agreement (requiring “access to toilets and\\nsinks, drinking water and food as appropriate,” and so on).\\nThe government’s brief maintains that as the enumerated\\nconditions said nothing about, for example, allowing the\\nchildren in government custody to sleep or to wash\\nthemselves with soap, reading the “safe and sanitary”\\nrequirement to cover those requirements is a modification of\\nthe Agreement rather than an interpretation of it.\\n\\n That cramped understanding of paragraph 12A is\\nuntenable. Construing the Agreement as requiring only the\\nparticular conditions specifically enumerated renders both\\nthe “safe and sanitary” and the “particular vulnerability of\\nminors” phrases wholly superfluous. We cannot accept that\\nthe parties to the Agreement included gratuitous standards\\nthat have no practical impact. “Courts interpreting the\\n\\f FLORES V. BARR 13\\n\\nlanguage of contracts ‘should give effect to every provision,’\\nand ‘an interpretation which renders part of the instrument\\nto be surplusage should be avoided.’” United States v. 1.377\\nAcres of Land, 352 F.3d 1259, 1265 (9th Cir. 2003) (quoting\\nAppalachian Ins. Co. v. McDonnell Douglas Corp., 214 Cal.\\nApp. 3d 1, 12 (Ct. App. 1989)); see also Restatement\\n(Second) of Contracts § 203 (Am. Law Inst. 1981). We\\nconclude that paragraph 12A’s provisions that facilities be\\n“safe and sanitary and . . . consistent with the INS’s concern\\nfor the particular vulnerability of minors” do have\\nindependent force and can be interpreted and enforced\\nwithout thereby modifying the Agreement.\\n\\n The government also argues that the phrase “safe and\\nsanitary” is so vague that either it cannot be enforced, see\\nOral Argument at 33:57–34:09, Flores v. Barr, No. 17-\\n56297 (9th Cir. Jun. 18, 2019), https://www.ca9.uscourts.g\\nov/media/view_video.php?pk_vid=0000015907, or it leaves\\n“the specifics of compliance [with paragraph 12A] up to” the\\ngovernment. Not so.\\n\\n The district court’s interpretation of the Agreement is\\nconsistent with the ordinary meaning of the language of\\nparagraph 12A, which does provide a standard sufficiently\\nclear to be enforced. The court found, among other things,\\nthat minors (1) were “not receiving hot, edible, or a sufficient\\nnumber of meals during a given day,” (2) “had no adequate\\naccess to clean drinking water,” (3) experienced “unsanitary\\nconditions with respect to the holding cells and bathroom\\nfacilities,” (4) lacked “access to clean bedding, and access to\\nhygiene products (i.e., toothbrushes, soap, towels),” and\\n(5) endured “sleep deprivation” as a result of “cold\\ntemperatures, overcrowding, lack of proper bedding (i.e.,\\nblankets, mats), [and] constant lighting.” After so finding,\\nthe district court concluded that these conditions fall short of\\n\\f14 FLORES V. BARR\\n\\nparagraph 12A’s requirement that facilities be “safe and\\nsanitary,” especially given “the particular vulnerability of\\nminors.” Those determinations reflect a commonsense\\nunderstanding of what the quoted language requires.\\nAssuring that children eat enough edible food, drink clean\\nwater, are housed in hygienic facilities with sanitary\\nbathrooms, have soap and toothpaste, and are not sleep-\\ndeprived are without doubt essential to the children’s\\nsafety. 6 The district court properly construed the Agreement\\nas requiring such conditions rather than allowing the\\ngovernment to decide whether to provide them.\\n\\n Moreover, contrary to the government’s assertions, the\\ndistrict court did not incorporate into the Agreement a\\nparticular set of standards, Customs and Border Protection’s\\n“National Standards on Transport, Escort, Detention, and\\nSearch” (or “TEDS”), with respect to food during detention.\\nWe doubt that the TEDS requirements—that minors “be\\noffered a snack upon arrival and a meal at least every six\\nhours thereafter,” have food that is “in edible condition (not\\nfrozen, expired, or spoiled),” and “have regular access to\\nsnacks, milk, and juice,”—extend beyond what paragraph\\n12A requires. But in any event, in context, the district court\\nreferred to TEDS not to interpret the Agreement as\\nincorporating the TEDS standards specifically, but to\\nconfirm that the government’s inattention to ensuring that\\nchildren were being adequately fed was egregious, as the\\ngovernment was not even complying with its own standards.\\n\\n\\n\\n 6\\n We note that, as the district court properly understood, assuring\\n“safe and sanitary” conditions includes protecting children from\\ndeveloping short- or long-term illnesses as well as protecting them from\\naccidental or intentional injury.\\n\\f FLORES V. BARR 15\\n\\n In short, the district court’s explanation of its\\nenforcement of paragraph 12A regarding the conditions at\\nBorder Patrol stations concerned only requirements\\nunarguably within the terms of the Agreement. As a result,\\nthe portion of the court’s order enforcing paragraph 12A did\\nnot constitute an “[i]nterlocutory order[] . . . modifying [an]\\ninjunction[], or refusing to . . . modify [an] injunction[].”\\n28 U.S.C. § 1292(a)(1). We therefore lack jurisdiction over\\nthis claim. 7\\n\\n B\\n\\n The government next argues that the district court\\nmodified the Agreement by concluding that it requires the\\ngovernment to consider releasing class members subject to\\nexpedited removal. The government contends that this\\ninterpretation of the Agreement is inconsistent with the\\nImmigration and Nationality Act (INA) and related\\nregulations—primarily with the expedited removal\\nprovisions, which provide that noncitizens in expedited\\nremoval proceedings “shall be detained for further\\nconsideration of the[ir] application[s] for asylum.” 8 U.S.C.\\n§ 1225(b)(1)(B)(ii).\\n\\n The Agreement “creates a presumption in favor of\\nreleasing minors.” Flores v. Lynch, 828 F.3d at 901; accord\\nFlores v. Sessions, 862 F.3d at 866. That presumption is fully\\nconsistent with the Act’s expedited removal provisions.\\n\\n To begin, not all noncitizens eligible to be placed in\\nexpedited removal proceedings are in fact placed in such\\n\\n 7\\n As we lack jurisdiction, we do not reach the government’s\\nevidentiary objections regarding the record considered by the district\\ncourt.\\n\\f16 FLORES V. BARR\\n\\nproceedings. The government has discretion to place\\nnoncitizens in standard removal proceedings even if the\\nexpedited removal statute could be applied to them. See\\nMatter of E-R-M- & L-R-M-, 25 I. & N. Dec. 520, 521–22\\n(B.I.A. 2011). In other words, the government may place\\nminors into standard, nonexpedited removal proceedings\\nand thus comply with the Agreement by avoiding any\\nmandatory detention allegedly required for expedited\\nremoval.\\n\\n Further, expedited removal does not require mandatory\\ndetention for minors. The INA provides that, even for\\nnoncitizens in expedited removal, “the Attorney General\\nmay . . . in his discretion parole into the United States\\ntemporarily” any noncitizen applying for admission “under\\nsuch conditions as he may prescribe.” 8 U.S.C.\\n§ 1182(d)(5)(A). The government has promulgated two\\nregulations that pertain to parole into the United States of\\nnoncitizens in expedited removal proceedings. One provides\\nthat such noncitizens “shall be detained pending\\ndetermination and removal, except that parole of such alien\\n. . . may be permitted only when the Attorney General\\ndetermines, in the exercise of discretion, that parole is\\nrequired to meet a medical emergency or is necessary for a\\nlegitimate law enforcement objective.” 8 C.F.R.\\n§ 235.3(b)(2)(iii). A second provides that all noncitizens\\nsubject to removal—specifically including, by cross-\\nreference, those in expedited removal 8—may be paroled “on\\na case-by-case basis for ‘urgent humanitarian reasons’ or\\n‘significant public benefit,’ provided the [noncitizens]\\n\\n 8\\n Section 212.5(b) addresses the parole of aliens “who have been or\\nare detained in accordance with § 235.3(b) or (c) of this chapter.”\\n8 C.F.R. § 212.5(b). Section 235.3(b) provides for expedited removal.\\nId. § 235.3.\\n\\f FLORES V. BARR 17\\n\\npresent neither a security risk nor a risk of absconding.” Id.\\n§ 212.5(b). Among the groups eligible for parole under this\\nregulation are “[a]liens who are defined as juveniles in\\n§ 236.3(a) of this chapter,” who may be paroled under “the\\nguidelines set forth in § 236.3(a) of this chapter and\\nparagraphs (b)(3)(i) through (iii) of this section.” Id.\\n§ 212.5(b)(3).\\n\\n Both regulations expressly cover noncitizens in\\nexpedited removal proceedings; nothing in section 235.3(b)\\nnegates section 212.5(b). The more specific regulatory\\nprovision providing an exception for minors governs, not the\\ngeneral expedited removal provisions. See Karczewski v.\\nDCH Mission Valley LLC, 862 F.3d 1006, 1015–16 (9th Cir.\\n2017) (applying the interpretive canon that “the specific\\ngoverns over the general” to regulations).\\n\\n The upshot is that the government’s own regulations\\ncontemplate that minors in expedited removal proceedings\\nmay be considered for release, just as the Agreement\\nrequires. Rather than modifying the Agreement, the district\\ncourt appropriately interpreted it as consistent with both the\\nINA and our prior interpretation of the Agreement. We\\ntherefore lack jurisdiction over this claim. 9\\n\\n\\n\\n\\n 9\\n For the first time on appeal, the government contends that the\\ndistrict court’s conclusion that the government must consider whether to\\nrelease minors to potential custodians other than a parent or legal\\nguardian conflicts with a provision of the TVPRA, 8 U.S.C.\\n§ 1232(b)(3). We decline to address this statutory defense in this appeal,\\nas the government did not raise it before the district court. See Cold\\nMountain v. Garber, 375 F.3d 884, 891 (9th Cir. 2004).\\n\\f18 FLORES V. BARR\\n\\n C\\n\\n Finally, the government argues that the district court\\nerred in concluding that the Agreement prohibits the\\ngovernment from detaining minors in secure, unlicensed\\nfamily detention centers. The district court addressed this\\nissue directly in its July 2015 order. Although the\\ngovernment appealed that order, it did not on appeal\\nchallenge the district court’s holding on this issue. See\\nFlores v. Lynch, 828 F.3d at 901. “[A] party cannot offer up\\nsuccessively different legal or factual theories that could\\nhave been presented in a prior request for review.” Sec. Inv’r\\nProt. Corp. v. Vigman, 74 F.3d 932, 937 (9th Cir. 1996). The\\nissue belatedly raised in this appeal is not properly before us.\\n\\n III\\n\\n We dismiss the appeal for lack of jurisdiction.\\n\\n DISMISSED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4425942/\", \"author_raw\": \"BERZON, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"A WALLACE TASHIMA"},"judge2":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge3":{"kind":"string","value":"MARSHA S BERZON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4648689/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Jenny Flores v. William Barr\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-15\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nJENNY LISETTE FLORES, et al., No. 17-56297\n Plaintiffs-Appellees,\n D.C. No.\n v. 2:85-cv-04544-\n DMG-AGR\nWILLIAM P. BARR, Attorney General;\nKEVIN K. MCALEENAN, Acting\nSecretary of Homeland Security; OPINION\nU.S. DEPARTMENT OF HOMELAND\nSECURITY; U.S. IMMIGRATION AND\nCUSTOMS ENFORCEMENT; U.S.\nCUSTOMS AND BORDER PROTECTION,\n Defendants-Appellants.\n\n Appeal from the United States District Court\n for the Central District of California\n Dolly M. Gee, District Judge, Presiding\n\n Argued and Submitted June 18, 2019\n San Francisco, California\n\n Filed August 15, 2019\n\n Before: A. Wallace Tashima, William A. Fletcher,\n and Marsha S. Berzon, Circuit Judges.\n\n Opinion by Judge Berzon\n\f2 FLORES V. BARR\n\n SUMMARY *\n\n\n Immigration / Juvenile Detention\n\n The panel dismissed for lack of jurisdiction an appeal\nbrought by the Department of Homeland Security and its\ncomponent agencies of the district court’s June 2017 order\ngranting in part the motion of a plaintiff class to enforce a\n1997 Settlement Agreement with the government which set\na nationwide policy for the detention, release, and treatment\nof minors detained in Immigration and Naturalization\nService custody.\n\n In 1997, the United States entered into a settlement with\na class of minors subject to detention by U.S. immigration\nauthorities. The Settlement Agreement, incorporated into a\nconsent decree, requires immigration agencies to hold such\nminors in their custody “in facilities that are safe and\nsanitary.” The Agreement also requires the government to\ntreat these “minors in its custody with dignity, respect, and\nspecial concern for their particular vulnerability as minors.”\n\n Plaintiffs filed a motion in district court to enforce the\nAgreement. The district court found that the government was\nviolating the Agreement by detaining minors in unsanitary\nand unsafe conditions at Border Patrol stations. These\nfindings were based on evidence that minors in U.S.\nCustoms and Border Protection custody were held in\nconditions that deprived them of sleep and did not provide\nadequate access to food, clean water, and basic hygiene\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f FLORES V. BARR 3\n\nitems. The court also found that the government was\nviolating the Agreement by failing to consider minors for\nrelease as specified in the Agreement and by detaining\nminors in detention facilities not licensed for the care of\nminors. The district court ordered “enforced” various\nparagraphs of the Agreement and also directed the\ngovernment to appoint an internal “Juvenile Coordinator,”\nas contemplated by the Agreement, to monitor the\ngovernment’s compliance with the Agreement and report to\nthe court.\n\n The parties agreed that this court has jurisdiction over the\nappeal of this post-judgment order only if it modified the\nAgreement.\n\n The government argued that the district court’s order did\nmodify the Agreement by requiring the government to\nprovide specific hygiene items and adequate sleeping\naccommodations not explicitly listed in the text of the\nAgreement. The panel held that the district court’s order did\nnot modify the Agreement, but instead interpreted the\nAgreement’s requirement that minors be held in “safe and\nsanitary” conditions “consistent with the [government’s]\nconcern for the particular vulnerability of minors.”\n\n The government also argued that the district court\nmodified the Agreement by concluding that it requires the\ngovernment to consider releasing class members subject to\nexpedited removal. The government contended that this\ninterpretation of the Agreement was inconsistent with the\nImmigration and Nationality Act and related regulations—\nprimarily with the expedited removal provisions, which\nprovide that noncitizens in expedited removal proceedings\n“shall be detained for further consideration of the[ir]\napplication[s] for asylum.” 8 U.S.C. § 1225(b)(1)(B)(ii).\n\f4 FLORES V. BARR\n\nThe panel held that, rather than modifying the Agreement,\nthe district court appropriately interpreted it as consistent\nwith both the INA and this court’s prior interpretation of the\nAgreement.\n\n Regarding the government’s argument that the district\ncourt erred in concluding that the Agreement prohibits the\ngovernment from detaining minors in secure, unlicensed\nfamily detention centers, the panel noted that the district\ncourt addressed this issue directly in its earlier July 2015\norder, and that although the government appealed that order,\nit did not on appeal challenge the district court’s holding on\nthis issue. The panel concluded that this issue belatedly\nraised in this appeal was not properly before the court.\n\n Because the panel concluded that the district court’s\norder did not modify the Agreement, it dismissed the appeal.\n\n\n COUNSEL\n\nSarah Fabian (argued), Senior Litigation Counsel; William\nC. Silvis, Assistant Director, District Court Section; William\nC. Peachey, Director, District Court Section; Office of\nImmigration Litigation, Civil Division, United States\nDepartment of Justice, Washington, D.C.; for Defendants-\nAppellants.\n\nPeter Anthony Schey (argued) and Carlos R. Holguin,\nCenter for Human Rights and Constitutional Law, Los\nAngeles, California; Elena Garcia, Orrick Herrington &\nSutcliffe LLP, Los Angeles, California; Michael S. Sorgen,\nLa Raza Centro Legal Inc., San Francisco, California;\nJennifer Kelleher Cloyd, Katherine H. Manning, and\nAnnette Kirkham, The Law Foundation of Silicon Valley\n\f FLORES V. BARR 5\n\nLegal Advocates for Children and Youth Public Interest Law\nFirm, San Jose, California; for Plaintiff-Appellee.\n\n\n OPINION\n\nBERZON, Circuit Judge:\n\n In 1997, the United States entered into a settlement with\na class of minors subject to detention by U.S. immigration\nauthorities (“Plaintiffs”). The settlement agreement,\nincorporated into a consent decree, requires immigration\nagencies to hold such minors in their custody “in facilities\nthat are safe and sanitary.” Flores Agreement (“the\nAgreement”) at ¶ 12A. The Agreement also requires the\ngovernment to treat these “minors in its custody with dignity,\nrespect, and special concern for their particular vulnerability\nas minors.” Id. at ¶ 11.\n\n This appeal began as a motion by the Plaintiffs to enforce\nthe Agreement. The district court found that the government\nwas violating the Agreement by detaining minors in\nunsanitary and unsafe conditions at Border Patrol stations.\nThese findings were based on evidence that minors in U.S.\nCustoms and Border Protection custody were held in\nconditions that deprived them of sleep and did not provide\nadequate access to food, clean water, and basic hygiene\nitems. The court also found that the government was\nviolating the Agreement by failing to consider minors for\nrelease as specified in the Agreement and by detaining\nminors in detention facilities not licensed for the care of\nminors. The district court ordered “enforced” various\nparagraphs of the Agreement and also directed the\ngovernment to appoint an internal “Juvenile Coordinator,”\nas contemplated by the Agreement, to monitor the\n\f6 FLORES V. BARR\n\ngovernment’s compliance with the Agreement and report to\nthe court.\n\n The parties agree that this court has jurisdiction over the\nappeal of this post-judgment order only if it modified the\nAgreement. The government argues that the district court’s\norder did modify the Agreement by requiring the\ngovernment to provide specific hygiene items and adequate\nsleeping accommodations not explicitly listed in the text of\nthe Agreement. We disagree. The district court’s order does\nnot modify the Agreement. Instead, it interprets the\nAgreement’s requirement that minors be held in “safe and\nsanitary” conditions “consistent with the [government’s]\nconcern for the particular vulnerability of minors.” See\nAgreement at ¶ 12A. The government also argues that the\ndistrict court’s order modifies the Agreement in other\nrespects, but those arguments likewise lack merit. As the\ndistrict court’s order did not modify the Agreement we\ndismiss the appeal.\n\n I\n\n A\n\n This case stems from a 1985 lawsuit filed on behalf of a\nclass of minors detained by U.S. immigration authorities.\nAfter considerable litigation, the parties negotiated the\nAgreement; it was entered by the district court as a consent\ndecree in January 1997. The Agreement remains in effect\ntoday. 1 The Agreement “sets out nationwide policy for the\n 1\n The Agreement included a specified termination date, but in 2001\nthe parties stipulated to extend the Agreement until “45 days following\ndefendants’ publication of final regulations implementing this\nAgreement.” The government has issued proposed regulations but the\nregulations have not been made final. See Apprehension, Processing,\n\f FLORES V. BARR 7\n\ndetention, release, and treatment of minors in the custody of\nthe INS.” Id. at ¶ 9. 2 It requires the government to “place\neach detained minor in the least restrictive setting\nappropriate to the minor’s age and special needs, provided\nthat such setting is consistent with its interests to ensure the\nminor’s timely appearance before the INS and the\nimmigration courts and to protect the minor’s well-being and\nthat of others.” Id. at ¶ 11. The Agreement’s provisions\n“create[] a presumption in favor of releasing minors and\nrequire[] placement of those not released in licensed, non-\nsecure facilities that meet certain standards.” Flores v.\nLynch, 828 F.3d 898, 901 (9th Cir. 2016).\n\n Under paragraph 12A of the Agreement, “[f]ollowing\narrest, the INS shall hold minors in facilities that are safe and\nsanitary and that are consistent with the INS’s concern for\nthe particular vulnerability of minors.” The Agreement\ncontinues: “Facilities will provide access to toilets and sinks,\ndrinking water and food as appropriate, medical assistance if\nthe minor is in need of emergency services, adequate\ntemperature control and ventilation, adequate supervision to\n\n\n\n\nCare, and Custody of Alien Minors and Unaccompanied Alien Children,\n83 Fed. Reg. 45,486 (proposed Sept. 7, 2018) (to be codified at 8 C.F.R.\n§§ 212, 236; 45 C.F.R. § 410); see also Apprehension, Processing, Care,\nand Custody of Alien Minors, Regulations.gov, https://www.regulations\n.gov/docket?D=ICEB-2018-0002 (last visited Aug. 6, 2019).\n\n 2\n Although the Agreement’s terms refer to “INS,” the Immigration\nand Naturalization Service’s obligations under the Agreement now apply\nto the Department of Homeland Security and the Department of Health\nand Human Services. See Flores v. Sessions, 862 F.3d 863, 870 (9th Cir.\n2017).\n\f8 FLORES V. BARR\n\nprotect minors from others, and contact with family\nmembers who were arrested with the minor.”\n\n Within a few days of initial detention—three days if a\nsuitable detention facility “is located and has space\navailable” or five days “in all other cases”—the government\nordinarily must choose between two options for placement\nof minors. 3 The first, and preferable, option, discussed in\nparagraph 14 of the Agreement, is releasing the minor to a\nparent, legal guardian, adult relative, or another “capable and\nwilling” designated adult. Alternatively, under paragraph 19,\nthe minor may be placed in a facility “licensed by an\nappropriate State agency to provide residential, group, or\nfoster care services for dependent children.” Licensed\nfacilities must be “non-secure as required under state law.”\n\n Finally, paragraph 28A provides that “[a]n INS Juvenile\nCoordinator . . . shall monitor compliance with the terms of\nthis Agreement and shall maintain an up-to-date record of all\nminors who are placed in proceedings and remain in INS\ncustody for longer than 72 hours.”\n\n\n\n\n 3\n Paragraph 21 of the Agreement provides a third option—\nplacement in a secure juvenile detention facility—in limited\ncircumstances, such as where the minor has been charged with a crime.\nAlthough the Agreement contemplates secure detention for minors that\nare “an escape-risk,” the William Wilberforce Trafficking Victims\nProtection Reauthorization Act of 2008 (“TVPRA”) limits the use of\nsecure custody, stating that “[a] child shall not be placed in a secure\nfacility absent a determination that the child poses a danger to self or\nothers or has been charged with having committed a criminal offense.”\n8 U.S.C. § 1232(c)(2)(A); see also Flores v. Sessions, 862 F.3d at 881\nn.19.\n\f FLORES V. BARR 9\n\n B\n\n In May 2016, the plaintiffs filed a motion to enforce the\nAgreement. The plaintiffs alleged that despite earlier such\nmotions and resulting orders enforcing the Agreement, the\ngovernment continued to violate it by detaining class\nmembers in unsafe and unsanitary conditions at Border\nPatrol stations and by detaining minors in secure, unlicensed\nfacilities. The plaintiffs requested that the district court\naddress these violations by appointing an independent\nmonitor.\n\n In June 2017, after an evidentiary hearing, the district\ncourt granted in part plaintiffs’ motion to enforce. In its order\n(“the Order”), the court found that the government was\nviolating the Agreement’s express requirements to provide\nadequate access to appropriate food and water and “adequate\ntemperature controls at a reasonable and comfortable range.”\nThe court further found that although the Agreement “makes\nno mention of the words ‘soap,’ ‘towels,’ ‘showers,’ ‘dry\nclothing,’ or ‘toothbrushes,’ . . . these hygiene products fall\nwithin the rubric of the Agreement’s language requiring\n‘safe and sanitary’ conditions.” Certain Border Patrol\nstations, the district court found, were violating paragraph\n12A of the Agreement by failing to provide such sanitary\nnecessities.\n\n The district court also determined that although “the\nword ‘sleep’ does not appear in the Agreement, . . . whether\nDefendants have set up conditions that allow class members\nto sleep in the [Border Patrol] facilities is relevant to the\nissue of whether they have acted in a manner that is\nconsistent with ‘the INS’s concern for the particular\nvulnerability of minors’ as well as the Agreement’s ‘safe and\nsanitary’ requirement.” Citing evidence that many minors in\nBorder Patrol custody are forced to sleep on concrete floors,\n\f10 FLORES V. BARR\n\nwith no bedding aside from pieces of thin polyester foil, and\nare subjected to cold temperatures, serious overcrowding,\nand constant lighting, the district court found that the\ngovernment was violating the Agreement at certain Border\nPatrol stations by holding children in facilities that deprived\nthem of adequate sleep.\n\n The court also found that the government was failing to\nmake and record ongoing efforts aimed at releasing or\nplacing class members in violation of paragraph 14 of the\nAgreement, and was detaining class members in secure,\nunlicensed facilities in violation of paragraph 19.\n\n The district court declined, however, to appoint an\nindependent monitor as the plaintiffs had requested. Instead,\nit directed the government to identify an internal “Juvenile\nCoordinator” in accordance with paragraph 28A of the\nAgreement. The court instructed that the Juvenile\nCoordinator, once appointed, “will monitor compliance with\nthose terms of the Flores Agreement, which this Court has\nfound must be enforced and shall report directly to the Court\nregarding the status of Defendants’ compliance.” 4\n\n II\n\n A\n\n This court has appellate jurisdiction over interlocutory\ndistrict court orders “granting, continuing, modifying,\nrefusing or dissolving injunctions, or refusing to dissolve or\n\n 4\n A Special Master has since been appointed. See Order Appointing\nSpecial Master/Independent Monitor, Flores v. Barr, No. 2:85-cv-4544\n(Order) (C.D. Cal. Oct. 5, 2018) (ECF No. 494). Neither the appointment\nof the Special Master nor the Special Master’s enforcement activities is\nthe subject of this appeal.\n\f FLORES V. BARR 11\n\nmodify injunctions.” 28 U.S.C. § 1292(a)(1). The parties\nagree that the district court’s Order did not grant, continue,\nrefuse, dissolve, or refuse to dissolve an injunction. The\nonly possible basis for appellate jurisdiction therefore\ndepends on whether the Order “modif[ied]” the Agreement.\nSee Thompson v. Enomoto, 815 F.2d 1323, 1327 (9th Cir.\n1987) (appointment of a special master cannot be appealed\nunder § 1292(a)(1) unless “the interlocutory order\nappointing a special master ‘modifies’ the consent decree”);\nsee also Bogard v. Wright, 159 F.3d 1060, 1064–65 (7th Cir.\n1998) (dismissing for lack of jurisdiction an appeal of an\norder that enforced but did not modify a consent decree, and\ndiscussing the “appellate remedy” available to each party\nunder section 1292(a)(1) with respect to orders after a final\njudgement).\n\n The text of the district court’s opinion in this case\ninterpreted the existing Agreement; its operative provisions,\nwhich appear at the end of the Order, granted enforcement\nof various provisions of the Agreement. The conclusion of\nthe Order reads, for example, “Plaintiffs’ motion to enforce\nParagraph 12A of the Agreement on the issue of unsanitary\nconditions is GRANTED . . .” The Order’s operative\nprovisions do not require the government to take any specific\naction other than to propose a Juvenile Coordinator for\nappointment. 5\n\n The government does not challenge the appointment of\na Juvenile Coordinator, as that appointment is expressly\n 5\n In contrast, the district court’s July 2015 order, over which this\ncourt exercised appellate jurisdiction, see Flores v. Lynch, 828 F.3d\nat 905, did direct the government to take specific actions, see id. at 908\n(“To comply with Paragraph 14A of the Agreement and as contemplated\nin Paragraph 15, a class member’s accompanying parent shall be released\nwith the class member . . . .”).\n\f12 FLORES V. BARR\n\nprovided for in the Agreement. Rather, the government\nargues that we have jurisdiction to review other parts of the\ndistrict court’s Order because they modified the Flores\nAgreement.\n\n Specifically, the government contends that, by\ninterpreting paragraph 12A in the body of its opinion to\nrequire that Border Patrol stations provide the most basic\nhuman necessities—accommodations that allow for\nadequate sleep, essential hygiene items, and adequate, clean\nfood and water—the district court modified the Agreement’s\nrequirement that minors be held in “safe and sanitary”\nconditions that comport with the “special concern for the\nparticular vulnerability of minors.” We emphatically\ndisagree.\n\n The government first suggests that the key phrases in\nparagraph 12A—“safe and sanitary” and “special concern\nfor the particular vulnerability of minors”—add nothing to\nthe enumerated specific requirements found in the next\nsentence of the Agreement (requiring “access to toilets and\nsinks, drinking water and food as appropriate,” and so on).\nThe government’s brief maintains that as the enumerated\nconditions said nothing about, for example, allowing the\nchildren in government custody to sleep or to wash\nthemselves with soap, reading the “safe and sanitary”\nrequirement to cover those requirements is a modification of\nthe Agreement rather than an interpretation of it.\n\n That cramped understanding of paragraph 12A is\nuntenable. Construing the Agreement as requiring only the\nparticular conditions specifically enumerated renders both\nthe “safe and sanitary” and the “particular vulnerability of\nminors” phrases wholly superfluous. We cannot accept that\nthe parties to the Agreement included gratuitous standards\nthat have no practical impact. “Courts interpreting the\n\f FLORES V. BARR 13\n\nlanguage of contracts ‘should give effect to every provision,’\nand ‘an interpretation which renders part of the instrument\nto be surplusage should be avoided.’” United States v. 1.377\nAcres of Land, 352 F.3d 1259, 1265 (9th Cir. 2003) (quoting\nAppalachian Ins. Co. v. McDonnell Douglas Corp., 214 Cal.\nApp. 3d 1, 12 (Ct. App. 1989)); see also Restatement\n(Second) of Contracts § 203 (Am. Law Inst. 1981). We\nconclude that paragraph 12A’s provisions that facilities be\n“safe and sanitary and . . . consistent with the INS’s concern\nfor the particular vulnerability of minors” do have\nindependent force and can be interpreted and enforced\nwithout thereby modifying the Agreement.\n\n The government also argues that the phrase “safe and\nsanitary” is so vague that either it cannot be enforced, see\nOral Argument at 33:57–34:09, Flores v. Barr, No. 17-\n56297 (9th Cir. Jun. 18, 2019), https://www.ca9.uscourts.g\nov/media/view_video.php?pk_vid=0000015907, or it leaves\n“the specifics of compliance [with paragraph 12A] up to” the\ngovernment. Not so.\n\n The district court’s interpretation of the Agreement is\nconsistent with the ordinary meaning of the language of\nparagraph 12A, which does provide a standard sufficiently\nclear to be enforced. The court found, among other things,\nthat minors (1) were “not receiving hot, edible, or a sufficient\nnumber of meals during a given day,” (2) “had no adequate\naccess to clean drinking water,” (3) experienced “unsanitary\nconditions with respect to the holding cells and bathroom\nfacilities,” (4) lacked “access to clean bedding, and access to\nhygiene products (i.e., toothbrushes, soap, towels),” and\n(5) endured “sleep deprivation” as a result of “cold\ntemperatures, overcrowding, lack of proper bedding (i.e.,\nblankets, mats), [and] constant lighting.” After so finding,\nthe district court concluded that these conditions fall short of\n\f14 FLORES V. BARR\n\nparagraph 12A’s requirement that facilities be “safe and\nsanitary,” especially given “the particular vulnerability of\nminors.” Those determinations reflect a commonsense\nunderstanding of what the quoted language requires.\nAssuring that children eat enough edible food, drink clean\nwater, are housed in hygienic facilities with sanitary\nbathrooms, have soap and toothpaste, and are not sleep-\ndeprived are without doubt essential to the children’s\nsafety. 6 The district court properly construed the Agreement\nas requiring such conditions rather than allowing the\ngovernment to decide whether to provide them.\n\n Moreover, contrary to the government’s assertions, the\ndistrict court did not incorporate into the Agreement a\nparticular set of standards, Customs and Border Protection’s\n“National Standards on Transport, Escort, Detention, and\nSearch” (or “TEDS”), with respect to food during detention.\nWe doubt that the TEDS requirements—that minors “be\noffered a snack upon arrival and a meal at least every six\nhours thereafter,” have food that is “in edible condition (not\nfrozen, expired, or spoiled),” and “have regular access to\nsnacks, milk, and juice,”—extend beyond what paragraph\n12A requires. But in any event, in context, the district court\nreferred to TEDS not to interpret the Agreement as\nincorporating the TEDS standards specifically, but to\nconfirm that the government’s inattention to ensuring that\nchildren were being adequately fed was egregious, as the\ngovernment was not even complying with its own standards.\n\n\n\n 6\n We note that, as the district court properly understood, assuring\n“safe and sanitary” conditions includes protecting children from\ndeveloping short- or long-term illnesses as well as protecting them from\naccidental or intentional injury.\n\f FLORES V. BARR 15\n\n In short, the district court’s explanation of its\nenforcement of paragraph 12A regarding the conditions at\nBorder Patrol stations concerned only requirements\nunarguably within the terms of the Agreement. As a result,\nthe portion of the court’s order enforcing paragraph 12A did\nnot constitute an “[i]nterlocutory order[] . . . modifying [an]\ninjunction[], or refusing to . . . modify [an] injunction[].”\n28 U.S.C. § 1292(a)(1). We therefore lack jurisdiction over\nthis claim. 7\n\n B\n\n The government next argues that the district court\nmodified the Agreement by concluding that it requires the\ngovernment to consider releasing class members subject to\nexpedited removal. The government contends that this\ninterpretation of the Agreement is inconsistent with the\nImmigration and Nationality Act (INA) and related\nregulations—primarily with the expedited removal\nprovisions, which provide that noncitizens in expedited\nremoval proceedings “shall be detained for further\nconsideration of the[ir] application[s] for asylum.” 8 U.S.C.\n§ 1225(b)(1)(B)(ii).\n\n The Agreement “creates a presumption in favor of\nreleasing minors.” Flores v. Lynch, 828 F.3d at 901; accord\nFlores v. Sessions, 862 F.3d at 866. That presumption is fully\nconsistent with the Act’s expedited removal provisions.\n\n To begin, not all noncitizens eligible to be placed in\nexpedited removal proceedings are in fact placed in such\n\n 7\n As we lack jurisdiction, we do not reach the government’s\nevidentiary objections regarding the record considered by the district\ncourt.\n\f16 FLORES V. BARR\n\nproceedings. The government has discretion to place\nnoncitizens in standard removal proceedings even if the\nexpedited removal statute could be applied to them. See\nMatter of E-R-M- & L-R-M-, 25 I. & N. Dec. 520, 521–22\n(B.I.A. 2011). In other words, the government may place\nminors into standard, nonexpedited removal proceedings\nand thus comply with the Agreement by avoiding any\nmandatory detention allegedly required for expedited\nremoval.\n\n Further, expedited removal does not require mandatory\ndetention for minors. The INA provides that, even for\nnoncitizens in expedited removal, “the Attorney General\nmay . . . in his discretion parole into the United States\ntemporarily” any noncitizen applying for admission “under\nsuch conditions as he may prescribe.” 8 U.S.C.\n§ 1182(d)(5)(A). The government has promulgated two\nregulations that pertain to parole into the United States of\nnoncitizens in expedited removal proceedings. One provides\nthat such noncitizens “shall be detained pending\ndetermination and removal, except that parole of such alien\n. . . may be permitted only when the Attorney General\ndetermines, in the exercise of discretion, that parole is\nrequired to meet a medical emergency or is necessary for a\nlegitimate law enforcement objective.” 8 C.F.R.\n§ 235.3(b)(2)(iii). A second provides that all noncitizens\nsubject to removal—specifically including, by cross-\nreference, those in expedited removal 8—may be paroled “on\na case-by-case basis for ‘urgent humanitarian reasons’ or\n‘significant public benefit,’ provided the [noncitizens]\n\n 8\n Section 212.5(b) addresses the parole of aliens “who have been or\nare detained in accordance with § 235.3(b) or (c) of this chapter.”\n8 C.F.R. § 212.5(b). Section 235.3(b) provides for expedited removal.\nId. § 235.3.\n\f FLORES V. BARR 17\n\npresent neither a security risk nor a risk of absconding.” Id.\n§ 212.5(b). Among the groups eligible for parole under this\nregulation are “[a]liens who are defined as juveniles in\n§ 236.3(a) of this chapter,” who may be paroled under “the\nguidelines set forth in § 236.3(a) of this chapter and\nparagraphs (b)(3)(i) through (iii) of this section.” Id.\n§ 212.5(b)(3).\n\n Both regulations expressly cover noncitizens in\nexpedited removal proceedings; nothing in section 235.3(b)\nnegates section 212.5(b). The more specific regulatory\nprovision providing an exception for minors governs, not the\ngeneral expedited removal provisions. See Karczewski v.\nDCH Mission Valley LLC, 862 F.3d 1006, 1015–16 (9th Cir.\n2017) (applying the interpretive canon that “the specific\ngoverns over the general” to regulations).\n\n The upshot is that the government’s own regulations\ncontemplate that minors in expedited removal proceedings\nmay be considered for release, just as the Agreement\nrequires. Rather than modifying the Agreement, the district\ncourt appropriately interpreted it as consistent with both the\nINA and our prior interpretation of the Agreement. We\ntherefore lack jurisdiction over this claim. 9\n\n\n\n\n 9\n For the first time on appeal, the government contends that the\ndistrict court’s conclusion that the government must consider whether to\nrelease minors to potential custodians other than a parent or legal\nguardian conflicts with a provision of the TVPRA, 8 U.S.C.\n§ 1232(b)(3). We decline to address this statutory defense in this appeal,\nas the government did not raise it before the district court. See Cold\nMountain v. Garber, 375 F.3d 884, 891 (9th Cir. 2004).\n\f18 FLORES V. BARR\n\n C\n\n Finally, the government argues that the district court\nerred in concluding that the Agreement prohibits the\ngovernment from detaining minors in secure, unlicensed\nfamily detention centers. The district court addressed this\nissue directly in its July 2015 order. Although the\ngovernment appealed that order, it did not on appeal\nchallenge the district court’s holding on this issue. See\nFlores v. Lynch, 828 F.3d at 901. “[A] party cannot offer up\nsuccessively different legal or factual theories that could\nhave been presented in a prior request for review.” Sec. Inv’r\nProt. Corp. v. Vigman, 74 F.3d 932, 937 (9th Cir. 1996). The\nissue belatedly raised in this appeal is not properly before us.\n\n III\n\n We dismiss the appeal for lack of jurisdiction.\n\n DISMISSED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Jenny Flores v. William Barr\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-15\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nJENNY LISETTE FLORES, et al., No. 17-56297\\n Plaintiffs-Appellees,\\n D.C. No.\\n v. 2:85-cv-04544-\\n DMG-AGR\\nWILLIAM P. BARR, Attorney General;\\nKEVIN K. MCALEENAN, Acting\\nSecretary of Homeland Security; OPINION\\nU.S. DEPARTMENT OF HOMELAND\\nSECURITY; U.S. IMMIGRATION AND\\nCUSTOMS ENFORCEMENT; U.S.\\nCUSTOMS AND BORDER PROTECTION,\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Dolly M. Gee, District Judge, Presiding\\n\\n Argued and Submitted June 18, 2019\\n San Francisco, California\\n\\n Filed August 15, 2019\\n\\n Before: A. Wallace Tashima, William A. Fletcher,\\n and Marsha S. Berzon, Circuit Judges.\\n\\n Opinion by Judge Berzon\\n\\f2 FLORES V. BARR\\n\\n SUMMARY *\\n\\n\\n Immigration / Juvenile Detention\\n\\n The panel dismissed for lack of jurisdiction an appeal\\nbrought by the Department of Homeland Security and its\\ncomponent agencies of the district court’s June 2017 order\\ngranting in part the motion of a plaintiff class to enforce a\\n1997 Settlement Agreement with the government which set\\na nationwide policy for the detention, release, and treatment\\nof minors detained in Immigration and Naturalization\\nService custody.\\n\\n In 1997, the United States entered into a settlement with\\na class of minors subject to detention by U.S. immigration\\nauthorities. The Settlement Agreement, incorporated into a\\nconsent decree, requires immigration agencies to hold such\\nminors in their custody “in facilities that are safe and\\nsanitary.” The Agreement also requires the government to\\ntreat these “minors in its custody with dignity, respect, and\\nspecial concern for their particular vulnerability as minors.”\\n\\n Plaintiffs filed a motion in district court to enforce the\\nAgreement. The district court found that the government was\\nviolating the Agreement by detaining minors in unsanitary\\nand unsafe conditions at Border Patrol stations. These\\nfindings were based on evidence that minors in U.S.\\nCustoms and Border Protection custody were held in\\nconditions that deprived them of sleep and did not provide\\nadequate access to food, clean water, and basic hygiene\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f FLORES V. BARR 3\\n\\nitems. The court also found that the government was\\nviolating the Agreement by failing to consider minors for\\nrelease as specified in the Agreement and by detaining\\nminors in detention facilities not licensed for the care of\\nminors. The district court ordered “enforced” various\\nparagraphs of the Agreement and also directed the\\ngovernment to appoint an internal “Juvenile Coordinator,”\\nas contemplated by the Agreement, to monitor the\\ngovernment’s compliance with the Agreement and report to\\nthe court.\\n\\n The parties agreed that this court has jurisdiction over the\\nappeal of this post-judgment order only if it modified the\\nAgreement.\\n\\n The government argued that the district court’s order did\\nmodify the Agreement by requiring the government to\\nprovide specific hygiene items and adequate sleeping\\naccommodations not explicitly listed in the text of the\\nAgreement. The panel held that the district court’s order did\\nnot modify the Agreement, but instead interpreted the\\nAgreement’s requirement that minors be held in “safe and\\nsanitary” conditions “consistent with the [government’s]\\nconcern for the particular vulnerability of minors.”\\n\\n The government also argued that the district court\\nmodified the Agreement by concluding that it requires the\\ngovernment to consider releasing class members subject to\\nexpedited removal. The government contended that this\\ninterpretation of the Agreement was inconsistent with the\\nImmigration and Nationality Act and related regulations—\\nprimarily with the expedited removal provisions, which\\nprovide that noncitizens in expedited removal proceedings\\n“shall be detained for further consideration of the[ir]\\napplication[s] for asylum.” 8 U.S.C. § 1225(b)(1)(B)(ii).\\n\\f4 FLORES V. BARR\\n\\nThe panel held that, rather than modifying the Agreement,\\nthe district court appropriately interpreted it as consistent\\nwith both the INA and this court’s prior interpretation of the\\nAgreement.\\n\\n Regarding the government’s argument that the district\\ncourt erred in concluding that the Agreement prohibits the\\ngovernment from detaining minors in secure, unlicensed\\nfamily detention centers, the panel noted that the district\\ncourt addressed this issue directly in its earlier July 2015\\norder, and that although the government appealed that order,\\nit did not on appeal challenge the district court’s holding on\\nthis issue. The panel concluded that this issue belatedly\\nraised in this appeal was not properly before the court.\\n\\n Because the panel concluded that the district court’s\\norder did not modify the Agreement, it dismissed the appeal.\\n\\n\\n COUNSEL\\n\\nSarah Fabian (argued), Senior Litigation Counsel; William\\nC. Silvis, Assistant Director, District Court Section; William\\nC. Peachey, Director, District Court Section; Office of\\nImmigration Litigation, Civil Division, United States\\nDepartment of Justice, Washington, D.C.; for Defendants-\\nAppellants.\\n\\nPeter Anthony Schey (argued) and Carlos R. Holguin,\\nCenter for Human Rights and Constitutional Law, Los\\nAngeles, California; Elena Garcia, Orrick Herrington &\\nSutcliffe LLP, Los Angeles, California; Michael S. Sorgen,\\nLa Raza Centro Legal Inc., San Francisco, California;\\nJennifer Kelleher Cloyd, Katherine H. Manning, and\\nAnnette Kirkham, The Law Foundation of Silicon Valley\\n\\f FLORES V. BARR 5\\n\\nLegal Advocates for Children and Youth Public Interest Law\\nFirm, San Jose, California; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nBERZON, Circuit Judge:\\n\\n In 1997, the United States entered into a settlement with\\na class of minors subject to detention by U.S. immigration\\nauthorities (“Plaintiffs”). The settlement agreement,\\nincorporated into a consent decree, requires immigration\\nagencies to hold such minors in their custody “in facilities\\nthat are safe and sanitary.” Flores Agreement (“the\\nAgreement”) at ¶ 12A. The Agreement also requires the\\ngovernment to treat these “minors in its custody with dignity,\\nrespect, and special concern for their particular vulnerability\\nas minors.” Id. at ¶ 11.\\n\\n This appeal began as a motion by the Plaintiffs to enforce\\nthe Agreement. The district court found that the government\\nwas violating the Agreement by detaining minors in\\nunsanitary and unsafe conditions at Border Patrol stations.\\nThese findings were based on evidence that minors in U.S.\\nCustoms and Border Protection custody were held in\\nconditions that deprived them of sleep and did not provide\\nadequate access to food, clean water, and basic hygiene\\nitems. The court also found that the government was\\nviolating the Agreement by failing to consider minors for\\nrelease as specified in the Agreement and by detaining\\nminors in detention facilities not licensed for the care of\\nminors. The district court ordered “enforced” various\\nparagraphs of the Agreement and also directed the\\ngovernment to appoint an internal “Juvenile Coordinator,”\\nas contemplated by the Agreement, to monitor the\\n\\f6 FLORES V. BARR\\n\\ngovernment’s compliance with the Agreement and report to\\nthe court.\\n\\n The parties agree that this court has jurisdiction over the\\nappeal of this post-judgment order only if it modified the\\nAgreement. The government argues that the district court’s\\norder did modify the Agreement by requiring the\\ngovernment to provide specific hygiene items and adequate\\nsleeping accommodations not explicitly listed in the text of\\nthe Agreement. We disagree. The district court’s order does\\nnot modify the Agreement. Instead, it interprets the\\nAgreement’s requirement that minors be held in “safe and\\nsanitary” conditions “consistent with the [government’s]\\nconcern for the particular vulnerability of minors.” See\\nAgreement at ¶ 12A. The government also argues that the\\ndistrict court’s order modifies the Agreement in other\\nrespects, but those arguments likewise lack merit. As the\\ndistrict court’s order did not modify the Agreement we\\ndismiss the appeal.\\n\\n I\\n\\n A\\n\\n This case stems from a 1985 lawsuit filed on behalf of a\\nclass of minors detained by U.S. immigration authorities.\\nAfter considerable litigation, the parties negotiated the\\nAgreement; it was entered by the district court as a consent\\ndecree in January 1997. The Agreement remains in effect\\ntoday. 1 The Agreement “sets out nationwide policy for the\\n 1\\n The Agreement included a specified termination date, but in 2001\\nthe parties stipulated to extend the Agreement until “45 days following\\ndefendants’ publication of final regulations implementing this\\nAgreement.” The government has issued proposed regulations but the\\nregulations have not been made final. See Apprehension, Processing,\\n\\f FLORES V. BARR 7\\n\\ndetention, release, and treatment of minors in the custody of\\nthe INS.” Id. at ¶ 9. 2 It requires the government to “place\\neach detained minor in the least restrictive setting\\nappropriate to the minor’s age and special needs, provided\\nthat such setting is consistent with its interests to ensure the\\nminor’s timely appearance before the INS and the\\nimmigration courts and to protect the minor’s well-being and\\nthat of others.” Id. at ¶ 11. The Agreement’s provisions\\n“create[] a presumption in favor of releasing minors and\\nrequire[] placement of those not released in licensed, non-\\nsecure facilities that meet certain standards.” Flores v.\\nLynch, 828 F.3d 898, 901 (9th Cir. 2016).\\n\\n Under paragraph 12A of the Agreement, “[f]ollowing\\narrest, the INS shall hold minors in facilities that are safe and\\nsanitary and that are consistent with the INS’s concern for\\nthe particular vulnerability of minors.” The Agreement\\ncontinues: “Facilities will provide access to toilets and sinks,\\ndrinking water and food as appropriate, medical assistance if\\nthe minor is in need of emergency services, adequate\\ntemperature control and ventilation, adequate supervision to\\n\\n\\n\\n\\nCare, and Custody of Alien Minors and Unaccompanied Alien Children,\\n83 Fed. Reg. 45,486 (proposed Sept. 7, 2018) (to be codified at 8 C.F.R.\\n§§ 212, 236; 45 C.F.R. § 410); see also Apprehension, Processing, Care,\\nand Custody of Alien Minors, Regulations.gov, https://www.regulations\\n.gov/docket?D=ICEB-2018-0002 (last visited Aug. 6, 2019).\\n\\n 2\\n Although the Agreement’s terms refer to “INS,” the Immigration\\nand Naturalization Service’s obligations under the Agreement now apply\\nto the Department of Homeland Security and the Department of Health\\nand Human Services. See Flores v. Sessions, 862 F.3d 863, 870 (9th Cir.\\n2017).\\n\\f8 FLORES V. BARR\\n\\nprotect minors from others, and contact with family\\nmembers who were arrested with the minor.”\\n\\n Within a few days of initial detention—three days if a\\nsuitable detention facility “is located and has space\\navailable” or five days “in all other cases”—the government\\nordinarily must choose between two options for placement\\nof minors. 3 The first, and preferable, option, discussed in\\nparagraph 14 of the Agreement, is releasing the minor to a\\nparent, legal guardian, adult relative, or another “capable and\\nwilling” designated adult. Alternatively, under paragraph 19,\\nthe minor may be placed in a facility “licensed by an\\nappropriate State agency to provide residential, group, or\\nfoster care services for dependent children.” Licensed\\nfacilities must be “non-secure as required under state law.”\\n\\n Finally, paragraph 28A provides that “[a]n INS Juvenile\\nCoordinator . . . shall monitor compliance with the terms of\\nthis Agreement and shall maintain an up-to-date record of all\\nminors who are placed in proceedings and remain in INS\\ncustody for longer than 72 hours.”\\n\\n\\n\\n\\n 3\\n Paragraph 21 of the Agreement provides a third option—\\nplacement in a secure juvenile detention facility—in limited\\ncircumstances, such as where the minor has been charged with a crime.\\nAlthough the Agreement contemplates secure detention for minors that\\nare “an escape-risk,” the William Wilberforce Trafficking Victims\\nProtection Reauthorization Act of 2008 (“TVPRA”) limits the use of\\nsecure custody, stating that “[a] child shall not be placed in a secure\\nfacility absent a determination that the child poses a danger to self or\\nothers or has been charged with having committed a criminal offense.”\\n8 U.S.C. § 1232(c)(2)(A); see also Flores v. Sessions, 862 F.3d at 881\\nn.19.\\n\\f FLORES V. BARR 9\\n\\n B\\n\\n In May 2016, the plaintiffs filed a motion to enforce the\\nAgreement. The plaintiffs alleged that despite earlier such\\nmotions and resulting orders enforcing the Agreement, the\\ngovernment continued to violate it by detaining class\\nmembers in unsafe and unsanitary conditions at Border\\nPatrol stations and by detaining minors in secure, unlicensed\\nfacilities. The plaintiffs requested that the district court\\naddress these violations by appointing an independent\\nmonitor.\\n\\n In June 2017, after an evidentiary hearing, the district\\ncourt granted in part plaintiffs’ motion to enforce. In its order\\n(“the Order”), the court found that the government was\\nviolating the Agreement’s express requirements to provide\\nadequate access to appropriate food and water and “adequate\\ntemperature controls at a reasonable and comfortable range.”\\nThe court further found that although the Agreement “makes\\nno mention of the words ‘soap,’ ‘towels,’ ‘showers,’ ‘dry\\nclothing,’ or ‘toothbrushes,’ . . . these hygiene products fall\\nwithin the rubric of the Agreement’s language requiring\\n‘safe and sanitary’ conditions.” Certain Border Patrol\\nstations, the district court found, were violating paragraph\\n12A of the Agreement by failing to provide such sanitary\\nnecessities.\\n\\n The district court also determined that although “the\\nword ‘sleep’ does not appear in the Agreement, . . . whether\\nDefendants have set up conditions that allow class members\\nto sleep in the [Border Patrol] facilities is relevant to the\\nissue of whether they have acted in a manner that is\\nconsistent with ‘the INS’s concern for the particular\\nvulnerability of minors’ as well as the Agreement’s ‘safe and\\nsanitary’ requirement.” Citing evidence that many minors in\\nBorder Patrol custody are forced to sleep on concrete floors,\\n\\f10 FLORES V. BARR\\n\\nwith no bedding aside from pieces of thin polyester foil, and\\nare subjected to cold temperatures, serious overcrowding,\\nand constant lighting, the district court found that the\\ngovernment was violating the Agreement at certain Border\\nPatrol stations by holding children in facilities that deprived\\nthem of adequate sleep.\\n\\n The court also found that the government was failing to\\nmake and record ongoing efforts aimed at releasing or\\nplacing class members in violation of paragraph 14 of the\\nAgreement, and was detaining class members in secure,\\nunlicensed facilities in violation of paragraph 19.\\n\\n The district court declined, however, to appoint an\\nindependent monitor as the plaintiffs had requested. Instead,\\nit directed the government to identify an internal “Juvenile\\nCoordinator” in accordance with paragraph 28A of the\\nAgreement. The court instructed that the Juvenile\\nCoordinator, once appointed, “will monitor compliance with\\nthose terms of the Flores Agreement, which this Court has\\nfound must be enforced and shall report directly to the Court\\nregarding the status of Defendants’ compliance.” 4\\n\\n II\\n\\n A\\n\\n This court has appellate jurisdiction over interlocutory\\ndistrict court orders “granting, continuing, modifying,\\nrefusing or dissolving injunctions, or refusing to dissolve or\\n\\n 4\\n A Special Master has since been appointed. See Order Appointing\\nSpecial Master/Independent Monitor, Flores v. Barr, No. 2:85-cv-4544\\n(Order) (C.D. Cal. Oct. 5, 2018) (ECF No. 494). Neither the appointment\\nof the Special Master nor the Special Master’s enforcement activities is\\nthe subject of this appeal.\\n\\f FLORES V. BARR 11\\n\\nmodify injunctions.” 28 U.S.C. § 1292(a)(1). The parties\\nagree that the district court’s Order did not grant, continue,\\nrefuse, dissolve, or refuse to dissolve an injunction. The\\nonly possible basis for appellate jurisdiction therefore\\ndepends on whether the Order “modif[ied]” the Agreement.\\nSee Thompson v. Enomoto, 815 F.2d 1323, 1327 (9th Cir.\\n1987) (appointment of a special master cannot be appealed\\nunder § 1292(a)(1) unless “the interlocutory order\\nappointing a special master ‘modifies’ the consent decree”);\\nsee also Bogard v. Wright, 159 F.3d 1060, 1064–65 (7th Cir.\\n1998) (dismissing for lack of jurisdiction an appeal of an\\norder that enforced but did not modify a consent decree, and\\ndiscussing the “appellate remedy” available to each party\\nunder section 1292(a)(1) with respect to orders after a final\\njudgement).\\n\\n The text of the district court’s opinion in this case\\ninterpreted the existing Agreement; its operative provisions,\\nwhich appear at the end of the Order, granted enforcement\\nof various provisions of the Agreement. The conclusion of\\nthe Order reads, for example, “Plaintiffs’ motion to enforce\\nParagraph 12A of the Agreement on the issue of unsanitary\\nconditions is GRANTED . . .” The Order’s operative\\nprovisions do not require the government to take any specific\\naction other than to propose a Juvenile Coordinator for\\nappointment. 5\\n\\n The government does not challenge the appointment of\\na Juvenile Coordinator, as that appointment is expressly\\n 5\\n In contrast, the district court’s July 2015 order, over which this\\ncourt exercised appellate jurisdiction, see Flores v. Lynch, 828 F.3d\\nat 905, did direct the government to take specific actions, see id. at 908\\n(“To comply with Paragraph 14A of the Agreement and as contemplated\\nin Paragraph 15, a class member’s accompanying parent shall be released\\nwith the class member . . . .”).\\n\\f12 FLORES V. BARR\\n\\nprovided for in the Agreement. Rather, the government\\nargues that we have jurisdiction to review other parts of the\\ndistrict court’s Order because they modified the Flores\\nAgreement.\\n\\n Specifically, the government contends that, by\\ninterpreting paragraph 12A in the body of its opinion to\\nrequire that Border Patrol stations provide the most basic\\nhuman necessities—accommodations that allow for\\nadequate sleep, essential hygiene items, and adequate, clean\\nfood and water—the district court modified the Agreement’s\\nrequirement that minors be held in “safe and sanitary”\\nconditions that comport with the “special concern for the\\nparticular vulnerability of minors.” We emphatically\\ndisagree.\\n\\n The government first suggests that the key phrases in\\nparagraph 12A—“safe and sanitary” and “special concern\\nfor the particular vulnerability of minors”—add nothing to\\nthe enumerated specific requirements found in the next\\nsentence of the Agreement (requiring “access to toilets and\\nsinks, drinking water and food as appropriate,” and so on).\\nThe government’s brief maintains that as the enumerated\\nconditions said nothing about, for example, allowing the\\nchildren in government custody to sleep or to wash\\nthemselves with soap, reading the “safe and sanitary”\\nrequirement to cover those requirements is a modification of\\nthe Agreement rather than an interpretation of it.\\n\\n That cramped understanding of paragraph 12A is\\nuntenable. Construing the Agreement as requiring only the\\nparticular conditions specifically enumerated renders both\\nthe “safe and sanitary” and the “particular vulnerability of\\nminors” phrases wholly superfluous. We cannot accept that\\nthe parties to the Agreement included gratuitous standards\\nthat have no practical impact. “Courts interpreting the\\n\\f FLORES V. BARR 13\\n\\nlanguage of contracts ‘should give effect to every provision,’\\nand ‘an interpretation which renders part of the instrument\\nto be surplusage should be avoided.’” United States v. 1.377\\nAcres of Land, 352 F.3d 1259, 1265 (9th Cir. 2003) (quoting\\nAppalachian Ins. Co. v. McDonnell Douglas Corp., 214 Cal.\\nApp. 3d 1, 12 (Ct. App. 1989)); see also Restatement\\n(Second) of Contracts § 203 (Am. Law Inst. 1981). We\\nconclude that paragraph 12A’s provisions that facilities be\\n“safe and sanitary and . . . consistent with the INS’s concern\\nfor the particular vulnerability of minors” do have\\nindependent force and can be interpreted and enforced\\nwithout thereby modifying the Agreement.\\n\\n The government also argues that the phrase “safe and\\nsanitary” is so vague that either it cannot be enforced, see\\nOral Argument at 33:57–34:09, Flores v. Barr, No. 17-\\n56297 (9th Cir. Jun. 18, 2019), https://www.ca9.uscourts.g\\nov/media/view_video.php?pk_vid=0000015907, or it leaves\\n“the specifics of compliance [with paragraph 12A] up to” the\\ngovernment. Not so.\\n\\n The district court’s interpretation of the Agreement is\\nconsistent with the ordinary meaning of the language of\\nparagraph 12A, which does provide a standard sufficiently\\nclear to be enforced. The court found, among other things,\\nthat minors (1) were “not receiving hot, edible, or a sufficient\\nnumber of meals during a given day,” (2) “had no adequate\\naccess to clean drinking water,” (3) experienced “unsanitary\\nconditions with respect to the holding cells and bathroom\\nfacilities,” (4) lacked “access to clean bedding, and access to\\nhygiene products (i.e., toothbrushes, soap, towels),” and\\n(5) endured “sleep deprivation” as a result of “cold\\ntemperatures, overcrowding, lack of proper bedding (i.e.,\\nblankets, mats), [and] constant lighting.” After so finding,\\nthe district court concluded that these conditions fall short of\\n\\f14 FLORES V. BARR\\n\\nparagraph 12A’s requirement that facilities be “safe and\\nsanitary,” especially given “the particular vulnerability of\\nminors.” Those determinations reflect a commonsense\\nunderstanding of what the quoted language requires.\\nAssuring that children eat enough edible food, drink clean\\nwater, are housed in hygienic facilities with sanitary\\nbathrooms, have soap and toothpaste, and are not sleep-\\ndeprived are without doubt essential to the children’s\\nsafety. 6 The district court properly construed the Agreement\\nas requiring such conditions rather than allowing the\\ngovernment to decide whether to provide them.\\n\\n Moreover, contrary to the government’s assertions, the\\ndistrict court did not incorporate into the Agreement a\\nparticular set of standards, Customs and Border Protection’s\\n“National Standards on Transport, Escort, Detention, and\\nSearch” (or “TEDS”), with respect to food during detention.\\nWe doubt that the TEDS requirements—that minors “be\\noffered a snack upon arrival and a meal at least every six\\nhours thereafter,” have food that is “in edible condition (not\\nfrozen, expired, or spoiled),” and “have regular access to\\nsnacks, milk, and juice,”—extend beyond what paragraph\\n12A requires. But in any event, in context, the district court\\nreferred to TEDS not to interpret the Agreement as\\nincorporating the TEDS standards specifically, but to\\nconfirm that the government’s inattention to ensuring that\\nchildren were being adequately fed was egregious, as the\\ngovernment was not even complying with its own standards.\\n\\n\\n\\n 6\\n We note that, as the district court properly understood, assuring\\n“safe and sanitary” conditions includes protecting children from\\ndeveloping short- or long-term illnesses as well as protecting them from\\naccidental or intentional injury.\\n\\f FLORES V. BARR 15\\n\\n In short, the district court’s explanation of its\\nenforcement of paragraph 12A regarding the conditions at\\nBorder Patrol stations concerned only requirements\\nunarguably within the terms of the Agreement. As a result,\\nthe portion of the court’s order enforcing paragraph 12A did\\nnot constitute an “[i]nterlocutory order[] . . . modifying [an]\\ninjunction[], or refusing to . . . modify [an] injunction[].”\\n28 U.S.C. § 1292(a)(1). We therefore lack jurisdiction over\\nthis claim. 7\\n\\n B\\n\\n The government next argues that the district court\\nmodified the Agreement by concluding that it requires the\\ngovernment to consider releasing class members subject to\\nexpedited removal. The government contends that this\\ninterpretation of the Agreement is inconsistent with the\\nImmigration and Nationality Act (INA) and related\\nregulations—primarily with the expedited removal\\nprovisions, which provide that noncitizens in expedited\\nremoval proceedings “shall be detained for further\\nconsideration of the[ir] application[s] for asylum.” 8 U.S.C.\\n§ 1225(b)(1)(B)(ii).\\n\\n The Agreement “creates a presumption in favor of\\nreleasing minors.” Flores v. Lynch, 828 F.3d at 901; accord\\nFlores v. Sessions, 862 F.3d at 866. That presumption is fully\\nconsistent with the Act’s expedited removal provisions.\\n\\n To begin, not all noncitizens eligible to be placed in\\nexpedited removal proceedings are in fact placed in such\\n\\n 7\\n As we lack jurisdiction, we do not reach the government’s\\nevidentiary objections regarding the record considered by the district\\ncourt.\\n\\f16 FLORES V. BARR\\n\\nproceedings. The government has discretion to place\\nnoncitizens in standard removal proceedings even if the\\nexpedited removal statute could be applied to them. See\\nMatter of E-R-M- & L-R-M-, 25 I. & N. Dec. 520, 521–22\\n(B.I.A. 2011). In other words, the government may place\\nminors into standard, nonexpedited removal proceedings\\nand thus comply with the Agreement by avoiding any\\nmandatory detention allegedly required for expedited\\nremoval.\\n\\n Further, expedited removal does not require mandatory\\ndetention for minors. The INA provides that, even for\\nnoncitizens in expedited removal, “the Attorney General\\nmay . . . in his discretion parole into the United States\\ntemporarily” any noncitizen applying for admission “under\\nsuch conditions as he may prescribe.” 8 U.S.C.\\n§ 1182(d)(5)(A). The government has promulgated two\\nregulations that pertain to parole into the United States of\\nnoncitizens in expedited removal proceedings. One provides\\nthat such noncitizens “shall be detained pending\\ndetermination and removal, except that parole of such alien\\n. . . may be permitted only when the Attorney General\\ndetermines, in the exercise of discretion, that parole is\\nrequired to meet a medical emergency or is necessary for a\\nlegitimate law enforcement objective.” 8 C.F.R.\\n§ 235.3(b)(2)(iii). A second provides that all noncitizens\\nsubject to removal—specifically including, by cross-\\nreference, those in expedited removal 8—may be paroled “on\\na case-by-case basis for ‘urgent humanitarian reasons’ or\\n‘significant public benefit,’ provided the [noncitizens]\\n\\n 8\\n Section 212.5(b) addresses the parole of aliens “who have been or\\nare detained in accordance with § 235.3(b) or (c) of this chapter.”\\n8 C.F.R. § 212.5(b). Section 235.3(b) provides for expedited removal.\\nId. § 235.3.\\n\\f FLORES V. BARR 17\\n\\npresent neither a security risk nor a risk of absconding.” Id.\\n§ 212.5(b). Among the groups eligible for parole under this\\nregulation are “[a]liens who are defined as juveniles in\\n§ 236.3(a) of this chapter,” who may be paroled under “the\\nguidelines set forth in § 236.3(a) of this chapter and\\nparagraphs (b)(3)(i) through (iii) of this section.” Id.\\n§ 212.5(b)(3).\\n\\n Both regulations expressly cover noncitizens in\\nexpedited removal proceedings; nothing in section 235.3(b)\\nnegates section 212.5(b). The more specific regulatory\\nprovision providing an exception for minors governs, not the\\ngeneral expedited removal provisions. See Karczewski v.\\nDCH Mission Valley LLC, 862 F.3d 1006, 1015–16 (9th Cir.\\n2017) (applying the interpretive canon that “the specific\\ngoverns over the general” to regulations).\\n\\n The upshot is that the government’s own regulations\\ncontemplate that minors in expedited removal proceedings\\nmay be considered for release, just as the Agreement\\nrequires. Rather than modifying the Agreement, the district\\ncourt appropriately interpreted it as consistent with both the\\nINA and our prior interpretation of the Agreement. We\\ntherefore lack jurisdiction over this claim. 9\\n\\n\\n\\n\\n 9\\n For the first time on appeal, the government contends that the\\ndistrict court’s conclusion that the government must consider whether to\\nrelease minors to potential custodians other than a parent or legal\\nguardian conflicts with a provision of the TVPRA, 8 U.S.C.\\n§ 1232(b)(3). We decline to address this statutory defense in this appeal,\\nas the government did not raise it before the district court. See Cold\\nMountain v. Garber, 375 F.3d 884, 891 (9th Cir. 2004).\\n\\f18 FLORES V. BARR\\n\\n C\\n\\n Finally, the government argues that the district court\\nerred in concluding that the Agreement prohibits the\\ngovernment from detaining minors in secure, unlicensed\\nfamily detention centers. The district court addressed this\\nissue directly in its July 2015 order. Although the\\ngovernment appealed that order, it did not on appeal\\nchallenge the district court’s holding on this issue. See\\nFlores v. Lynch, 828 F.3d at 901. “[A] party cannot offer up\\nsuccessively different legal or factual theories that could\\nhave been presented in a prior request for review.” Sec. Inv’r\\nProt. Corp. v. Vigman, 74 F.3d 932, 937 (9th Cir. 1996). The\\nissue belatedly raised in this appeal is not properly before us.\\n\\n III\\n\\n We dismiss the appeal for lack of jurisdiction.\\n\\n DISMISSED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4070,"cells":{"id":{"kind":"number","value":4648690,"string":"4,648,690"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Apollo Education Group, Inc. v. National Union Fire Insurance"},"decision_date":{"kind":"string","value":"2019-08-15"},"docket_number":{"kind":"string","value":"17-17293"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Richard A. Paez and Richard R. Clifton, Circuit Judges, and Gary S. Katzmann, * Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"The Honorable Gary S. Katzmann, Judge for the United States\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n APOLLO EDUCATION GROUP, No. 17-17293\\n INC., FKA Apollo Group, Inc.,\\n Plaintiff-Appellant, D.C. No.\\n 2:15-cv-01948-SPL\\n v.\\n\\n NATIONAL UNION FIRE ORDER\\n INSURANCE COMPANY OF CERTIFYING\\n PITTSBURGH, PA, a QUESTION TO THE\\n Pennsylvania corporation, SUPREME COURT\\n Defendant-Appellee. OF ARIZONA\\n\\n Filed August 15, 2019\\n\\n Before: Richard A. Paez and Richard R. Clifton, Circuit\\n Judges, and Gary S. Katzmann, * Judge.\\n\\n Order\\n\\n\\n\\n\\n *\\n The Honorable Gary S. Katzmann, Judge for the United States\\nCourt of International Trade, sitting by designation.\\n\\f2 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n SUMMARY **\\n\\n\\n Certified Question to the Supreme Court of Arizona\\n\\n The panel certified the following question of state law to\\nthe Supreme Court of Arizona:\\n\\n What is the standard for determining whether\\n National Union unreasonably withheld\\n consent to Apollo’s settlement with\\n shareholders in breach of contract under a\\n policy where the insurer has no duty to\\n defend?\\n\\n\\n ORDER\\n\\n Apollo Education Group, Inc., (“Apollo”) has asserted\\nbreach of contract and bad faith claims against its insurer,\\nNational Union Fire Insurance Company (“National Union”)\\nunder a governing insurance policy. National Union was\\ncontractually bound not to unreasonably withhold consent to\\nsettlement from its insured, Apollo. This appeal presents the\\nquestion of the standard imposed by Arizona law on an\\ninsurer’s obligation not to unreasonably withhold consent to\\na settlement, where the insurer has no duty to defend.\\nShould the federal district court assess the objective\\nreasonableness of National Union’s decision to withhold\\nconsent from the perspective of an insurer or an insured?\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 3\\n\\n As detailed below, the Teamsters Local 617 Pension &\\nWelfare Funds (“Teamsters”) filed a securities class action\\nlawsuit against Apollo in the U.S. District Court for the\\nDistrict of Arizona. After five years of litigation, Apollo and\\nthe Teamsters settled. National Union, as Apollo’s insurer,\\ndeclined to consent to the settlement, and Apollo sued\\nNational Union for breach of its contractual obligation not to\\nunreasonably withhold consent. The district court held that\\nbecause National Union reasonably considered the\\nsettlement, Apollo’s breach of contract claim failed as a\\nmatter of law. 1 1ER 4–5. 2\\n\\n Apollo now argues before this court that summary\\njudgment was improper because the district court erred in\\nconcluding that the insurer need only “consider[] the terms”\\nof the settlement. Pl.’s Br. at 38. Apollo contends that under\\nArizona law, an insurer must consent to an objectively\\nreasonable settlement. Id. National Union instead frames\\nthe district court’s decision as construing “the insurance\\npolicy’s consent-to-settlement provision to focus on the\\nobjective reasonableness of National Union’s decision to\\nwithhold consent, rather than whether the settlement was\\nreasonable from Apollo’s perspective.” Def.’s Br. at 30.\\nThus, while both Apollo and National Union contend that\\nthe reasonableness standard is “objective,” they dispute from\\n\\n\\n 1\\n The district court also held that Apollo’s claim that National Union\\nbreached its duty to act in good faith failed as a matter of law. 1ER 5.\\nBecause the district court’s analysis of this claim largely relied upon its\\nbreach of contract analysis, we forgo addressing this claim pending\\nresolution of this open question of Arizona state law on the breach of\\ncontract claim.\\n 2\\n “ER” references are to the excerpts of record filed by the parties in\\nthis court for the purposes of this appeal.\\n\\f4 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nwhose perspective the district court should evaluate\\nreasonableness.\\n\\n Because this issue is governed by Arizona law but is not\\nclearly addressed by relevant Arizona authorities, we certify\\nit to the Supreme Court of Arizona pursuant to the\\nprocedures established by Arizona Revised Statutes § 12-\\n1861 and Rule 27 of the Rules of the Supreme Court of\\nArizona.\\n\\nI. Factual Background\\n\\n Pursuant to Supreme Court of Arizona Rule 27(a)(3)(B),\\nwe set forth the facts relevant to the question certified.\\n\\n In 2005, Apollo purchased a $15 million directors and\\nofficers (“D&O”) insurance policy (“the Policy”) from\\nNational Union. In March 2006, the Wall Street Journal\\npublished an article in which it identified the practice of\\nbackdating stock options for corporate executives. 2ER 95–\\n103. Backdating is particularly problematic when it results\\nin companies under-reporting executive compensation to\\ntheir shareholders. The article did not identify Apollo as one\\nof the companies affected. Id.\\n\\n In June 2006, the U.S. Attorney’s Office for the Southern\\nDistrict of New York issued a grand jury subpoena, and the\\nSecurities and Exchange Commission issued a letter\\nnotifying Apollo of its intent to investigate the company for\\nbackdating. Neither resulted in charges or enforcement\\nactivities. 8SER 1558; 2SER 71. Apollo simultaneously\\nconducted its own internal investigation. Apollo\\nsubsequently disclosed publicly that “57 of the 100 total\\ngrants made during [the relevant] time period used incorrect\\nmeasurement dates for accounting purposes.” 2ER 155. On\\nOctober 18, 2006, after Apollo issued a statement admitting\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 5\\n\\n“various deficiencies in the process of granting and\\ndocumenting stock options,” the Teamsters alleged Apollo’s\\nstock price fell by 22.9% in one day. 4ER 543–44. Apollo’s\\nfinal report, however, found no direct evidence of\\nbackdating. 2ER 154.\\n\\n In November 2006, the Teamsters, as lead plaintiff, filed\\na securities class action in the U.S. District Court for the\\nDistrict of Arizona, alleging fraudulent misrepresentations\\nand backdating by Apollo in violation of the Securities\\nExchange Act of 1934. 3ER 290. In March 2009, the district\\ncourt denied Apollo’s motion to dismiss the Teamsters’\\ncomplaint, finding that the Teamsters “sufficiently pled\\nbackdating.” 3ER 429–500, 460–62. In April 2009, the\\nTeamsters filed an amended complaint that included 54\\nallegedly false statements by Apollo and the reasons why the\\nstatements were false. 4ER 524–42. In March 2011, the\\ndistrict court concluded that the falsity allegations were not\\npled with particularity, and thus, “[b]ecause the court has\\nexpressly found that at least one of the elements of plaintiff’s\\n[backdating] claim is missing, i.e. falsity, that claim fails,\\nand there is no need for the court to consider defendants’ loss\\ncausation and scienter arguments.” 4ER 602. The district\\ncourt dismissed the complaint with prejudice and entered\\nfinal judgment for Apollo. 4ER 604.\\n\\n The Teamsters filed a motion to reconsider and a\\nproposed amended complaint. In its March 2012 order, the\\ndistrict court denied the motion, reasoning that (1) the\\nTeamsters had not met its burden to show that subsequent\\ncase law warranted setting aside the judgment, 4ER 624; and\\n(2) the Teamsters had not shown clear error sufficient to\\njustify reopening a final judgment, 4ER 634.\\n\\n The Teamsters timely appealed the district court’s order\\nto this court. 3ER 314. In January 2013, while the appeal\\n\\f6 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nwas pending, Apollo, National Union, and Apollo’s excess\\ninsurance carriers agreed to mediation with the Teamsters.\\n2ER 27. The parties entered into a Mediation\\nConfidentiality Agreement. 6ER 1075–77. After over a\\nyear of mediation, Apollo agreed to settle with the Teamsters\\nfor $13,125,000. 2ER 27. Apollo had $13,500,000\\nremaining on the Policy with National Union. 2ER 25.\\n\\n Apollo’s counsel believed the settlement was “a great\\ndeal” for Apollo, 5ER 959, and that the appeal of the\\nbackdating litigation would be an “enormously expensive\\ncase to defend” if the case was reversed by this court,\\n5ER 823. Apollo feared the Teamsters could recover “a\\nsubstantial multiple of insurance coverage” if successful on\\nappeal. 5ER 803.\\n\\n Pursuant to the Policy, National Union had no duty to\\ndefend Apollo. Instead, Apollo controlled its own defense\\nand settlement negotiations. 2SER 75, 9SER 1838. The\\nPolicy provides:\\n\\n The Insurer does not, however, under this\\n policy, assume any duty to defend. The\\n Insureds shall defend and contest any Claim\\n made against them. The Insureds shall not\\n admit or assume any liability, enter into\\n settlement agreement, stipulate any\\n judgment, or incur any Defense Costs\\n without the prior written consent of the\\n Insurer. Only those settlements, stipulated\\n judgments and Defense Costs which have\\n been consented to by the Insurer shall be\\n recoverable as Loss under the terms of this\\n policy. The Insurer’s consent shall not be\\n unreasonably withheld, provided that the\\n Insurer shall be entitled to effectively\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 7\\n\\n associate in the defense, the prosecution and\\n the negotiation of any settlement of any\\n Claim that involves or appears reasonably\\n likely to involve the Insurer.\\n\\n2ER 48 (emphasis added).\\n\\n National Union declined to consent to or fund the\\nsettlement. 4ER 770. Apollo nonetheless settled with the\\nTeamsters, using its own funds. Id. Apollo then filed the\\npresent action in district court to recover $13,125,000 under\\nits Policy from National Union. Id. Apollo alleged breach\\nof contract under the Policy and bad faith, arguing that\\nNational Union unreasonably withheld consent to the\\nsettlement. 2ER 23–33.\\n\\n In February 2017, National Union filed a motion for\\nsummary judgment. ECF 74; 6ER 1137. National Union\\nargued that summary judgment was appropriate on Apollo’s\\nbreach of contract, bad faith, and punitive damages claims\\nbecause National Union did not unreasonably withhold its\\nconsent to the settlement. Id. Apollo argued against\\nsummary judgment, contending that there were genuine\\nissues of material fact regarding the reasonableness of its\\nsettlement and National Union’s failure to consent. ECF 87;\\n6ER 1138. In October 2017, the district court granted\\nsummary judgment in favor of National Union. 3 ECF 104;\\n6ER 1139.\\n\\n 3\\n The district court also precluded Apollo from introducing\\ndocuments protected by California’s mediation privilege. 1ER 7.\\nApollo challenged the district court’s ruling on the admissibility of those\\ndocuments in the present appeal, which we affirmed in a concurrently-\\nfiled separate disposition.\\n\\f8 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n The district court first found no genuine issue of material\\nfact that National Union’s withholding of consent to\\nsettlement was reasonable. 4 The district court stated that it\\nconstrued the Policy “in accordance with [its] ‘plain and\\nordinary meaning’” and “‘to effectuate the parties’ intent.’”\\n1ER 3 (internal citations omitted). The district court noted\\nthat “[i]n the context of insurance, questions of\\nreasonableness can be resolved by the Court as a matter of\\nlaw when the position of the parties turns on ‘the\\ninterpretation of standard language in the form policies.’”\\n1ER 3 (quoting Lennar Corp. v. Transamerica Ins. Co.,\\n256 P.3d 635, 641 (Ariz. Ct. App. 2011). The district court\\ndefined National Union’s obligation as “giv[ing] equal\\nconsideration to both the interests of itself and of its\\ninsured.” 1ER 4 (citing Bills v. Ariz. Prop. & Cas. Ins. Guar.\\nFund, 984 P.2d 574, 578 (Ariz. Ct. App. 1999)). The district\\ncourt then noted the undisputed steps National Union took to\\nconsider the settlement and concluded that in light of\\nNational Union’s “extensive analysis weighing the\\nTeamsters’ settlement, taken in conjunction with the express\\nterms of Plaintiff’s policy, it is clear that Defendant fulfilled\\nits obligation to Plaintiff by considering the terms of the\\nTeamsters’ settlement and that Plaintiff’s breach of contract\\nclaim must fail as a matter of law.” 5 1ER 4.\\n\\n\\n 4\\n The district court also concluded that Apollo’s bad faith claim must\\nfail as a matter of law because it “failed to set forth facts from which a\\nreasonable juror could find that Defendant acted unreasonably, or knew\\nthat it was acting unreasonably, when it refused to consent to Plaintiff’s\\nsettlement with the Teamsters.” 1ER 5.\\n 5\\n The district court stated:\\n\\n Defendant concluded that settlement was premature –\\n and likely unnecessary – because of the probability\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 9\\n\\n Apollo filed its Notice of Appeal to this court on\\nNovember 10, 2017. We heard argument on April 8, 2019,\\nand we issued a memorandum disposition on the issue of\\nmediation privilege concurrently with this Certification\\nOrder.\\n\\nII. Question of Law\\n\\n This case turns on Arizona state law. The basic question\\nis: What is the standard for determining whether National\\nUnion unreasonably withheld Apollo’s settlement with\\nshareholders in breach of contract under a policy where the\\ninsurer has no duty to defend? 6 For the reasons discussed\\n\\n that Plaintiff would prevail on appeal. Moreover,\\n Defendant concluded that even in the unlikely event\\n that Plaintiff did not prevail on appeal, there were a\\n variety of other hurdles that Teamster would have to\\n overcome in order to recover a “substantial judgment”\\n against Plaintiff.\\n\\n1 ER 4.\\n 6\\n Windt’s insurance law treatise explains the conflict of interest that\\nan insurer and insured may face in considering a settlement:\\n\\n When a claim against an insured can be settled for the\\n policy limit, a conflict of interest exists between the\\n insured and the insurance company. It is generally in\\n the insured’s interest to settle for the policy limit, since\\n doing so protects the insured from any potential\\n personal liability. However, it is generally not in the\\n insurance company’s interest to settle for the policy\\n limit. Apart from considerations of defense costs, the\\n insurance company cannot do any worse than paying\\n its policy limit. As long as any chance existed, no\\n matter how remote, that the insured might prevail, it\\n would generally be in the insurance company's interest\\n to litigate rather than settle for the policy limit.\\n\\f10 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nbelow, Arizona case law does not allow us to reliably\\ndetermine how the Supreme Court of Arizona would answer\\nthis basic question of the standard for evaluating the\\nreasonableness of National Union’s withholding of consent\\nat issue here.\\n\\n In Arizona, as in other jurisdictions, judicial analysis of\\nan insurance coverage dispute begins with the principle that:\\n\\n [a]n insurance policy is a contract, and in an\\n action based thereon the terms of the policy\\n must govern . . . [W]here the provisions of\\n the contract are plain and unambiguous upon\\n their face, they must be applied as written,\\n and the court will not pervert or do violence\\n to the language used, or expand it beyond its\\n plain and ordinary meaning or add something\\n to the contract which the parties have not put\\n there.\\n\\nHome Indem. Co. v. Wilson, 489 P.2d 244, 247 (Ariz. 1971)\\n(internal quotes and citations omitted). Here, in construing\\nthe phrase “the Insurer’s consent [to settlement] shall not be\\nunreasonably withheld,” the district court relied upon the\\nCourt of Appeals of Arizona’s holding in Bills that the\\n\\n Moreover, liability insurance policies, by their terms,\\n give insurers complete discretion in deciding whether\\n to settle or litigate. In light of that fact, and the conflict\\n of interest that can arise between the insurer and the\\n insured in connection with whether a claim against an\\n insured should be settled, the law has imposed on\\n insurance companies an implied contractual obligation\\n to settle under certain circumstances.\\n\\n1 Allan D. Windt, Insurance Claims and Disputes § 5:1 (6th ed. 2016).\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 11\\n\\ninsurer was obligated “to give equal consideration to both\\nthe interests of itself and of its insured.” 1ER 4 (quoting\\nBills, 984 P.2d at 578). The district court interpreted this to\\nrequire that National Union “consider[] the terms and\\nundert[ake] an assessment of a variety of factors” and\\nconcluded that National Union “fulfilled its obligation to\\n[Apollo] by considering the terms of the Teamsters’\\nsettlement.” 1ER 4. In Bills, however, the court was plainly\\nsetting forth the meaning of the insurer’s obligations under\\n“the implied covenant of good faith and fair dealing,” not an\\ninsurer’s obligations under a consent-to-settle provision.\\n984 P.2d at 578. Bills, therefore, does not answer the\\nquestion before us.\\n\\n Apollo relies primarily on two Arizona cases, United\\nServices Automobile Association v. Morris, 741 P.2d 246\\n(Ariz. 1987) and Himes v. Safeway Ins. Co., 66 P.3d 74\\n(Ariz. Ct. App. 2003), to argue that the district court, in its\\nbreach of contract analysis, should have evaluated the\\nobjective reasonableness of the settlement, not the extent to\\nwhich National Union may have considered settlement\\nfactors. Pl.’s Br. at 44–45. In both cases, Arizona courts set\\nforth an objective standard, focused on the merits of the\\nunderlying claim, to determine whether a settlement was\\nreasonable, and thus whether the insurer must accept it.\\n\\n In Morris, the insurer, United States Auto Insurance\\n(“USAA”), had a duty to defend the insured homeowners in\\na tort action. 741 P.2d 246. USAA defended the case but\\nreserved its right to deny coverage, claiming the\\nhomeowners may have committed intentional acts outside\\nthe policy’s coverage. Id. at 249. The Supreme Court of\\nArizona, however, noted that this reservation of rights put\\nthe homeowners in a precarious position, incentivizing them\\nto limit their personal liability should they not be covered by\\n\\f12 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nthe policy. Id. at 251. USAA’s position, however, was that\\nthe homeowners’ duty to cooperate meant USAA could\\nforce the homeowners to “reject any settlement, no matter\\nhow reasonable, risk trial, and place [the homeowners] at\\ndanger of a judgment larger than the policy limits or one that\\nmight not be covered.” Id.\\n\\n The Supreme Court of Arizona explained why USAA’s\\ninterpretation of its right – to reject a settlement, no matter\\nhow reasonable – was problematic. The Court noted that this\\nwould “hamstring[] insureds while granting the insurer a\\ndouble bite at escaping liability.” Id. “If the verdict were in\\nfavor of [plaintiff] . . . USAA would have another chance at\\nescaping the obligation to indemnify because it would be\\nable to relitigate the intentional act exclusion coverage issue\\n. . . In addition, absent bad faith conduct, USAA will never\\nbe at risk for more than [the] . . . policy limit.” Id. On the\\nother hand, the court observed that insureds have an\\nincentive to settle for any amount that allows them to escape\\npersonal liability – that is, any amount at or below the policy\\nlimit – even if the amount is unreasonable and unwarranted.\\nId. The Supreme Court of Arizona thus recognized the\\nconflict between an insurer and insured’s interests,\\nparticularly when a settlement may be at or near policy limits\\nand the insured could face greater personal liability at trial.\\n\\n In response to this problem, the Supreme Court of\\nArizona adopted the standard set forth by Minnesota’s\\nSupreme Court in Miller v. Shugart, 316 N.W.2d 729 (Minn.\\n1982), which held that “neither the fact nor amount of\\nliability to the claimant is binding on the insurer unless the\\ninsured or claimant can show that the settlement was\\nreasonable and prudent,” in accordance with indemnification\\nprinciples. Morris, 741 P.2d at 253 (quoting Miller,\\n316 N.W.2d at 735). The Supreme Court of Arizona then\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 13\\n\\napplied the Miller test to determine “whether the settlement\\nwas reasonable and prudent.” Id. at 254. The Court looked\\nto “what a reasonably prudent person in the insureds’\\nposition would have settled for on the merits of the\\nclaimant’s case. This involves evaluating the facts bearing\\non the liability and damage aspects of a claimant’s case, as\\nwell as the risks of going to trial.” Id. (citing Miller,\\n316 N.W. 2d at 735). Thus, if the Morris standard applies to\\nthis case, the district court would look to whether it was\\nobjectively reasonable for Apollo to reach the settlement it\\ndid, not whether National Union thought it was reasonable\\nor the steps it took to consider and evaluate the settlement.\\n\\n In Himes, the Court of Appeals of Arizona applied the\\nreasonableness test set forth in Morris. 66 P.3d 74. There,\\nan insured involved in an auto accident settled with the\\naccident victim, giving the victim his breach of duty claims\\nagainst the insurer in exchange for an agreement that the\\nvictim would not execute against the insured’s personal\\nassets. Id. at 87. The trial court judge held an evidentiary\\nhearing and found the settlement reasonable. Id. at 78. The\\ninsurer appealed. Id. at 79. The Himes court held that the\\nburden of proving the settlement was reasonable was\\nsquarely on the insured, but that the trial judge’s conclusion\\nthat there was “no evidence that the agreement was\\nunreasonable” was “plainly contradicted by the record.” Id.\\nat 79–80.\\n\\n The Himes court reiterated the “reasonably prudent\\nperson” standard adopted in Morris and enumerated factors\\npotentially relevant to evaluating reasonableness:\\n\\n [T]he releasing person’s damages; the merits\\n of the releasing person’s liability theory; the\\n merits of the released person’s defense\\n theory; the released person’s relative faults;\\n\\f14 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n the risks and expenses of continued litigation\\n [on the merits]; . . . any evidence of bad faith,\\n collusion, or fraud; the extent of the releasing\\n person’s investigation and preparation of the\\n case; and the interests of the parties not being\\n released.\\n\\nId. at 85 (quoting Chaussee v. Md. Cas. Co., 803 P.2d 1339,\\n1343 (Wash. Ct. App. 1991)). The Himes court further noted\\nthat, “it is only those costs, expenses, and risks that relate to\\nthe merits of the underlying case that should be considered.\\nThe costs, expenses, and risk of litigation for any coverage\\nor bad faith litigation . . . should not be considered.” Id. The\\nfocus should only be on the merits of the claimant’s case,\\n“reflect[ing] what would occur in an arm’s-length\\nnegotiation.” Id. The Himes court also noted that under\\nMorris and related case law an insured’s ability to pay\\nshould not be considered, except in punitive damages cases.\\nId. The Himes court defined “reasonably prudent person” as\\none who “has the ability to pay a reasonable settlement from\\nhis or her own funds” to avoid the need to introduce evidence\\nof the insured’s ability to pay into the reasonableness\\nanalysis. Id. Lastly, the Himes court noted that its list of\\nfactors was not exhaustive; courts should generally consider\\nfactors which are “relevant in re-creating a genuine arm’s\\nlength negotiation.” Id.\\n\\n National Union counters that unlike the policy before us,\\nin Morris and Himes the insurer had a duty to defend the\\ninsured; that is, these cases on which Apollo relies are\\ninapposite because they do not involve “reimbursement\\npolic[ies] under which the insurer has no duty to defend the\\ninsured.” Def.’s Br. at 37. Both cases, moreover, do not\\ninvolve the contract language, “insurer’s consent shall not be\\nunreasonably withheld,” at issue here. National Union\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 15\\n\\ncontends that the consent to settle line of cases, involving\\nthird-party claims, cannot apply because, “[u]nder Arizona\\nlaw, a failure to reimburse the cost of a settlement entered by\\nthe insured implicates a first-party claim for failure to pay\\nbenefits, not a third-party claim for failure to defend the\\ninsured.” Def.’s Br. at 37 (citing Trus Joist Corp. v. Safeco\\nIns. Co. of Am., 735 P.2d 125, 135–36 (Ariz. Ct. App.\\n1986)). This court cannot say whether the Supreme Court of\\nArizona would apply the reasonable settlement standard in\\nMorris and Himes here.\\n\\n The parties disagree fundamentally on the purpose of the\\nconsent to settlement provision in the policy before us.\\nNational Union, citing the Windt treatise, contends that\\n“when an insured desires to settle a lawsuit that has been\\nmade against it with an insurer’s money, the insured does not\\nhave complete discretion to spend the insurer’s money.”\\nDef.’s Br. at 31 (quoting 2 Allan D. Windt, Insurance Claims\\n& Disputes § 6:29, at 6-293 n.1 (6th ed. 2016)). “Since the\\ninsured wants to spend not its own money, but someone\\nelse’s money, the issue is not whether the insured had a\\nreasonable basis to pay a particular amount in settlement, but\\nwhether the insurer had a reasonable basis not to agree to\\npay that amount of money in settlement.” Id. (quoting Windt\\nat 6-293 n.1) (emphasis added by National Union). Apollo,\\nciting Morris, argues that the objectively reasonable\\nsettlement rule is necessary to protect the insured because of\\nthe inherent conflict that arises when an insured is presented\\nwith a settlement opportunity at or near the policy limits and\\nthe insurer has issued a reservation of rights and strong\\nincentives to avoid a settlement because it can rely on both\\nliability and coverage defenses and absent bad faith will\\nnever face exposure beyond its policy limits. Pl.’s Br. at 44–\\n47.\\n\\f16 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n Though in disagreement as to how consent to settlement\\nprovisions should be evaluated, the parties are in agreement\\nas to their essential importance to the system of liability\\ninsurance. We think that the resolution of that disagreement\\nand consideration of an open question of Arizona\\njurisprudence are better entrusted to the Supreme Court of\\nArizona than to us.\\n\\nIII. Certified Question\\n\\n In light of the above discussion, we certify the following\\nquestion of law: What is the standard for determining\\nwhether National Union unreasonably withheld consent to\\nApollo’s settlement with shareholders in breach of contract\\nunder a policy where the insurer has no duty to defend? This\\ncourt will accept the Arizona Supreme Court’s decision on\\nthis question. Our phrasing of the question is not intended\\nto restrict the Arizona Supreme Court’s consideration of the\\ncase or formulation of the question. See Broad v.\\nMannesmann Anglagenbau AG, 196 F.3d 1075, 1076 (9th\\nCir. 1999).\\n\\n The clerk of this court shall forward the original and six\\ncertified copies of this order, under official seal, to the\\nSupreme Court of Arizona, along with copies of all briefs\\nand excerpts of record that have been filed with this court, as\\ndirected by Supreme Court of Arizona Rule 27.\\n\\nIV. Counsel\\n\\n For Plaintiff-Appellant Apollo Education Group, Inc.,\\nFKA Apollo Group, Inc., an Arizona Corporation:\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 17\\n\\n Douglas E. Whitney\\n Douglas Whitney Law Offices LLC\\n 321 North Clark Street, Suite 1301\\n Chicago, Illinois 60654\\n (312) 279-0510\\n\\n Mark J. DePasquale\\n Mark J. DePasquale, P.C.\\n 3300 North Central, Suite 2070\\n Phoenix, Arizona 85012\\n (602) 744-7777\\n\\n Mark S. Hersh\\n Reed Smith, LLP\\n Chicago, Illinois 60606\\n (312) 207-6400\\n\\n For Defendant-Appellee National Union Fire Insurance\\nCompany of Pittsburgh, PA., a Pennsylvania Corporation:\\n\\n Bennett Evan Cooper\\n Dickinson Wright\\n 1850 North Central Avenue, Suite 1400\\n Phoenix, Arizona 85004\\n (602) 285-5044\\n\\n Timothy M. Strong\\n Steptoe & Johnson LLP\\n Collier Center, Suite #1600\\n 201 E. Washington St.\\n Phoenix, Arizona 85004-2382\\n (602) 257-5219\\n\\f18 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nV. Filing fee\\n\\n The parties shall equally share the required filing fees\\npursuant to Supreme Court of Arizona Rule 27(a)(3)(D).\\n\\nVI. Stay of Proceedings\\n\\n In light of our decision to certify the issue described here,\\nall further proceedings in this case before our court are\\nstayed pending final action by the Supreme Court of\\nArizona, save for any petition for rehearing regarding the\\nmemorandum disposition filed concurrently with this order.\\nThe Clerk is directed to administratively close this docket\\npending further order. The parties shall notify the clerk of\\nthis court after the Supreme Court of Arizona accepts or\\nrejects certification, and again within fourteen days if the\\nSupreme Court of Arizona issues a decision.\\n\\n QUESTION CERTIFIED; PROCEEDINGS\\nSTAYED.\\n\\n It is so ORDERED.\\n\\n\\n\\n _____________________________\\n Richard A. Paez, Presiding Judge\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4425943/\", \"author_raw\": \"The Honorable Gary S. Katzmann, Judge for the United States\"}]}"},"judge1":{"kind":"string","value":"RICHARD A PAEZ"},"judge2":{"kind":"string","value":"RICHARD R CLIFTON"},"judge3":{"kind":"string","value":"GARY S KATZMANN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4648690/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Apollo Education Group, Inc. v. National Union Fire Insurance\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-15\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n APOLLO EDUCATION GROUP, No. 17-17293\n INC., FKA Apollo Group, Inc.,\n Plaintiff-Appellant, D.C. No.\n 2:15-cv-01948-SPL\n v.\n\n NATIONAL UNION FIRE ORDER\n INSURANCE COMPANY OF CERTIFYING\n PITTSBURGH, PA, a QUESTION TO THE\n Pennsylvania corporation, SUPREME COURT\n Defendant-Appellee. OF ARIZONA\n\n Filed August 15, 2019\n\n Before: Richard A. Paez and Richard R. Clifton, Circuit\n Judges, and Gary S. Katzmann, * Judge.\n\n Order\n\n\n\n\n *\n The Honorable Gary S. Katzmann, Judge for the United States\nCourt of International Trade, sitting by designation.\n\f2 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\n SUMMARY **\n\n\n Certified Question to the Supreme Court of Arizona\n\n The panel certified the following question of state law to\nthe Supreme Court of Arizona:\n\n What is the standard for determining whether\n National Union unreasonably withheld\n consent to Apollo’s settlement with\n shareholders in breach of contract under a\n policy where the insurer has no duty to\n defend?\n\n\n ORDER\n\n Apollo Education Group, Inc., (“Apollo”) has asserted\nbreach of contract and bad faith claims against its insurer,\nNational Union Fire Insurance Company (“National Union”)\nunder a governing insurance policy. National Union was\ncontractually bound not to unreasonably withhold consent to\nsettlement from its insured, Apollo. This appeal presents the\nquestion of the standard imposed by Arizona law on an\ninsurer’s obligation not to unreasonably withhold consent to\na settlement, where the insurer has no duty to defend.\nShould the federal district court assess the objective\nreasonableness of National Union’s decision to withhold\nconsent from the perspective of an insurer or an insured?\n\n\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 3\n\n As detailed below, the Teamsters Local 617 Pension &\nWelfare Funds (“Teamsters”) filed a securities class action\nlawsuit against Apollo in the U.S. District Court for the\nDistrict of Arizona. After five years of litigation, Apollo and\nthe Teamsters settled. National Union, as Apollo’s insurer,\ndeclined to consent to the settlement, and Apollo sued\nNational Union for breach of its contractual obligation not to\nunreasonably withhold consent. The district court held that\nbecause National Union reasonably considered the\nsettlement, Apollo’s breach of contract claim failed as a\nmatter of law. 1 1ER 4–5. 2\n\n Apollo now argues before this court that summary\njudgment was improper because the district court erred in\nconcluding that the insurer need only “consider[] the terms”\nof the settlement. Pl.’s Br. at 38. Apollo contends that under\nArizona law, an insurer must consent to an objectively\nreasonable settlement. Id. National Union instead frames\nthe district court’s decision as construing “the insurance\npolicy’s consent-to-settlement provision to focus on the\nobjective reasonableness of National Union’s decision to\nwithhold consent, rather than whether the settlement was\nreasonable from Apollo’s perspective.” Def.’s Br. at 30.\nThus, while both Apollo and National Union contend that\nthe reasonableness standard is “objective,” they dispute from\n\n\n 1\n The district court also held that Apollo’s claim that National Union\nbreached its duty to act in good faith failed as a matter of law. 1ER 5.\nBecause the district court’s analysis of this claim largely relied upon its\nbreach of contract analysis, we forgo addressing this claim pending\nresolution of this open question of Arizona state law on the breach of\ncontract claim.\n 2\n “ER” references are to the excerpts of record filed by the parties in\nthis court for the purposes of this appeal.\n\f4 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\nwhose perspective the district court should evaluate\nreasonableness.\n\n Because this issue is governed by Arizona law but is not\nclearly addressed by relevant Arizona authorities, we certify\nit to the Supreme Court of Arizona pursuant to the\nprocedures established by Arizona Revised Statutes § 12-\n1861 and Rule 27 of the Rules of the Supreme Court of\nArizona.\n\nI. Factual Background\n\n Pursuant to Supreme Court of Arizona Rule 27(a)(3)(B),\nwe set forth the facts relevant to the question certified.\n\n In 2005, Apollo purchased a $15 million directors and\nofficers (“D&O”) insurance policy (“the Policy”) from\nNational Union. In March 2006, the Wall Street Journal\npublished an article in which it identified the practice of\nbackdating stock options for corporate executives. 2ER 95–\n103. Backdating is particularly problematic when it results\nin companies under-reporting executive compensation to\ntheir shareholders. The article did not identify Apollo as one\nof the companies affected. Id.\n\n In June 2006, the U.S. Attorney’s Office for the Southern\nDistrict of New York issued a grand jury subpoena, and the\nSecurities and Exchange Commission issued a letter\nnotifying Apollo of its intent to investigate the company for\nbackdating. Neither resulted in charges or enforcement\nactivities. 8SER 1558; 2SER 71. Apollo simultaneously\nconducted its own internal investigation. Apollo\nsubsequently disclosed publicly that “57 of the 100 total\ngrants made during [the relevant] time period used incorrect\nmeasurement dates for accounting purposes.” 2ER 155. On\nOctober 18, 2006, after Apollo issued a statement admitting\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 5\n\n“various deficiencies in the process of granting and\ndocumenting stock options,” the Teamsters alleged Apollo’s\nstock price fell by 22.9% in one day. 4ER 543–44. Apollo’s\nfinal report, however, found no direct evidence of\nbackdating. 2ER 154.\n\n In November 2006, the Teamsters, as lead plaintiff, filed\na securities class action in the U.S. District Court for the\nDistrict of Arizona, alleging fraudulent misrepresentations\nand backdating by Apollo in violation of the Securities\nExchange Act of 1934. 3ER 290. In March 2009, the district\ncourt denied Apollo’s motion to dismiss the Teamsters’\ncomplaint, finding that the Teamsters “sufficiently pled\nbackdating.” 3ER 429–500, 460–62. In April 2009, the\nTeamsters filed an amended complaint that included 54\nallegedly false statements by Apollo and the reasons why the\nstatements were false. 4ER 524–42. In March 2011, the\ndistrict court concluded that the falsity allegations were not\npled with particularity, and thus, “[b]ecause the court has\nexpressly found that at least one of the elements of plaintiff’s\n[backdating] claim is missing, i.e. falsity, that claim fails,\nand there is no need for the court to consider defendants’ loss\ncausation and scienter arguments.” 4ER 602. The district\ncourt dismissed the complaint with prejudice and entered\nfinal judgment for Apollo. 4ER 604.\n\n The Teamsters filed a motion to reconsider and a\nproposed amended complaint. In its March 2012 order, the\ndistrict court denied the motion, reasoning that (1) the\nTeamsters had not met its burden to show that subsequent\ncase law warranted setting aside the judgment, 4ER 624; and\n(2) the Teamsters had not shown clear error sufficient to\njustify reopening a final judgment, 4ER 634.\n\n The Teamsters timely appealed the district court’s order\nto this court. 3ER 314. In January 2013, while the appeal\n\f6 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\nwas pending, Apollo, National Union, and Apollo’s excess\ninsurance carriers agreed to mediation with the Teamsters.\n2ER 27. The parties entered into a Mediation\nConfidentiality Agreement. 6ER 1075–77. After over a\nyear of mediation, Apollo agreed to settle with the Teamsters\nfor $13,125,000. 2ER 27. Apollo had $13,500,000\nremaining on the Policy with National Union. 2ER 25.\n\n Apollo’s counsel believed the settlement was “a great\ndeal” for Apollo, 5ER 959, and that the appeal of the\nbackdating litigation would be an “enormously expensive\ncase to defend” if the case was reversed by this court,\n5ER 823. Apollo feared the Teamsters could recover “a\nsubstantial multiple of insurance coverage” if successful on\nappeal. 5ER 803.\n\n Pursuant to the Policy, National Union had no duty to\ndefend Apollo. Instead, Apollo controlled its own defense\nand settlement negotiations. 2SER 75, 9SER 1838. The\nPolicy provides:\n\n The Insurer does not, however, under this\n policy, assume any duty to defend. The\n Insureds shall defend and contest any Claim\n made against them. The Insureds shall not\n admit or assume any liability, enter into\n settlement agreement, stipulate any\n judgment, or incur any Defense Costs\n without the prior written consent of the\n Insurer. Only those settlements, stipulated\n judgments and Defense Costs which have\n been consented to by the Insurer shall be\n recoverable as Loss under the terms of this\n policy. The Insurer’s consent shall not be\n unreasonably withheld, provided that the\n Insurer shall be entitled to effectively\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 7\n\n associate in the defense, the prosecution and\n the negotiation of any settlement of any\n Claim that involves or appears reasonably\n likely to involve the Insurer.\n\n2ER 48 (emphasis added).\n\n National Union declined to consent to or fund the\nsettlement. 4ER 770. Apollo nonetheless settled with the\nTeamsters, using its own funds. Id. Apollo then filed the\npresent action in district court to recover $13,125,000 under\nits Policy from National Union. Id. Apollo alleged breach\nof contract under the Policy and bad faith, arguing that\nNational Union unreasonably withheld consent to the\nsettlement. 2ER 23–33.\n\n In February 2017, National Union filed a motion for\nsummary judgment. ECF 74; 6ER 1137. National Union\nargued that summary judgment was appropriate on Apollo’s\nbreach of contract, bad faith, and punitive damages claims\nbecause National Union did not unreasonably withhold its\nconsent to the settlement. Id. Apollo argued against\nsummary judgment, contending that there were genuine\nissues of material fact regarding the reasonableness of its\nsettlement and National Union’s failure to consent. ECF 87;\n6ER 1138. In October 2017, the district court granted\nsummary judgment in favor of National Union. 3 ECF 104;\n6ER 1139.\n\n 3\n The district court also precluded Apollo from introducing\ndocuments protected by California’s mediation privilege. 1ER 7.\nApollo challenged the district court’s ruling on the admissibility of those\ndocuments in the present appeal, which we affirmed in a concurrently-\nfiled separate disposition.\n\f8 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\n The district court first found no genuine issue of material\nfact that National Union’s withholding of consent to\nsettlement was reasonable. 4 The district court stated that it\nconstrued the Policy “in accordance with [its] ‘plain and\nordinary meaning’” and “‘to effectuate the parties’ intent.’”\n1ER 3 (internal citations omitted). The district court noted\nthat “[i]n the context of insurance, questions of\nreasonableness can be resolved by the Court as a matter of\nlaw when the position of the parties turns on ‘the\ninterpretation of standard language in the form policies.’”\n1ER 3 (quoting Lennar Corp. v. Transamerica Ins. Co.,\n256 P.3d 635, 641 (Ariz. Ct. App. 2011). The district court\ndefined National Union’s obligation as “giv[ing] equal\nconsideration to both the interests of itself and of its\ninsured.” 1ER 4 (citing Bills v. Ariz. Prop. & Cas. Ins. Guar.\nFund, 984 P.2d 574, 578 (Ariz. Ct. App. 1999)). The district\ncourt then noted the undisputed steps National Union took to\nconsider the settlement and concluded that in light of\nNational Union’s “extensive analysis weighing the\nTeamsters’ settlement, taken in conjunction with the express\nterms of Plaintiff’s policy, it is clear that Defendant fulfilled\nits obligation to Plaintiff by considering the terms of the\nTeamsters’ settlement and that Plaintiff’s breach of contract\nclaim must fail as a matter of law.” 5 1ER 4.\n\n\n 4\n The district court also concluded that Apollo’s bad faith claim must\nfail as a matter of law because it “failed to set forth facts from which a\nreasonable juror could find that Defendant acted unreasonably, or knew\nthat it was acting unreasonably, when it refused to consent to Plaintiff’s\nsettlement with the Teamsters.” 1ER 5.\n 5\n The district court stated:\n\n Defendant concluded that settlement was premature –\n and likely unnecessary – because of the probability\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 9\n\n Apollo filed its Notice of Appeal to this court on\nNovember 10, 2017. We heard argument on April 8, 2019,\nand we issued a memorandum disposition on the issue of\nmediation privilege concurrently with this Certification\nOrder.\n\nII. Question of Law\n\n This case turns on Arizona state law. The basic question\nis: What is the standard for determining whether National\nUnion unreasonably withheld Apollo’s settlement with\nshareholders in breach of contract under a policy where the\ninsurer has no duty to defend? 6 For the reasons discussed\n\n that Plaintiff would prevail on appeal. Moreover,\n Defendant concluded that even in the unlikely event\n that Plaintiff did not prevail on appeal, there were a\n variety of other hurdles that Teamster would have to\n overcome in order to recover a “substantial judgment”\n against Plaintiff.\n\n1 ER 4.\n 6\n Windt’s insurance law treatise explains the conflict of interest that\nan insurer and insured may face in considering a settlement:\n\n When a claim against an insured can be settled for the\n policy limit, a conflict of interest exists between the\n insured and the insurance company. It is generally in\n the insured’s interest to settle for the policy limit, since\n doing so protects the insured from any potential\n personal liability. However, it is generally not in the\n insurance company’s interest to settle for the policy\n limit. Apart from considerations of defense costs, the\n insurance company cannot do any worse than paying\n its policy limit. As long as any chance existed, no\n matter how remote, that the insured might prevail, it\n would generally be in the insurance company's interest\n to litigate rather than settle for the policy limit.\n\f10 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\nbelow, Arizona case law does not allow us to reliably\ndetermine how the Supreme Court of Arizona would answer\nthis basic question of the standard for evaluating the\nreasonableness of National Union’s withholding of consent\nat issue here.\n\n In Arizona, as in other jurisdictions, judicial analysis of\nan insurance coverage dispute begins with the principle that:\n\n [a]n insurance policy is a contract, and in an\n action based thereon the terms of the policy\n must govern . . . [W]here the provisions of\n the contract are plain and unambiguous upon\n their face, they must be applied as written,\n and the court will not pervert or do violence\n to the language used, or expand it beyond its\n plain and ordinary meaning or add something\n to the contract which the parties have not put\n there.\n\nHome Indem. Co. v. Wilson, 489 P.2d 244, 247 (Ariz. 1971)\n(internal quotes and citations omitted). Here, in construing\nthe phrase “the Insurer’s consent [to settlement] shall not be\nunreasonably withheld,” the district court relied upon the\nCourt of Appeals of Arizona’s holding in Bills that the\n\n Moreover, liability insurance policies, by their terms,\n give insurers complete discretion in deciding whether\n to settle or litigate. In light of that fact, and the conflict\n of interest that can arise between the insurer and the\n insured in connection with whether a claim against an\n insured should be settled, the law has imposed on\n insurance companies an implied contractual obligation\n to settle under certain circumstances.\n\n1 Allan D. Windt, Insurance Claims and Disputes § 5:1 (6th ed. 2016).\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 11\n\ninsurer was obligated “to give equal consideration to both\nthe interests of itself and of its insured.” 1ER 4 (quoting\nBills, 984 P.2d at 578). The district court interpreted this to\nrequire that National Union “consider[] the terms and\nundert[ake] an assessment of a variety of factors” and\nconcluded that National Union “fulfilled its obligation to\n[Apollo] by considering the terms of the Teamsters’\nsettlement.” 1ER 4. In Bills, however, the court was plainly\nsetting forth the meaning of the insurer’s obligations under\n“the implied covenant of good faith and fair dealing,” not an\ninsurer’s obligations under a consent-to-settle provision.\n984 P.2d at 578. Bills, therefore, does not answer the\nquestion before us.\n\n Apollo relies primarily on two Arizona cases, United\nServices Automobile Association v. Morris, 741 P.2d 246\n(Ariz. 1987) and Himes v. Safeway Ins. Co., 66 P.3d 74\n(Ariz. Ct. App. 2003), to argue that the district court, in its\nbreach of contract analysis, should have evaluated the\nobjective reasonableness of the settlement, not the extent to\nwhich National Union may have considered settlement\nfactors. Pl.’s Br. at 44–45. In both cases, Arizona courts set\nforth an objective standard, focused on the merits of the\nunderlying claim, to determine whether a settlement was\nreasonable, and thus whether the insurer must accept it.\n\n In Morris, the insurer, United States Auto Insurance\n(“USAA”), had a duty to defend the insured homeowners in\na tort action. 741 P.2d 246. USAA defended the case but\nreserved its right to deny coverage, claiming the\nhomeowners may have committed intentional acts outside\nthe policy’s coverage. Id. at 249. The Supreme Court of\nArizona, however, noted that this reservation of rights put\nthe homeowners in a precarious position, incentivizing them\nto limit their personal liability should they not be covered by\n\f12 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\nthe policy. Id. at 251. USAA’s position, however, was that\nthe homeowners’ duty to cooperate meant USAA could\nforce the homeowners to “reject any settlement, no matter\nhow reasonable, risk trial, and place [the homeowners] at\ndanger of a judgment larger than the policy limits or one that\nmight not be covered.” Id.\n\n The Supreme Court of Arizona explained why USAA’s\ninterpretation of its right – to reject a settlement, no matter\nhow reasonable – was problematic. The Court noted that this\nwould “hamstring[] insureds while granting the insurer a\ndouble bite at escaping liability.” Id. “If the verdict were in\nfavor of [plaintiff] . . . USAA would have another chance at\nescaping the obligation to indemnify because it would be\nable to relitigate the intentional act exclusion coverage issue\n. . . In addition, absent bad faith conduct, USAA will never\nbe at risk for more than [the] . . . policy limit.” Id. On the\nother hand, the court observed that insureds have an\nincentive to settle for any amount that allows them to escape\npersonal liability – that is, any amount at or below the policy\nlimit – even if the amount is unreasonable and unwarranted.\nId. The Supreme Court of Arizona thus recognized the\nconflict between an insurer and insured’s interests,\nparticularly when a settlement may be at or near policy limits\nand the insured could face greater personal liability at trial.\n\n In response to this problem, the Supreme Court of\nArizona adopted the standard set forth by Minnesota’s\nSupreme Court in Miller v. Shugart, 316 N.W.2d 729 (Minn.\n1982), which held that “neither the fact nor amount of\nliability to the claimant is binding on the insurer unless the\ninsured or claimant can show that the settlement was\nreasonable and prudent,” in accordance with indemnification\nprinciples. Morris, 741 P.2d at 253 (quoting Miller,\n316 N.W.2d at 735). The Supreme Court of Arizona then\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 13\n\napplied the Miller test to determine “whether the settlement\nwas reasonable and prudent.” Id. at 254. The Court looked\nto “what a reasonably prudent person in the insureds’\nposition would have settled for on the merits of the\nclaimant’s case. This involves evaluating the facts bearing\non the liability and damage aspects of a claimant’s case, as\nwell as the risks of going to trial.” Id. (citing Miller,\n316 N.W. 2d at 735). Thus, if the Morris standard applies to\nthis case, the district court would look to whether it was\nobjectively reasonable for Apollo to reach the settlement it\ndid, not whether National Union thought it was reasonable\nor the steps it took to consider and evaluate the settlement.\n\n In Himes, the Court of Appeals of Arizona applied the\nreasonableness test set forth in Morris. 66 P.3d 74. There,\nan insured involved in an auto accident settled with the\naccident victim, giving the victim his breach of duty claims\nagainst the insurer in exchange for an agreement that the\nvictim would not execute against the insured’s personal\nassets. Id. at 87. The trial court judge held an evidentiary\nhearing and found the settlement reasonable. Id. at 78. The\ninsurer appealed. Id. at 79. The Himes court held that the\nburden of proving the settlement was reasonable was\nsquarely on the insured, but that the trial judge’s conclusion\nthat there was “no evidence that the agreement was\nunreasonable” was “plainly contradicted by the record.” Id.\nat 79–80.\n\n The Himes court reiterated the “reasonably prudent\nperson” standard adopted in Morris and enumerated factors\npotentially relevant to evaluating reasonableness:\n\n [T]he releasing person’s damages; the merits\n of the releasing person’s liability theory; the\n merits of the released person’s defense\n theory; the released person’s relative faults;\n\f14 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\n the risks and expenses of continued litigation\n [on the merits]; . . . any evidence of bad faith,\n collusion, or fraud; the extent of the releasing\n person’s investigation and preparation of the\n case; and the interests of the parties not being\n released.\n\nId. at 85 (quoting Chaussee v. Md. Cas. Co., 803 P.2d 1339,\n1343 (Wash. Ct. App. 1991)). The Himes court further noted\nthat, “it is only those costs, expenses, and risks that relate to\nthe merits of the underlying case that should be considered.\nThe costs, expenses, and risk of litigation for any coverage\nor bad faith litigation . . . should not be considered.” Id. The\nfocus should only be on the merits of the claimant’s case,\n“reflect[ing] what would occur in an arm’s-length\nnegotiation.” Id. The Himes court also noted that under\nMorris and related case law an insured’s ability to pay\nshould not be considered, except in punitive damages cases.\nId. The Himes court defined “reasonably prudent person” as\none who “has the ability to pay a reasonable settlement from\nhis or her own funds” to avoid the need to introduce evidence\nof the insured’s ability to pay into the reasonableness\nanalysis. Id. Lastly, the Himes court noted that its list of\nfactors was not exhaustive; courts should generally consider\nfactors which are “relevant in re-creating a genuine arm’s\nlength negotiation.” Id.\n\n National Union counters that unlike the policy before us,\nin Morris and Himes the insurer had a duty to defend the\ninsured; that is, these cases on which Apollo relies are\ninapposite because they do not involve “reimbursement\npolic[ies] under which the insurer has no duty to defend the\ninsured.” Def.’s Br. at 37. Both cases, moreover, do not\ninvolve the contract language, “insurer’s consent shall not be\nunreasonably withheld,” at issue here. National Union\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 15\n\ncontends that the consent to settle line of cases, involving\nthird-party claims, cannot apply because, “[u]nder Arizona\nlaw, a failure to reimburse the cost of a settlement entered by\nthe insured implicates a first-party claim for failure to pay\nbenefits, not a third-party claim for failure to defend the\ninsured.” Def.’s Br. at 37 (citing Trus Joist Corp. v. Safeco\nIns. Co. of Am., 735 P.2d 125, 135–36 (Ariz. Ct. App.\n1986)). This court cannot say whether the Supreme Court of\nArizona would apply the reasonable settlement standard in\nMorris and Himes here.\n\n The parties disagree fundamentally on the purpose of the\nconsent to settlement provision in the policy before us.\nNational Union, citing the Windt treatise, contends that\n“when an insured desires to settle a lawsuit that has been\nmade against it with an insurer’s money, the insured does not\nhave complete discretion to spend the insurer’s money.”\nDef.’s Br. at 31 (quoting 2 Allan D. Windt, Insurance Claims\n& Disputes § 6:29, at 6-293 n.1 (6th ed. 2016)). “Since the\ninsured wants to spend not its own money, but someone\nelse’s money, the issue is not whether the insured had a\nreasonable basis to pay a particular amount in settlement, but\nwhether the insurer had a reasonable basis not to agree to\npay that amount of money in settlement.” Id. (quoting Windt\nat 6-293 n.1) (emphasis added by National Union). Apollo,\nciting Morris, argues that the objectively reasonable\nsettlement rule is necessary to protect the insured because of\nthe inherent conflict that arises when an insured is presented\nwith a settlement opportunity at or near the policy limits and\nthe insurer has issued a reservation of rights and strong\nincentives to avoid a settlement because it can rely on both\nliability and coverage defenses and absent bad faith will\nnever face exposure beyond its policy limits. Pl.’s Br. at 44–\n47.\n\f16 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\n Though in disagreement as to how consent to settlement\nprovisions should be evaluated, the parties are in agreement\nas to their essential importance to the system of liability\ninsurance. We think that the resolution of that disagreement\nand consideration of an open question of Arizona\njurisprudence are better entrusted to the Supreme Court of\nArizona than to us.\n\nIII. Certified Question\n\n In light of the above discussion, we certify the following\nquestion of law: What is the standard for determining\nwhether National Union unreasonably withheld consent to\nApollo’s settlement with shareholders in breach of contract\nunder a policy where the insurer has no duty to defend? This\ncourt will accept the Arizona Supreme Court’s decision on\nthis question. Our phrasing of the question is not intended\nto restrict the Arizona Supreme Court’s consideration of the\ncase or formulation of the question. See Broad v.\nMannesmann Anglagenbau AG, 196 F.3d 1075, 1076 (9th\nCir. 1999).\n\n The clerk of this court shall forward the original and six\ncertified copies of this order, under official seal, to the\nSupreme Court of Arizona, along with copies of all briefs\nand excerpts of record that have been filed with this court, as\ndirected by Supreme Court of Arizona Rule 27.\n\nIV. Counsel\n\n For Plaintiff-Appellant Apollo Education Group, Inc.,\nFKA Apollo Group, Inc., an Arizona Corporation:\n\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 17\n\n Douglas E. Whitney\n Douglas Whitney Law Offices LLC\n 321 North Clark Street, Suite 1301\n Chicago, Illinois 60654\n (312) 279-0510\n\n Mark J. DePasquale\n Mark J. DePasquale, P.C.\n 3300 North Central, Suite 2070\n Phoenix, Arizona 85012\n (602) 744-7777\n\n Mark S. Hersh\n Reed Smith, LLP\n Chicago, Illinois 60606\n (312) 207-6400\n\n For Defendant-Appellee National Union Fire Insurance\nCompany of Pittsburgh, PA., a Pennsylvania Corporation:\n\n Bennett Evan Cooper\n Dickinson Wright\n 1850 North Central Avenue, Suite 1400\n Phoenix, Arizona 85004\n (602) 285-5044\n\n Timothy M. Strong\n Steptoe & Johnson LLP\n Collier Center, Suite #1600\n 201 E. Washington St.\n Phoenix, Arizona 85004-2382\n (602) 257-5219\n\f18 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\n\nV. Filing fee\n\n The parties shall equally share the required filing fees\npursuant to Supreme Court of Arizona Rule 27(a)(3)(D).\n\nVI. Stay of Proceedings\n\n In light of our decision to certify the issue described here,\nall further proceedings in this case before our court are\nstayed pending final action by the Supreme Court of\nArizona, save for any petition for rehearing regarding the\nmemorandum disposition filed concurrently with this order.\nThe Clerk is directed to administratively close this docket\npending further order. The parties shall notify the clerk of\nthis court after the Supreme Court of Arizona accepts or\nrejects certification, and again within fourteen days if the\nSupreme Court of Arizona issues a decision.\n\n QUESTION CERTIFIED; PROCEEDINGS\nSTAYED.\n\n It is so ORDERED.\n\n\n\n _____________________________\n Richard A. Paez, Presiding Judge\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Apollo Education Group, Inc. v. National Union Fire Insurance\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-15\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n APOLLO EDUCATION GROUP, No. 17-17293\\n INC., FKA Apollo Group, Inc.,\\n Plaintiff-Appellant, D.C. No.\\n 2:15-cv-01948-SPL\\n v.\\n\\n NATIONAL UNION FIRE ORDER\\n INSURANCE COMPANY OF CERTIFYING\\n PITTSBURGH, PA, a QUESTION TO THE\\n Pennsylvania corporation, SUPREME COURT\\n Defendant-Appellee. OF ARIZONA\\n\\n Filed August 15, 2019\\n\\n Before: Richard A. Paez and Richard R. Clifton, Circuit\\n Judges, and Gary S. Katzmann, * Judge.\\n\\n Order\\n\\n\\n\\n\\n *\\n The Honorable Gary S. Katzmann, Judge for the United States\\nCourt of International Trade, sitting by designation.\\n\\f2 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n SUMMARY **\\n\\n\\n Certified Question to the Supreme Court of Arizona\\n\\n The panel certified the following question of state law to\\nthe Supreme Court of Arizona:\\n\\n What is the standard for determining whether\\n National Union unreasonably withheld\\n consent to Apollo’s settlement with\\n shareholders in breach of contract under a\\n policy where the insurer has no duty to\\n defend?\\n\\n\\n ORDER\\n\\n Apollo Education Group, Inc., (“Apollo”) has asserted\\nbreach of contract and bad faith claims against its insurer,\\nNational Union Fire Insurance Company (“National Union”)\\nunder a governing insurance policy. National Union was\\ncontractually bound not to unreasonably withhold consent to\\nsettlement from its insured, Apollo. This appeal presents the\\nquestion of the standard imposed by Arizona law on an\\ninsurer’s obligation not to unreasonably withhold consent to\\na settlement, where the insurer has no duty to defend.\\nShould the federal district court assess the objective\\nreasonableness of National Union’s decision to withhold\\nconsent from the perspective of an insurer or an insured?\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 3\\n\\n As detailed below, the Teamsters Local 617 Pension &\\nWelfare Funds (“Teamsters”) filed a securities class action\\nlawsuit against Apollo in the U.S. District Court for the\\nDistrict of Arizona. After five years of litigation, Apollo and\\nthe Teamsters settled. National Union, as Apollo’s insurer,\\ndeclined to consent to the settlement, and Apollo sued\\nNational Union for breach of its contractual obligation not to\\nunreasonably withhold consent. The district court held that\\nbecause National Union reasonably considered the\\nsettlement, Apollo’s breach of contract claim failed as a\\nmatter of law. 1 1ER 4–5. 2\\n\\n Apollo now argues before this court that summary\\njudgment was improper because the district court erred in\\nconcluding that the insurer need only “consider[] the terms”\\nof the settlement. Pl.’s Br. at 38. Apollo contends that under\\nArizona law, an insurer must consent to an objectively\\nreasonable settlement. Id. National Union instead frames\\nthe district court’s decision as construing “the insurance\\npolicy’s consent-to-settlement provision to focus on the\\nobjective reasonableness of National Union’s decision to\\nwithhold consent, rather than whether the settlement was\\nreasonable from Apollo’s perspective.” Def.’s Br. at 30.\\nThus, while both Apollo and National Union contend that\\nthe reasonableness standard is “objective,” they dispute from\\n\\n\\n 1\\n The district court also held that Apollo’s claim that National Union\\nbreached its duty to act in good faith failed as a matter of law. 1ER 5.\\nBecause the district court’s analysis of this claim largely relied upon its\\nbreach of contract analysis, we forgo addressing this claim pending\\nresolution of this open question of Arizona state law on the breach of\\ncontract claim.\\n 2\\n “ER” references are to the excerpts of record filed by the parties in\\nthis court for the purposes of this appeal.\\n\\f4 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nwhose perspective the district court should evaluate\\nreasonableness.\\n\\n Because this issue is governed by Arizona law but is not\\nclearly addressed by relevant Arizona authorities, we certify\\nit to the Supreme Court of Arizona pursuant to the\\nprocedures established by Arizona Revised Statutes § 12-\\n1861 and Rule 27 of the Rules of the Supreme Court of\\nArizona.\\n\\nI. Factual Background\\n\\n Pursuant to Supreme Court of Arizona Rule 27(a)(3)(B),\\nwe set forth the facts relevant to the question certified.\\n\\n In 2005, Apollo purchased a $15 million directors and\\nofficers (“D&O”) insurance policy (“the Policy”) from\\nNational Union. In March 2006, the Wall Street Journal\\npublished an article in which it identified the practice of\\nbackdating stock options for corporate executives. 2ER 95–\\n103. Backdating is particularly problematic when it results\\nin companies under-reporting executive compensation to\\ntheir shareholders. The article did not identify Apollo as one\\nof the companies affected. Id.\\n\\n In June 2006, the U.S. Attorney’s Office for the Southern\\nDistrict of New York issued a grand jury subpoena, and the\\nSecurities and Exchange Commission issued a letter\\nnotifying Apollo of its intent to investigate the company for\\nbackdating. Neither resulted in charges or enforcement\\nactivities. 8SER 1558; 2SER 71. Apollo simultaneously\\nconducted its own internal investigation. Apollo\\nsubsequently disclosed publicly that “57 of the 100 total\\ngrants made during [the relevant] time period used incorrect\\nmeasurement dates for accounting purposes.” 2ER 155. On\\nOctober 18, 2006, after Apollo issued a statement admitting\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 5\\n\\n“various deficiencies in the process of granting and\\ndocumenting stock options,” the Teamsters alleged Apollo’s\\nstock price fell by 22.9% in one day. 4ER 543–44. Apollo’s\\nfinal report, however, found no direct evidence of\\nbackdating. 2ER 154.\\n\\n In November 2006, the Teamsters, as lead plaintiff, filed\\na securities class action in the U.S. District Court for the\\nDistrict of Arizona, alleging fraudulent misrepresentations\\nand backdating by Apollo in violation of the Securities\\nExchange Act of 1934. 3ER 290. In March 2009, the district\\ncourt denied Apollo’s motion to dismiss the Teamsters’\\ncomplaint, finding that the Teamsters “sufficiently pled\\nbackdating.” 3ER 429–500, 460–62. In April 2009, the\\nTeamsters filed an amended complaint that included 54\\nallegedly false statements by Apollo and the reasons why the\\nstatements were false. 4ER 524–42. In March 2011, the\\ndistrict court concluded that the falsity allegations were not\\npled with particularity, and thus, “[b]ecause the court has\\nexpressly found that at least one of the elements of plaintiff’s\\n[backdating] claim is missing, i.e. falsity, that claim fails,\\nand there is no need for the court to consider defendants’ loss\\ncausation and scienter arguments.” 4ER 602. The district\\ncourt dismissed the complaint with prejudice and entered\\nfinal judgment for Apollo. 4ER 604.\\n\\n The Teamsters filed a motion to reconsider and a\\nproposed amended complaint. In its March 2012 order, the\\ndistrict court denied the motion, reasoning that (1) the\\nTeamsters had not met its burden to show that subsequent\\ncase law warranted setting aside the judgment, 4ER 624; and\\n(2) the Teamsters had not shown clear error sufficient to\\njustify reopening a final judgment, 4ER 634.\\n\\n The Teamsters timely appealed the district court’s order\\nto this court. 3ER 314. In January 2013, while the appeal\\n\\f6 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nwas pending, Apollo, National Union, and Apollo’s excess\\ninsurance carriers agreed to mediation with the Teamsters.\\n2ER 27. The parties entered into a Mediation\\nConfidentiality Agreement. 6ER 1075–77. After over a\\nyear of mediation, Apollo agreed to settle with the Teamsters\\nfor $13,125,000. 2ER 27. Apollo had $13,500,000\\nremaining on the Policy with National Union. 2ER 25.\\n\\n Apollo’s counsel believed the settlement was “a great\\ndeal” for Apollo, 5ER 959, and that the appeal of the\\nbackdating litigation would be an “enormously expensive\\ncase to defend” if the case was reversed by this court,\\n5ER 823. Apollo feared the Teamsters could recover “a\\nsubstantial multiple of insurance coverage” if successful on\\nappeal. 5ER 803.\\n\\n Pursuant to the Policy, National Union had no duty to\\ndefend Apollo. Instead, Apollo controlled its own defense\\nand settlement negotiations. 2SER 75, 9SER 1838. The\\nPolicy provides:\\n\\n The Insurer does not, however, under this\\n policy, assume any duty to defend. The\\n Insureds shall defend and contest any Claim\\n made against them. The Insureds shall not\\n admit or assume any liability, enter into\\n settlement agreement, stipulate any\\n judgment, or incur any Defense Costs\\n without the prior written consent of the\\n Insurer. Only those settlements, stipulated\\n judgments and Defense Costs which have\\n been consented to by the Insurer shall be\\n recoverable as Loss under the terms of this\\n policy. The Insurer’s consent shall not be\\n unreasonably withheld, provided that the\\n Insurer shall be entitled to effectively\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 7\\n\\n associate in the defense, the prosecution and\\n the negotiation of any settlement of any\\n Claim that involves or appears reasonably\\n likely to involve the Insurer.\\n\\n2ER 48 (emphasis added).\\n\\n National Union declined to consent to or fund the\\nsettlement. 4ER 770. Apollo nonetheless settled with the\\nTeamsters, using its own funds. Id. Apollo then filed the\\npresent action in district court to recover $13,125,000 under\\nits Policy from National Union. Id. Apollo alleged breach\\nof contract under the Policy and bad faith, arguing that\\nNational Union unreasonably withheld consent to the\\nsettlement. 2ER 23–33.\\n\\n In February 2017, National Union filed a motion for\\nsummary judgment. ECF 74; 6ER 1137. National Union\\nargued that summary judgment was appropriate on Apollo’s\\nbreach of contract, bad faith, and punitive damages claims\\nbecause National Union did not unreasonably withhold its\\nconsent to the settlement. Id. Apollo argued against\\nsummary judgment, contending that there were genuine\\nissues of material fact regarding the reasonableness of its\\nsettlement and National Union’s failure to consent. ECF 87;\\n6ER 1138. In October 2017, the district court granted\\nsummary judgment in favor of National Union. 3 ECF 104;\\n6ER 1139.\\n\\n 3\\n The district court also precluded Apollo from introducing\\ndocuments protected by California’s mediation privilege. 1ER 7.\\nApollo challenged the district court’s ruling on the admissibility of those\\ndocuments in the present appeal, which we affirmed in a concurrently-\\nfiled separate disposition.\\n\\f8 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n The district court first found no genuine issue of material\\nfact that National Union’s withholding of consent to\\nsettlement was reasonable. 4 The district court stated that it\\nconstrued the Policy “in accordance with [its] ‘plain and\\nordinary meaning’” and “‘to effectuate the parties’ intent.’”\\n1ER 3 (internal citations omitted). The district court noted\\nthat “[i]n the context of insurance, questions of\\nreasonableness can be resolved by the Court as a matter of\\nlaw when the position of the parties turns on ‘the\\ninterpretation of standard language in the form policies.’”\\n1ER 3 (quoting Lennar Corp. v. Transamerica Ins. Co.,\\n256 P.3d 635, 641 (Ariz. Ct. App. 2011). The district court\\ndefined National Union’s obligation as “giv[ing] equal\\nconsideration to both the interests of itself and of its\\ninsured.” 1ER 4 (citing Bills v. Ariz. Prop. & Cas. Ins. Guar.\\nFund, 984 P.2d 574, 578 (Ariz. Ct. App. 1999)). The district\\ncourt then noted the undisputed steps National Union took to\\nconsider the settlement and concluded that in light of\\nNational Union’s “extensive analysis weighing the\\nTeamsters’ settlement, taken in conjunction with the express\\nterms of Plaintiff’s policy, it is clear that Defendant fulfilled\\nits obligation to Plaintiff by considering the terms of the\\nTeamsters’ settlement and that Plaintiff’s breach of contract\\nclaim must fail as a matter of law.” 5 1ER 4.\\n\\n\\n 4\\n The district court also concluded that Apollo’s bad faith claim must\\nfail as a matter of law because it “failed to set forth facts from which a\\nreasonable juror could find that Defendant acted unreasonably, or knew\\nthat it was acting unreasonably, when it refused to consent to Plaintiff’s\\nsettlement with the Teamsters.” 1ER 5.\\n 5\\n The district court stated:\\n\\n Defendant concluded that settlement was premature –\\n and likely unnecessary – because of the probability\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 9\\n\\n Apollo filed its Notice of Appeal to this court on\\nNovember 10, 2017. We heard argument on April 8, 2019,\\nand we issued a memorandum disposition on the issue of\\nmediation privilege concurrently with this Certification\\nOrder.\\n\\nII. Question of Law\\n\\n This case turns on Arizona state law. The basic question\\nis: What is the standard for determining whether National\\nUnion unreasonably withheld Apollo’s settlement with\\nshareholders in breach of contract under a policy where the\\ninsurer has no duty to defend? 6 For the reasons discussed\\n\\n that Plaintiff would prevail on appeal. Moreover,\\n Defendant concluded that even in the unlikely event\\n that Plaintiff did not prevail on appeal, there were a\\n variety of other hurdles that Teamster would have to\\n overcome in order to recover a “substantial judgment”\\n against Plaintiff.\\n\\n1 ER 4.\\n 6\\n Windt’s insurance law treatise explains the conflict of interest that\\nan insurer and insured may face in considering a settlement:\\n\\n When a claim against an insured can be settled for the\\n policy limit, a conflict of interest exists between the\\n insured and the insurance company. It is generally in\\n the insured’s interest to settle for the policy limit, since\\n doing so protects the insured from any potential\\n personal liability. However, it is generally not in the\\n insurance company’s interest to settle for the policy\\n limit. Apart from considerations of defense costs, the\\n insurance company cannot do any worse than paying\\n its policy limit. As long as any chance existed, no\\n matter how remote, that the insured might prevail, it\\n would generally be in the insurance company's interest\\n to litigate rather than settle for the policy limit.\\n\\f10 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nbelow, Arizona case law does not allow us to reliably\\ndetermine how the Supreme Court of Arizona would answer\\nthis basic question of the standard for evaluating the\\nreasonableness of National Union’s withholding of consent\\nat issue here.\\n\\n In Arizona, as in other jurisdictions, judicial analysis of\\nan insurance coverage dispute begins with the principle that:\\n\\n [a]n insurance policy is a contract, and in an\\n action based thereon the terms of the policy\\n must govern . . . [W]here the provisions of\\n the contract are plain and unambiguous upon\\n their face, they must be applied as written,\\n and the court will not pervert or do violence\\n to the language used, or expand it beyond its\\n plain and ordinary meaning or add something\\n to the contract which the parties have not put\\n there.\\n\\nHome Indem. Co. v. Wilson, 489 P.2d 244, 247 (Ariz. 1971)\\n(internal quotes and citations omitted). Here, in construing\\nthe phrase “the Insurer’s consent [to settlement] shall not be\\nunreasonably withheld,” the district court relied upon the\\nCourt of Appeals of Arizona’s holding in Bills that the\\n\\n Moreover, liability insurance policies, by their terms,\\n give insurers complete discretion in deciding whether\\n to settle or litigate. In light of that fact, and the conflict\\n of interest that can arise between the insurer and the\\n insured in connection with whether a claim against an\\n insured should be settled, the law has imposed on\\n insurance companies an implied contractual obligation\\n to settle under certain circumstances.\\n\\n1 Allan D. Windt, Insurance Claims and Disputes § 5:1 (6th ed. 2016).\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 11\\n\\ninsurer was obligated “to give equal consideration to both\\nthe interests of itself and of its insured.” 1ER 4 (quoting\\nBills, 984 P.2d at 578). The district court interpreted this to\\nrequire that National Union “consider[] the terms and\\nundert[ake] an assessment of a variety of factors” and\\nconcluded that National Union “fulfilled its obligation to\\n[Apollo] by considering the terms of the Teamsters’\\nsettlement.” 1ER 4. In Bills, however, the court was plainly\\nsetting forth the meaning of the insurer’s obligations under\\n“the implied covenant of good faith and fair dealing,” not an\\ninsurer’s obligations under a consent-to-settle provision.\\n984 P.2d at 578. Bills, therefore, does not answer the\\nquestion before us.\\n\\n Apollo relies primarily on two Arizona cases, United\\nServices Automobile Association v. Morris, 741 P.2d 246\\n(Ariz. 1987) and Himes v. Safeway Ins. Co., 66 P.3d 74\\n(Ariz. Ct. App. 2003), to argue that the district court, in its\\nbreach of contract analysis, should have evaluated the\\nobjective reasonableness of the settlement, not the extent to\\nwhich National Union may have considered settlement\\nfactors. Pl.’s Br. at 44–45. In both cases, Arizona courts set\\nforth an objective standard, focused on the merits of the\\nunderlying claim, to determine whether a settlement was\\nreasonable, and thus whether the insurer must accept it.\\n\\n In Morris, the insurer, United States Auto Insurance\\n(“USAA”), had a duty to defend the insured homeowners in\\na tort action. 741 P.2d 246. USAA defended the case but\\nreserved its right to deny coverage, claiming the\\nhomeowners may have committed intentional acts outside\\nthe policy’s coverage. Id. at 249. The Supreme Court of\\nArizona, however, noted that this reservation of rights put\\nthe homeowners in a precarious position, incentivizing them\\nto limit their personal liability should they not be covered by\\n\\f12 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nthe policy. Id. at 251. USAA’s position, however, was that\\nthe homeowners’ duty to cooperate meant USAA could\\nforce the homeowners to “reject any settlement, no matter\\nhow reasonable, risk trial, and place [the homeowners] at\\ndanger of a judgment larger than the policy limits or one that\\nmight not be covered.” Id.\\n\\n The Supreme Court of Arizona explained why USAA’s\\ninterpretation of its right – to reject a settlement, no matter\\nhow reasonable – was problematic. The Court noted that this\\nwould “hamstring[] insureds while granting the insurer a\\ndouble bite at escaping liability.” Id. “If the verdict were in\\nfavor of [plaintiff] . . . USAA would have another chance at\\nescaping the obligation to indemnify because it would be\\nable to relitigate the intentional act exclusion coverage issue\\n. . . In addition, absent bad faith conduct, USAA will never\\nbe at risk for more than [the] . . . policy limit.” Id. On the\\nother hand, the court observed that insureds have an\\nincentive to settle for any amount that allows them to escape\\npersonal liability – that is, any amount at or below the policy\\nlimit – even if the amount is unreasonable and unwarranted.\\nId. The Supreme Court of Arizona thus recognized the\\nconflict between an insurer and insured’s interests,\\nparticularly when a settlement may be at or near policy limits\\nand the insured could face greater personal liability at trial.\\n\\n In response to this problem, the Supreme Court of\\nArizona adopted the standard set forth by Minnesota’s\\nSupreme Court in Miller v. Shugart, 316 N.W.2d 729 (Minn.\\n1982), which held that “neither the fact nor amount of\\nliability to the claimant is binding on the insurer unless the\\ninsured or claimant can show that the settlement was\\nreasonable and prudent,” in accordance with indemnification\\nprinciples. Morris, 741 P.2d at 253 (quoting Miller,\\n316 N.W.2d at 735). The Supreme Court of Arizona then\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 13\\n\\napplied the Miller test to determine “whether the settlement\\nwas reasonable and prudent.” Id. at 254. The Court looked\\nto “what a reasonably prudent person in the insureds’\\nposition would have settled for on the merits of the\\nclaimant’s case. This involves evaluating the facts bearing\\non the liability and damage aspects of a claimant’s case, as\\nwell as the risks of going to trial.” Id. (citing Miller,\\n316 N.W. 2d at 735). Thus, if the Morris standard applies to\\nthis case, the district court would look to whether it was\\nobjectively reasonable for Apollo to reach the settlement it\\ndid, not whether National Union thought it was reasonable\\nor the steps it took to consider and evaluate the settlement.\\n\\n In Himes, the Court of Appeals of Arizona applied the\\nreasonableness test set forth in Morris. 66 P.3d 74. There,\\nan insured involved in an auto accident settled with the\\naccident victim, giving the victim his breach of duty claims\\nagainst the insurer in exchange for an agreement that the\\nvictim would not execute against the insured’s personal\\nassets. Id. at 87. The trial court judge held an evidentiary\\nhearing and found the settlement reasonable. Id. at 78. The\\ninsurer appealed. Id. at 79. The Himes court held that the\\nburden of proving the settlement was reasonable was\\nsquarely on the insured, but that the trial judge’s conclusion\\nthat there was “no evidence that the agreement was\\nunreasonable” was “plainly contradicted by the record.” Id.\\nat 79–80.\\n\\n The Himes court reiterated the “reasonably prudent\\nperson” standard adopted in Morris and enumerated factors\\npotentially relevant to evaluating reasonableness:\\n\\n [T]he releasing person’s damages; the merits\\n of the releasing person’s liability theory; the\\n merits of the released person’s defense\\n theory; the released person’s relative faults;\\n\\f14 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n the risks and expenses of continued litigation\\n [on the merits]; . . . any evidence of bad faith,\\n collusion, or fraud; the extent of the releasing\\n person’s investigation and preparation of the\\n case; and the interests of the parties not being\\n released.\\n\\nId. at 85 (quoting Chaussee v. Md. Cas. Co., 803 P.2d 1339,\\n1343 (Wash. Ct. App. 1991)). The Himes court further noted\\nthat, “it is only those costs, expenses, and risks that relate to\\nthe merits of the underlying case that should be considered.\\nThe costs, expenses, and risk of litigation for any coverage\\nor bad faith litigation . . . should not be considered.” Id. The\\nfocus should only be on the merits of the claimant’s case,\\n“reflect[ing] what would occur in an arm’s-length\\nnegotiation.” Id. The Himes court also noted that under\\nMorris and related case law an insured’s ability to pay\\nshould not be considered, except in punitive damages cases.\\nId. The Himes court defined “reasonably prudent person” as\\none who “has the ability to pay a reasonable settlement from\\nhis or her own funds” to avoid the need to introduce evidence\\nof the insured’s ability to pay into the reasonableness\\nanalysis. Id. Lastly, the Himes court noted that its list of\\nfactors was not exhaustive; courts should generally consider\\nfactors which are “relevant in re-creating a genuine arm’s\\nlength negotiation.” Id.\\n\\n National Union counters that unlike the policy before us,\\nin Morris and Himes the insurer had a duty to defend the\\ninsured; that is, these cases on which Apollo relies are\\ninapposite because they do not involve “reimbursement\\npolic[ies] under which the insurer has no duty to defend the\\ninsured.” Def.’s Br. at 37. Both cases, moreover, do not\\ninvolve the contract language, “insurer’s consent shall not be\\nunreasonably withheld,” at issue here. National Union\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 15\\n\\ncontends that the consent to settle line of cases, involving\\nthird-party claims, cannot apply because, “[u]nder Arizona\\nlaw, a failure to reimburse the cost of a settlement entered by\\nthe insured implicates a first-party claim for failure to pay\\nbenefits, not a third-party claim for failure to defend the\\ninsured.” Def.’s Br. at 37 (citing Trus Joist Corp. v. Safeco\\nIns. Co. of Am., 735 P.2d 125, 135–36 (Ariz. Ct. App.\\n1986)). This court cannot say whether the Supreme Court of\\nArizona would apply the reasonable settlement standard in\\nMorris and Himes here.\\n\\n The parties disagree fundamentally on the purpose of the\\nconsent to settlement provision in the policy before us.\\nNational Union, citing the Windt treatise, contends that\\n“when an insured desires to settle a lawsuit that has been\\nmade against it with an insurer’s money, the insured does not\\nhave complete discretion to spend the insurer’s money.”\\nDef.’s Br. at 31 (quoting 2 Allan D. Windt, Insurance Claims\\n& Disputes § 6:29, at 6-293 n.1 (6th ed. 2016)). “Since the\\ninsured wants to spend not its own money, but someone\\nelse’s money, the issue is not whether the insured had a\\nreasonable basis to pay a particular amount in settlement, but\\nwhether the insurer had a reasonable basis not to agree to\\npay that amount of money in settlement.” Id. (quoting Windt\\nat 6-293 n.1) (emphasis added by National Union). Apollo,\\nciting Morris, argues that the objectively reasonable\\nsettlement rule is necessary to protect the insured because of\\nthe inherent conflict that arises when an insured is presented\\nwith a settlement opportunity at or near the policy limits and\\nthe insurer has issued a reservation of rights and strong\\nincentives to avoid a settlement because it can rely on both\\nliability and coverage defenses and absent bad faith will\\nnever face exposure beyond its policy limits. Pl.’s Br. at 44–\\n47.\\n\\f16 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\n Though in disagreement as to how consent to settlement\\nprovisions should be evaluated, the parties are in agreement\\nas to their essential importance to the system of liability\\ninsurance. We think that the resolution of that disagreement\\nand consideration of an open question of Arizona\\njurisprudence are better entrusted to the Supreme Court of\\nArizona than to us.\\n\\nIII. Certified Question\\n\\n In light of the above discussion, we certify the following\\nquestion of law: What is the standard for determining\\nwhether National Union unreasonably withheld consent to\\nApollo’s settlement with shareholders in breach of contract\\nunder a policy where the insurer has no duty to defend? This\\ncourt will accept the Arizona Supreme Court’s decision on\\nthis question. Our phrasing of the question is not intended\\nto restrict the Arizona Supreme Court’s consideration of the\\ncase or formulation of the question. See Broad v.\\nMannesmann Anglagenbau AG, 196 F.3d 1075, 1076 (9th\\nCir. 1999).\\n\\n The clerk of this court shall forward the original and six\\ncertified copies of this order, under official seal, to the\\nSupreme Court of Arizona, along with copies of all briefs\\nand excerpts of record that have been filed with this court, as\\ndirected by Supreme Court of Arizona Rule 27.\\n\\nIV. Counsel\\n\\n For Plaintiff-Appellant Apollo Education Group, Inc.,\\nFKA Apollo Group, Inc., an Arizona Corporation:\\n\\f APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO. 17\\n\\n Douglas E. Whitney\\n Douglas Whitney Law Offices LLC\\n 321 North Clark Street, Suite 1301\\n Chicago, Illinois 60654\\n (312) 279-0510\\n\\n Mark J. DePasquale\\n Mark J. DePasquale, P.C.\\n 3300 North Central, Suite 2070\\n Phoenix, Arizona 85012\\n (602) 744-7777\\n\\n Mark S. Hersh\\n Reed Smith, LLP\\n Chicago, Illinois 60606\\n (312) 207-6400\\n\\n For Defendant-Appellee National Union Fire Insurance\\nCompany of Pittsburgh, PA., a Pennsylvania Corporation:\\n\\n Bennett Evan Cooper\\n Dickinson Wright\\n 1850 North Central Avenue, Suite 1400\\n Phoenix, Arizona 85004\\n (602) 285-5044\\n\\n Timothy M. Strong\\n Steptoe & Johnson LLP\\n Collier Center, Suite #1600\\n 201 E. Washington St.\\n Phoenix, Arizona 85004-2382\\n (602) 257-5219\\n\\f18 APOLLO EDUC. GRP. V. NAT’L UNION FIRE INS. CO.\\n\\nV. Filing fee\\n\\n The parties shall equally share the required filing fees\\npursuant to Supreme Court of Arizona Rule 27(a)(3)(D).\\n\\nVI. Stay of Proceedings\\n\\n In light of our decision to certify the issue described here,\\nall further proceedings in this case before our court are\\nstayed pending final action by the Supreme Court of\\nArizona, save for any petition for rehearing regarding the\\nmemorandum disposition filed concurrently with this order.\\nThe Clerk is directed to administratively close this docket\\npending further order. The parties shall notify the clerk of\\nthis court after the Supreme Court of Arizona accepts or\\nrejects certification, and again within fourteen days if the\\nSupreme Court of Arizona issues a decision.\\n\\n QUESTION CERTIFIED; PROCEEDINGS\\nSTAYED.\\n\\n It is so ORDERED.\\n\\n\\n\\n _____________________________\\n Richard A. Paez, Presiding Judge\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4071,"cells":{"id":{"kind":"number","value":4649090,"string":"4,649,090"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"amazon.com Inc. & Subsidiaries v. Cir"},"decision_date":{"kind":"string","value":"2019-08-16"},"docket_number":{"kind":"string","value":"17-72922"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: William A. Fletcher, Consuelo M. Callahan, and Morgan Christen, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"CALLAHAN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nAMAZON.COM, INC. & SUBSIDIARIES, No. 17-72922\\n Petitioner-Appellee,\\n Tax Ct. No.\\n v. 31197-12\\n\\nCOMMISSIONER OF INTERNAL\\nREVENUE, OPINION\\n Respondent-Appellant.\\n\\n Appeal from a Decision of the\\n United States Tax Court\\n\\n Argued and Submitted April 12, 2019\\n Seattle, Washington\\n\\n Filed August 16, 2019\\n\\n Before: William A. Fletcher, Consuelo M. Callahan,\\n and Morgan Christen, Circuit Judges.\\n\\n Opinion by Judge Callahan\\n\\f2 AMAZON.COM V. CIR\\n\\n SUMMARY *\\n\\n\\n Tax\\n\\n The panel affirmed the Tax Court’s decision on a petition\\nfor redetermination of federal income tax deficiencies, in an\\nappeal involving the regulatory definition of intangible\\nassets and the method of their valuation in a cost-sharing\\narrangement.\\n\\n In the course of restructuring its European businesses in\\na way that would shift a substantial amount of income from\\nU.S.-based entities to the European subsidiaries, appellee\\nAmazon.com, Inc. entered into a cost sharing arrangement\\nin which a holding company for the European subsidiaries\\nmade a “buy-in” payment for Amazon’s assets that met the\\nregulatory definition of an “intangible.” See 26 U.S.C.\\n§ 482. Tax regulations required that the buy-in payment\\nreflect the fair market value of Amazon’s pre-existing\\nintangibles. After the Commissioner of Internal Revenue\\nconcluded that the buy-in payment had not been determined\\nat arm’s length in accordance with the transfer pricing\\nregulations, the Internal Revenue Service performed its own\\ncalculation, and Amazon filed a petition in the Tax Court\\nchallenging that valuation.\\n\\n At issue is the correct method for valuing the pre-\\nexisting intangibles under the then-applicable transfer\\npricing regulations. The Commissioner sought to include all\\nintangible assets of value, including “residual-business\\nassets” such as Amazon’s culture of innovcation, the value\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f AMAZON.COM V. CIR 3\\n\\nof workforce in place, going concern value, goodwill, and\\ngrowth options. The panel concluded that the definition of\\n“intangible” does not include residual-business assets, and\\nthat the definition is limited to independently transferrable\\nassets.\\n\\n\\n COUNSEL\\n\\nJudith A. Hagley (argued), Gilbert S. Rothenberg, and\\nArthur T. Catterall, Attorneys, Tax Division; Richard E.\\nZuckerman, Principal Deputy Assistant Attorney General;\\nTravis A. Greaves, Deputy Assistant Attorney General; Tax\\nDivision, United States Department of Justice, Washington,\\nD.C.; for Respondent-Appellant.\\n\\nCarter G. Phillips (argued), Joseph R. Guerra, and Matthew\\nD. Lerner, Sidley Austin LLP, Washington, D.C.; David R.\\nCarpenter, Sidley Austin LLP, Los Angeles, California; for\\nPetitioner-Appellee.\\n\\nChristopher J. Walker, The Ohio State University Moritz\\nCollege of Law, Columbus, Ohio; Steven P. Lehotsky, U.S.\\nChamber Litigation Center, Washington, D.C.; for Amicus\\nCuriae The Chamber of Commerce of the United States of\\nAmerica.\\n\\nElizabeth J. Stevens, Caplin & Drysdale, Chartered,\\nWashington, D.C., for Amici Curiae H. David Rosenbloom\\nand John P. Steines, Jr.\\n\\nAlice E. Loughran, Charles G. Cole, Michael C. Durst, and\\nRobert J. Kovacev, Steptoe & Johnson LLP, Washington\\nD.C., for Amici Curiae Semiconductor Industry Association,\\nInformation Technology Industry Council, National Foreign\\n\\f4 AMAZON.COM V. CIR\\n\\nTrade Council, Software Finance and Tax Executives\\nCouncil, and TechNet.\\n\\nA. Duane Webber, George M. Clarke, and Phillip J. Taylor,\\nBaker & McKenzie LLP, Washington, D.C.; Mark A. Oates,\\nSusan E. Ryba, and Cameron C. Reilly, Baker & McKenzie\\nLLP, Chicago, Illinois; Scott H. Frewing, Baker &\\nMcKenzie LLP, Palo Alto, California; for Amici Curiae\\nSilicon Valley Tax Directors Group, Agilent Technologies,\\nInc., Cisco Systems, Inc., Dell Technologies Inc., Dolby\\nLaboratories, Inc., Expedia Group, Inc., FireEye, Inc.,\\nGenesys Int’l Corp. Ltd., Informatica, Inc., NetApp, Inc.,\\nPalo Alto Networks, Inc., Surveymonkey Inc., and\\nVMware, Inc.\\n\\n\\n OPINION\\n\\nCALLAHAN, Circuit Judge:\\n\\n Appellee, Amazon.com, Inc., is a U.S.-based online\\nretailer with highly profitable intangible assets. In 2005 and\\n2006, Amazon restructured its European businesses in a way\\nthat would shift a substantial amount of its income from\\nU.S.-based entities to newly created European subsidiaries.\\nBecause the restructuring would allow the European entities\\nto generate income using Amazon’s pre-existing intangible\\nassets developed in the United States, the tax code and\\ncorresponding regulations required that the European\\nentities compensate Amazon for the use of assets that meet\\nthe regulatory definition of an “intangible.”\\n\\n The compensation was provided through a cost sharing\\narrangement, whereby Amazon and a holding company for\\nthe European subsidiaries would be treated as co-owners of\\n\\f AMAZON.COM V. CIR 5\\n\\nthe intangibles. Under the arrangement, the holding\\ncompany was required to make a “buy-in” payment for the\\npre-existing intangibles Amazon contributed to the\\narrangement and to make cost sharing payments going\\nforward for its share of future research and development\\n(R&D) efforts. The buy-in payment was taxable income to\\nAmazon, and the holding company’s cost sharing payments\\nwould reduce Amazon’s U.S. tax deductions for R&D costs.\\n\\n To guard against manipulation by jointly controlled\\nentities, the regulations require that the buy-in payment\\nreflect the fair market value of the pre-existing intangibles\\nmade available under a cost sharing arrangement. Amazon\\ninitially reported a buy-in payment of about $255 million.\\nAppellant, the Commissioner of Internal Revenue,\\nconcluded that the buy-in payment had not been determined\\nat arm’s length in accordance with the transfer pricing\\nregulations, so the IRS performed its own calculation,\\nvaluing the buy-in at about $3.6 billion. Amazon filed a\\npetition in the United States Tax Court challenging the IRS’s\\nvaluation.\\n\\n In the tax court proceedings, Amazon and the\\nCommissioner offered competing methods for valuing\\nAmazon’s pre-existing intangibles. There was a key\\ndifference between the parties’ respective approaches.\\nAmazon’s methodology isolated and valued only the\\nspecific intangible assets that it transferred to the European\\nholding company under the cost sharing arrangement,\\nincluding website technology, trademarks, and customer\\nlists. The Commissioner’s methodology essentially valued\\nthe entire European business, minus pre-existing tangible\\nassets. That method necessarily swept into the calculation\\nall contributions of value, including those that are more\\nnebulous and inseparable from the business itself, like the\\n\\f6 AMAZON.COM V. CIR\\n\\nvalue of employees’ experience, education, and training\\n(known as “workforce in place”), going concern value,\\ngoodwill, and other unique business attributes and\\nexpectancies (which the parties refer to as “growth\\noptions”). The tax court sided primarily with Amazon, and\\nthe Commissioner appealed.\\n\\n This case requires us to interpret the meaning of an\\n“intangible” in the applicable (but now outdated) transfer\\npricing regulations. 1 The case turns on whether, as the\\nCommissioner argues, the regulatory definition is broad\\nenough to include all intangible assets of value, even the\\nmore nebulous ones that the Commissioner refers to as\\n“residual-business assets” (i.e., Amazon’s culture of\\ninnovation, the value of workforce in place, going concern\\nvalue, goodwill, and growth options). We conclude that the\\ndefinition does not include residual-business assets.\\nAlthough the language of the definition is ambiguous, the\\ndrafting history of the regulations shows that “intangible”\\nwas understood to be limited to independently transferrable\\nassets. We thus affirm.\\n\\n\\n\\n\\n 1\\n This case is governed by regulations promulgated in 1994 and\\n1995. In 2009, more than three years after the tax years at issue here, the\\nDepartment of Treasury issued temporary regulations broadening the\\nscope of contributions for which compensation must be made as part of\\nthe buy-in payment. See 74 Fed. Reg. 340 (Jan. 5, 2009). In 2017,\\nCongress amended the definition of “intangible property” in 26 U.S.C.\\n§ 936(h)(3)(B) (which is incorporated by reference in 26 U.S.C. § 482).\\nTax Cuts and Jobs Act of 2017, Pub. L. 115-97, § 14221(a), 131 Stat.\\n2054, 2218 (2017). If this case were governed by the 2009 regulations\\nor by the 2017 statutory amendment, there is no doubt the\\nCommissioner’s position would be correct.\\n\\f AMAZON.COM V. CIR 7\\n\\n I.\\n\\n Before summarizing Amazon’s corporate restructuring\\nand the procedural history of this case, we first provide an\\noverview of the statutory and regulatory framework for\\ntransfer pricing.\\n\\n A.\\n\\n When a taxpayer sells or licenses its property, including\\nintangible assets, to another entity, the purchase price or\\nlicense royalty is taxable income. Often, such transactions\\noccur between entities “owned or controlled directly or\\nindirectly by the same interests.” 26 U.S.C. § 482. 2 The\\nparties in a controlled transaction are in a position to\\npotentially manipulate the terms of the transaction to\\nminimize taxable income artificially. But section 482 gives\\nthe Department of the Treasury the power to reallocate the\\ndollar figures of such controlled transactions if necessary “to\\nprevent evasion of taxes or clearly to reflect the income of”\\na taxpayer. Id.; see also Comm’r v. First Sec. Bank of Utah,\\nN.A., 405 U.S. 394, 400 (1972) (“[Section] 482 is designed\\nto prevent ‘artificial shifting, milking, or distorting of the\\ntrue net incomes of commonly controlled enterprises.’”\\n(quoting B. Bittker & J. Eustice, Federal Income Taxation of\\nCorporations and Shareholders p. 15–21 (3d ed. 1971)).\\n\\n The implementing regulations state that “[t]he purpose\\nof section 482 is to ensure that taxpayers clearly reflect\\nincome attributable to controlled transactions, and to prevent\\nthe avoidance of taxes with respect to such transactions.”\\n\\n 2\\n Unless otherwise indicated, references to section 482 of the\\nInternal Revenue Code and the implementing regulations are to the\\nversions in effect during the tax years at issue here (2005 and 2006).\\n\\f8 AMAZON.COM V. CIR\\n\\nTreas. Reg. § 1.482‑1(a)(1). A reallocation under\\nsection 482 and the implementing regulations is intended to\\n“place[] a controlled taxpayer on a tax parity with an\\nuncontrolled taxpayer by determining the true taxable\\nincome of the controlled taxpayer.” Id. The true taxable\\nincome is determined as if the parties to the controlled\\ntransaction had conducted their affairs in the manner of\\nunrelated parties “dealing at arm’s length.” Treas. Reg.\\n§ 1.482‑1(b)(1). The “arm’s length” standard is met “if the\\nresults of the transaction are consistent with the results that\\nwould have been realized if uncontrolled taxpayers had\\nengaged in the same transaction under the same\\ncircumstances.” Id.\\n\\n The key regulations were promulgated in 1994 and 1995.\\nThe regulations divide property into two categories: tangible\\nproperty (Treas. Reg. § 1.482‑3) and intangible property\\n(Treas. Reg. § 1.482‑4). This case concerns intangible\\nproperty. The regulations provide:\\n\\n (b) Definition of intangible. For purposes of\\n section 482, an intangible is an asset that\\n comprises any of the following items and has\\n substantial value independent of the services\\n of any individual—\\n\\n (1) Patents, inventions, formulae,\\n processes, designs, patterns, or know-\\n how;\\n\\n (2) Copyrights and literary, musical, or\\n artistic compositions;\\n\\n (3) Trademarks, trade names, or brand\\n names;\\n\\f AMAZON.COM V. CIR 9\\n\\n (4) Franchises, licenses, or contracts;\\n\\n (5) Methods, programs, systems,\\n procedures, campaigns, surveys, studies,\\n forecasts, estimates, customer lists, or\\n technical data; and\\n\\n (6) Other similar items. For purposes of\\n section 482, an item is considered similar\\n to those listed in paragraph (b)(1) through\\n (5) of this section if it derives its value not\\n from its physical attributes but from its\\n intellectual content or other intangible\\n properties.\\n\\nTreas. Reg. § 1.482‑4(b).\\n\\n A controlled taxpayer may make its intangibles available\\nto a foreign affiliate by entering a licensing agreement, under\\nwhich the foreign affiliate’s royalty fee is taxable income to\\nthe controlled taxpayer. See generally Treas. Reg. §§ 1.482-\\n1, 1.482-4. The royalty fee must, of course, reflect the\\n“arm’s length” value of the license. As new intangibles\\ncontinue to be developed, the value of the license changes\\nand becomes open to dispute for tax purposes on an ongoing\\nbasis.\\n\\n As an alternative to licensing, the regulations authorize\\njointly controlled entities to enter a cost sharing\\narrangement, under which they become co-owners of\\nintangibles developed as a result of the entities’ joint R&D\\n\\f10 AMAZON.COM V. CIR\\n\\nefforts. See generally Treas. Reg. § 1.482‑7A. 3 This\\narrangement provides the taxpayer the benefit of certainty\\nbecause, assuming the arrangement satisfies the\\nrequirements of the regulations, new intangibles need not be\\nvalued as they are developed.\\n\\n Under the regulations, a subsidiary that enters a cost\\nsharing arrangement with its parent must make two distinct\\npayments. First, the subsidiary must make an arm’s length\\n“buy-in” payment reflecting the value of the pre-existing\\nintangibles the parent contributes to the arrangement:\\n\\n If a controlled participant makes pre-existing\\n intangible property in which it owns an\\n interest available to other controlled\\n participants for purposes of research in the\\n intangible development area under a\\n qualified cost sharing arrangement, then each\\n such other controlled participant must make\\n a buy-in payment to the owner.\\n\\nTreas. Reg. § 1.482‑7A(g)(2). The buy-in payment is\\ntaxable income for the entity contributing the intangibles—\\nhere, Amazon.\\n\\n Second, the subsidiary must pay a share of the intangible\\ndevelopment costs (or R&D) “equal to its share of\\nreasonably anticipated benefits attributable to such\\ndevelopment.” Treas. Reg. § 1.482‑7A(a)(2). If the\\ncontrolled taxpayer incurs the lion’s share of the intangible\\ndevelopment costs, the subsidiary is required to make cost\\n\\n 3\\n We follow the parties’ convention of referring to the regulations\\non cost sharing arrangements by reference to the numbering as\\nredesignated in 2009 (i.e., Treas. Reg. § 1.482‑7A).\\n\\f AMAZON.COM V. CIR 11\\n\\nsharing payments to reflect its share of the anticipated\\nbenefits. The subsidiary’s cost sharing payments serve to\\nreduce the deductions the controlled taxpayer can take for\\nR&D costs (thereby increasing tax liability).\\n\\n B.\\n\\n Amazon is an online retailer that began operating in the\\nUnited States in 1995. In the late 1990s and early 2000s,\\nAmazon expanded operations into France, Germany, and the\\nUnited Kingdom. Amazon’s business in Europe had a siloed\\nstructure, with separate European subsidiaries independently\\noperating and managing the business, each subsidiary\\nhaving its own website, fulfillment centers, and customer\\nbase. The European subsidiaries licensed the right to use\\nAmazon’s website technology, customer information, and\\nmarketing intangibles (including trademarks and domain\\nnames).\\n\\n To address operational inefficiencies in Europe, in the\\nearly 2000s Amazon investigated options for creating a\\ncentralized European headquarters. Amazon ultimately\\nlocated its new headquarters in Luxembourg, which offered\\na central location, the lowest value-added tax rate in Europe,\\nand a relatively low corporate tax rate. Beginning in 2004,\\nAmazon undertook a series of transactions to implement its\\nplan for centralizing business operations in Europe.\\n\\n Amazon formed Amazon Europe Holding Technologies\\nSCS (“AEHT”) and transferred to AEHT the pre-existing\\nEuropean subsidiaries, their operating assets, and their pre-\\nexisting intangible rights (i.e., those developed in Europe).\\nAmazon and AEHT also entered a cost sharing\\n\\f12 AMAZON.COM V. CIR\\n\\narrangement—the transaction most pertinent to this appeal. 4\\nFor the arrangement to be a “qualified cost sharing\\narrangement” under the transfer pricing regulations outlined\\nabove, AEHT needed to pay Amazon for the pre-existing\\nintangibles Amazon contributed to the arrangement. To\\ndetermine the amount of this “buy-in,” Amazon hired a tax\\nfirm, which concluded the pre-existing intangibles were\\nworth $217 million. 5\\n\\n Under the cost sharing arrangement, AEHT also makes\\ncost sharing payments to Amazon for its share of ongoing\\nintangible development costs. Amazon reported cost\\nsharing payments from AEHT of about $116 million for\\n2005 and about $77 million for 2006.\\n\\n The IRS rejected Amazon’s calculation of AEHT’s buy-\\nin, concluding that Amazon grossly undervalued the\\nintangibles Amazon made available to AEHT. Applying a\\ndiscounted-cash-flow valuation methodology, the IRS\\ndetermined a buy-in payment of $3.6 billion.6 Amazon filed\\na petition in the tax court challenging the IRS’s valuation.\\nAmazon argued that the cost sharing arrangement covered\\n\\n 4\\n The following transactions were also part of the restructuring:\\n(1) Amazon granted AEHT a license to use Amazon’s existing\\ntechnology-related intellectual property; (2) Amazon assigned to AEHT\\ncustomer data and certain marketing intangibles (including trademarks,\\nwebsite content, and domain names) relating to the European business;\\n(3) Amazon transferred the stock of the European subsidiaries to AEHT;\\n(4) Amazon transferred other business assets (other than intellectual\\nproperty) to AEHT; and (5) the pre-existing European subsidiaries\\nlicensed to AEHT intellectual property titled in their names.\\n 5\\n The tax firm set the buy-in price at $254.5 million, to be paid over\\na seven-year period (resulting in a present value of $217 million).\\n\\n 6\\n The IRS’s calculation was later reduced to $3.468 billion.\\n\\f AMAZON.COM V. CIR 13\\n\\nthree distinct groups of transferred assets—website\\ntechnology, marketing intangibles, and European-customer\\ninformation—that must be valued separately using a\\nmethodology referred to in the transfer pricing regulations as\\nthe comparable uncontrolled transaction method.\\n\\n After a six-week trial that included testimony and written\\nreports of thirty expert witnesses, the tax court concluded\\nthat the Commissioner abused his discretion in determining\\nthat the discounted cash flow methodology supplied the best\\nmethod for determining an arm’s length buy-in payment and\\nin determining that the required payment is $3.468 billion.\\nAmazon.Com, Inc. v. Comm’r, 148 T.C. 108, 150 (2017).\\nRelying on the rationale of its prior decision in Veritas\\nSoftware Corp. v. Commissioner, 133 T.C. 297 (2009),\\nnonacq., 2010-49 I.R.B (2010), action on dec., 2010-05\\n(Nov. 12, 2010), the tax court reasoned that the\\nCommissioner’s valuation methodology “is based in essence\\non an ‘akin to a sale’ theory” and “necessarily sweeps into\\n[the] calculation assets that were not transferred under the\\n[cost sharing arrangement] and assets that were not\\ncompensable ‘intangibles’ to begin with.” Amazon.Com,\\n148 T.C. at 156–57. The tax court concluded that “[a]n\\nenterprise valuation of a business,” like the one conducted\\nby the Commissioner’s primary expert, “includes many\\nitems of value that are not ‘intangibles’” under the cost\\nsharing regulations. Id. at 157. Such items include\\n“workforce in place, going concern value, goodwill, and\\nwhat trial witnesses described as ‘growth options’ and\\ncorporate ‘resources’ or ‘opportunities.’” Id. The tax court\\nreasoned that such items are “[u]nlike the ‘intangibles’ listed\\n\\f14 AMAZON.COM V. CIR\\n\\nin the statutory and regulatory definitions” in that they\\n“cannot be bought and sold independently.” Id. 7\\n\\n The tax court adopted the comparable uncontrolled\\ntransaction method as the best way to value the buy-in\\npayment because that method allows for the intangibles at\\nissue to be isolated and separately valued. Id. at 164.\\nAlthough the tax court agreed with Amazon’s general\\nvaluation approach, it disagreed with certain aspects of\\nAmazon’s implementation of that method. Id. After making\\nadjustments, the tax court calculated the value of the buy-in\\npayment to be about $779 million. The Commissioner\\ntimely appealed, challenging the tax court’s rejection of his\\nexpert’s discounted cash flow methodology. 8\\n\\n II.\\n\\n The tax court had jurisdiction over this action under\\n26 U.S.C. § 6213(a). We have jurisdiction under 26 U.S.C.\\n§ 7482(a)(1), (b)(1).\\n\\n “[W]e review the tax court’s conclusions of law de novo\\nand its factual findings for clear error.” MK Hillside\\n\\n 7\\n The tax court made several other legal conclusions and factual\\nfindings not relevant to the Commissioner’s appeal.\\n 8\\n Although we affirm the tax court, nothing in our decision should\\nbe construed as an outright rejection of any particular valuation\\nmethodology. Here, the parties’ respective valuation methods did not\\ndiffer only in how they valued Amazon’s assets; they differed in what\\nassets they valued. The selection of the “best method” here, see Treas.\\nReg. § 1.482-1(c), thus turns on an interpretation of the regulatory\\ndefinition of an “intangible.” Nothing in our opinion should be\\nconstrued as favoring one valuation method over another in\\ncircumstances not present here.\\n\\f AMAZON.COM V. CIR 15\\n\\nPartners v. Comm’r, 826 F.3d 1200, 1203 (9th Cir. 2016)\\n(quoting DHL Corp. & Subsidiaries v. Comm’r, 285 F.3d\\n1210, 1216 (9th Cir. 2002)).\\n\\n III.\\n\\n The dispositive issue in this case is whether, under the\\n1994/1995 regulations, the “buy-in” required for “pre-\\nexisting intangible property” must include compensation for\\nresidual-business assets. 9 To answer this legal question, we\\nconsider the regulatory definition of an “intangible,” the\\noverall transfer pricing regulatory framework, the\\nrulemaking history of the regulations, and whether the\\nCommissioner’s position is entitled to deference under Auer\\nv. Robbins, 519 U.S. 452 (1997). We agree with the tax court\\nthat the definition of an “intangible” in § 1.482-4(b) was not\\nintended to embrace residual-business assets.\\n\\n A.\\n\\n The Commissioner argues that Amazon’s valuable\\nresidual-business assets meet “§ 1.482‑4(b)’s definition of\\nintangibles.” “Regulations are interpreted according to the\\nsame rules as statutes, applying traditional rules of\\nconstruction.” Minnick v. Comm’r, 796 F.3d 1156, 1159\\n(9th Cir. 2015); see Kisor v. Wilkie, 139 S. Ct. 2400, 2415\\n(2019) (recognizing that “all the ‘traditional tools’ of\\nconstruction” are the same for both statutes and regulations\\n\\n 9\\n The Commissioner also challenges some of the tax court’s other\\nreasons for rejecting the valuation conducted by the Commissioner’s\\nexpert. But in each of his other challenges, the Commissioner’s\\nargument assumes he is correct on his primary contention that the\\ndefinition of “intangible” in the 1994/1995 regulations embraces\\nresidual-business assets. Because we reject the Commissioner’s view on\\nthe primary issue, we need not address his other contentions.\\n\\f16 AMAZON.COM V. CIR\\n\\n(quoting Chevron, U.S.A., Inc. v. Nat. Res. Def. Council,\\nInc., 467 U.S. 837, 843 n.9 (1984))). Our “legal toolkit”\\nincludes careful examination of “the text, structure, history,\\nand purpose of a regulation.” Kisor, 139 S. Ct. at 2415.\\n\\n We begin with the language of § 1.482-4(b). If the\\nregulation is unambiguous, its plain meaning governs. Safe\\nAir For Everyone v. EPA, 488 F.3d 1088, 1097 (9th Cir.\\n2007).\\n\\n The regulation defines an “intangible” as an asset that\\nboth “has substantial value independent of the services of\\nany individual” and is one of the items listed in subsection\\n(b)(1)–(6). Treas. Reg. § 1.482‑4(b). Each of the 28 specific\\nitems in subsection (b) is independently transferrable—none\\nis a residual-business asset. The Commissioner thus relies\\non the catchall provision for “[o]ther similar items.” Treas.\\nReg. § 1.482‑4(b)(6). “[A]n item is considered similar” to\\nthe other items in the subsection “if it derives its value not\\nfrom its physical attributes but from its intellectual content\\nor other intangible properties.” Id.\\n\\n Reading the catchall provision together with the\\nintroductory language of the definition, residual-business\\nassets are intangibles if they (1) have substantial value\\nindependent of the services of any individual and (2) derive\\ntheir value from intellectual content or other intangible\\nproperties. The Commissioner argues both elements are\\nsatisfied. First, he argues that Amazon’s growth options\\n“derive their value from intangible, rather than physical,\\nattributes.” He then cites testimony from Amazon’s expert\\n(Dr. Bradford Cornell) that Amazon’s growth options are\\nprimarily attributable to its culture of innovation. Second,\\nhe argues that the value of Amazon’s growth options is\\nindependent of the services of any individual because they\\n\\f AMAZON.COM V. CIR 17\\n\\nare “part of the culture of Amazon to be able to have creative\\nideas bubble up in their organization and actually use them.”\\n\\n Amazon argues that the Commissioner’s interpretation\\nof the catchall provision is too sweeping for several reasons.\\nAmazon’s central argument is that to qualify as an\\n“intangible” under the regulation, an item must be capable\\nof being bought and sold independently of the business—\\nand residual-business assets are inseparable from the\\nbusiness. The tax court agreed. See Amazon.Com, 148 T.C.\\nat 157 (concluding that, unlike the specific intangibles listed\\nin the regulation, “workforce in place, going concern value,\\ngoodwill, and what trial witnesses described as ‘growth\\noptions’ and corporate ‘resources’ or ‘opportunities’ . . .\\ncannot be bought and sold independently”).\\n\\n Amazon offers several arguments in support of its\\nposition that assets that are inseparable from the business do\\nnot meet the regulatory definition of an “intangible.”\\nQuoting Arcadia v. Ohio Power Co., 498 U.S. 73, 78 (1990),\\nAmazon argues that the Commissioner’s interpretation\\nrenders the “enumeration of specific subjects entirely\\nsuperfluous—in effect adding to that detailed list ‘or\\nanything else.’” Amazon invokes the canon ejusdem\\ngeneris—of the same kind or class—and asserts that “[t]he\\ncommon attribute of all 28 specified items is that they can be\\nsold independently.” Amazon argues this canon should be\\napplied to prevent the catchall clause from swallowing the\\npreceding language of the regulation.\\n\\n Amazon’s focus on the commonality of the 28 specified\\nitems has some force. After all, if all 28 listed items share a\\ncommon attribute, why would anyone understand a catchall\\nfor “similar items” to include a non-listed item that doesn’t\\nshare that attribute? Amazon’s commonality argument\\nfalters, however, because the catchall provision did not\\n\\f18 AMAZON.COM V. CIR\\n\\nsimply say “[o]ther similar items.” Instead, the catchall\\nelaborated by explaining how an item is determined to be\\n“similar” to the other items: “if it derives its value not from\\nits physical attributes but from its intellectual content or\\nother intangible properties.” Treas. Reg. § 1.482-4(b)(6).\\nThis explanation of what is “similar” leaves open the\\npossibility of a non-listed item being included in the\\ndefinition even if it doesn’t share the attribute of being\\nseparately transferrable.\\n\\n Amazon also argues that the regulation’s requirement\\nthat an intangible have “substantial value independent of the\\nservices of any individual,” Treas. Reg. § 1.482‑4(b),\\nsupports its position. The tax court in Veritas agreed. In that\\ncase, the tax court rejected the Commissioner’s argument\\nthat a foreign subsidiary’s buy-in under a cost sharing\\narrangement must include the value of the subsidiary’s\\naccess to the U.S. corporate parent’s R&D and marketing\\nteams—i.e., workforce in place. Veritas, 133 T.C. at 323.\\nAlthough the court found insufficient evidence that the cost\\nsharing arrangement made such a transfer, it also stated that\\n“[e]ven if such evidence existed, these items would not be\\ntaken into account in calculating the requisite buy-in\\npayment because they do not have ‘substantial value\\nindependent of the services of any individual’ and thus do\\nnot meet the” definition of “intangible.” Id. at 323 n.31\\n(quoting Treas. Reg. § 1.482‑4(b)). The court reasoned that\\nthe value of access to the parent’s “R&D and marketing\\nteams is based primarily on the services of individuals (i.e.,\\nthe work, knowledge, and skills of team members).” Id. The\\nCommissioner clearly disagrees with Veritas on this point,\\nbut he doesn’t explain why the tax court’s analysis is wrong.\\n\\n Analysis of the regulatory text alone does not\\ndefinitively resolve the question here. The definition of an\\n\\f AMAZON.COM V. CIR 19\\n\\n“intangible” is susceptible to, but does not compel, an\\ninterpretation that embraces residual-business assets. The\\nproblem is that residual-business assets, such as “growth\\noptions” and a “culture of innovation,” are amorphous, and\\nit’s not self-evident whether such assets have “substantial\\nvalue independent of the services of any individual.” See\\nTreas. Reg. § 1.482‑4(b). Amazon raises legitimate\\nconcerns about the regulation’s catchall being stretched too\\nfar, and those concerns likely bear on which party has the\\nmore reasonable view of the regulatory definition. It\\nnonetheless remains that the definition of “intangible” could\\nbe construed as covering residual-business assets if the\\nlanguage of § 1.482‑4(b) is viewed in isolation. 10\\n\\n B.\\n\\n But we are required to look at the regulatory scheme “as\\na whole,” viewing the regulatory “definition in the context\\nof the entire [transfer pricing] regulations.” Alaska Trojan\\nP’ship v. Gutierrez, 425 F.3d 620, 628 (9th Cir. 2005). The\\nCommissioner argues that other regulations governing\\ntransfer pricing support his view that the regulatory\\ndefinition of an “intangible” embraces residual-business\\n\\n 10\\n The Commissioner also argues that the definition of an\\n“intangible” includes residual-business assets because an uncontrolled\\nparty would pay for access to those assets in an arm’s length transaction.\\nHe cites the testimony of Amazon’s expert that parties dealing at arm’s\\nlength “[d]efinitely” pay for “growth options” because “[n]o company is\\ngoing to give away something of value without compensation.” The\\nCommissioner’s argument misses the mark. Under the regulations, the\\narm’s length standard governs the valuation of intangibles; it doesn’t\\nanswer whether an item is an intangible. The definition of an\\n“intangible” is provided in § 1.482‑4(b). The Commissioner points to no\\nlanguage in the statute or regulations suggesting that the definition of\\nwhat constitutes an intangible is determined by asking whether an\\nuncontrolled party would pay for it.\\n\\f20 AMAZON.COM V. CIR\\n\\nassets. He cites two sections in the regulations, §§ 1.482‑7A\\nand 1.482‑1.\\n\\n Section 1.482‑7A specifies the requirements of a\\nqualified cost sharing arrangement and the methods for\\ndetermining the taxable income resulting from such an\\narrangement. The Commissioner cites the subsection that\\nimposes the “buy-in” requirement where “[a] controlled\\nparticipant . . . makes intangible property available to a\\nqualified cost sharing arrangement.” Treas. Reg.\\n§ 1.482‑7A(g)(1). The Commissioner argues that this\\nprovision mandates that residual-business assets be paid for\\n“if they are made ‘available’ to the cost-sharing participants”\\neven though such assets generally cannot be transferred\\nindependently from the business.\\n\\n The Commissioner’s argument based on\\n§ 1.482‑7A(g)(1) presupposes the very point he attempts to\\nprove—that residual-business assets are “intangible\\nproperty” within the meaning of the regulations. The\\nprovision requiring a “buy-in” does not expand but instead\\nincorporates the meaning of an “intangible” given in\\n§ 1.482-4(b). See Treas. Reg. § 1.482‑7A(g)(2) (“The buy-\\nin payment by each such other controlled participant is the\\narm’s length charge for the use of the intangible . . . .”\\n(emphasis added)). The “makes . . . available” language thus\\nprovides no meaningful insight into the regulatory definition\\nof an “intangible.” 11\\n\\n\\n 11\\n Considered in context of the regulatory framework, it appears that\\nthe purpose of the “makes . . . available” phrasing in § 1.482‑7A(g) is to\\naccount for taxpayers who might seek to avoid the requirements of the\\ntransfer pricing regulations by not formally transferring their intangible\\nassets to a related party and instead informally making available the\\n\\f AMAZON.COM V. CIR 21\\n\\n The Commissioner also finds support for his position in\\nthe regulations’ preamble stating that “[t]he purpose of\\nsection 482 is to ensure that taxpayers clearly reflect income\\nattributable to controlled transactions and to prevent the\\navoidance of taxes with respect to such transactions.” Treas.\\nReg. § 1.482‑1(a)(1). Relying on Xilinx, Inc. v.\\nCommissioner, 598 F.3d 1191 (9th Cir. 2010), the\\nCommissioner argues that “anything of value that is made\\navailable between related parties must be paid for” in the\\nbuy-in, regardless of whether it is defined as an intangible.\\n\\n In Xilinx, we considered whether the parties to a cost\\nsharing arrangement “must include the value of certain stock\\noption compensation one participant gives to its employees\\nin the pool of costs to be shared.” Id. at 1192. We found\\ntwo provisions of the regulations in conflict. We first cited\\n§ 1.482‑1(b)(1)’s statement that “the standard to be applied\\nin every case is that of a taxpayer dealing at arm’s length\\nwith an uncontrolled taxpayer.” Id. at 1195 (quoting Treas.\\nReg. § 1.482‑1(b)(1)). It was undisputed on appeal that\\nunrelated parties would not have shared the employee stock\\noption costs. Id. at 1194. Another provision, however,\\nrequired that “controlled parties in a cost sharing agreement\\n. . . share all ‘costs . . . related to the intangible development\\narea,’” which by definition would include the employee\\nstock options. Id. at 1196. In resolving this conflict, the\\nXilinx majority affirmed the tax court’s conclusion that the\\nstock options need not be paid for as part of the buy-in. Id.\\n\\n\\n\\n\\nassets. The regulations thus treat such taxpayers “as having transferred\\ninterests in such [intangible] property.” Treas. Reg. § 1.482‑7A(g)(1).\\n\\f22 AMAZON.COM V. CIR\\n\\nat 1196–97. The primary opinion reasoned that “[p]urpose\\nis paramount.” Id. at 1196. 12\\n\\n Relying on the logic of the primary opinion in Xilinx, the\\nCommissioner asserts that “it is undisputed that a company\\nentering into the same transaction under the same\\ncircumstances with an unrelated party would have required\\ncompensation.” To evaluate this claim, it’s important to be\\nclear about what exactly is meant by the phrase “the same\\ntransaction under the same circumstances.” The\\nCommissioner relies on deposition testimony from one of\\nAmazon’s experts that parties dealing at arm’s length would\\npay for growth options. But the expert explained at trial the\\ndifference between an investor or purchaser of the entire\\nbusiness (who would pay for the full value of the business)\\nand a partner (who would not). The question becomes\\nwhether a cost sharing arrangement is akin to the sale of a\\nbusiness or like a partnership in certain assets or aspects of\\nthe business. The Commissioner assumes, but does not\\nexplain why, the transfer of intangible assets under\\n\\n\\n\\n 12\\n Caution should be taken before relying on the rationale of the\\nprimary opinion in Xilinx. One panel member concurred—thus\\nproviding a majority vote for the disposition—but wrote separately “to\\nexplain [his] particular reasons for rejecting the Commissioner’s\\nposition.” Id. at 1197 (Fisher, J., concurring). Although the reasoning\\nof the concurring opinion overlaps somewhat with the reasoning of the\\nprimary opinion, the concurrence does not seem to adopt the “[p]urpose\\nis paramount” logic. Instead, based on the language of the regulations,\\nthe legislative and regulatory history, international tax treaties, and the\\nunderstanding of the business community and tax professionals (i.e.,\\namici curiae), the concurrence concluded that the taxpayer’s\\n“understanding of the regulations is the more reasonable even if the\\nCommissioner’s current interpretation may be theoretically plausible.”\\nId. at 1198.\\n\\f AMAZON.COM V. CIR 23\\n\\nAmazon’s cost sharing arrangement with AEHT should be\\ntreated the same as the sale of the business.\\n\\n The Commissioner’s reliance on Xilinx thus suffers the\\nsame defect as his “made available” argument based on\\n§ 1.482-7A(g)—he assumes the very conclusion he’s aiming\\nto prove. Although the regulatory provisions the\\nCommissioner cites are consistent with his position, they do\\nnot provide independent support and they are likewise\\nconsistent with Amazon’s view.\\n\\n If the cost sharing regulations as a whole tip the scale\\neither direction, they tend to favor Amazon on the issue\\npresented here. The regulations describe a cost sharing\\narrangement as an agreement “to share the costs of\\ndevelopment of one or more intangibles.” Treas. Reg.\\n§ 1.482‑7A(a)(1). The written document memorializing the\\narrangement must, among other things, describe both “the\\nscope of the research and development to be undertaken,\\nincluding the intangible or class of intangibles intended to be\\ndeveloped.” Treas. Reg. § 1.482‑7A(b)(4)(iii). The writing\\nmust also describe “each participant’s interest in any covered\\nintangibles,” which is “any intangible property that is\\ndeveloped as a result of the research and development\\nundertaken under the cost sharing arrangement.” Treas.\\nReg. § 1.482‑7A(b)(4)(iv). By identifying intangibles as\\nbeing the product of R&D efforts, the regulations seem to\\ncontemplate a meaning of “intangible” that excludes items\\nlike goodwill and going concern value, which “are generated\\nby earning income, not by incurring deductions.” Staff of J.\\nComm. on Taxation, 98th Cong., General Explanation of the\\nRevenue Provisions of the Deficit Reduction Act of 1984,\\n428 (Comm. Print 1984).\\n\\f24 AMAZON.COM V. CIR\\n\\n The overall regulatory scheme doesn’t definitively\\nresolve the issue, but it favors Amazon more than the\\nCommissioner.\\n\\n C.\\n\\n We next turn to the drafting history of the regulatory\\ndefinition. Both parties claim support for their respective\\npositions in the historical development of the regulations.\\n\\n Treasury first defined intangible property for purposes of\\nsection 482 in regulations adopted in 1968:\\n\\n [I]ntangible property shall consist of the\\n items described in subdivision (ii) of this\\n subparagraph, provided that such items have\\n substantial value independent of the services\\n of individual persons.\\n\\n (ii) The items referred to in subdivision (i) of\\n this subparagraph are as follows:\\n\\n (a) Patents, inventions, formulas,\\n processes, designs, patterns, and other\\n similar items;\\n\\n (b) Copyrights, literary, musical, or\\n artistic compositions, and other similar\\n items;\\n\\n (c) Trademarks, trade names, brand\\n names, and other similar items;\\n\\n (d) Franchises, licenses, contracts, and\\n other similar items;\\n\\f AMAZON.COM V. CIR 25\\n\\n (e) Methods, programs, systems,\\n procedures, campaigns, surveys, studies,\\n forecasts, estimates, customer lists,\\n technical data, and other similar items.\\n\\n33 Fed. Reg. 5848, 5854 (Apr. 16, 1968).\\n\\n Fourteen years later, Congress enacted the Tax Equity\\nand Fiscal Responsibility Act of 1982, adopting a definition\\nof intangible property in section 936 of the Internal Revenue\\nCode that was nearly identical to the one in the regulations\\nimplementing section 482. 13 The Commissioner plucks a\\nphrase from a Senate Report for the 1982 Act suggesting that\\nthe statute “defines intangible assets broadly.” With more\\ncontext, the quoted sentence from the Senate Report states\\nthat “[t]he bill defines intangible assets broadly to include”\\nthe 28 specifically-listed items “and other items similar to\\nany of those listed, so long as the item has substantial value\\nindependent of the services of individual persons.” S. Rep.\\n97-494, at 161 (1982), as reprinted in 1982 U.S.C.C.A.N.\\n781, 924 (emphasis added). 14\\n\\n\\n 13\\n The new statutory definition differed from the regulatory one in\\nfour, relatively minor respects: (1) making all enumerated items singular;\\n(2) adding “know-how” to the first category; (3) inserting “any similar\\nitem” as a stand-alone category in lieu of “other similar items” in each\\ncategory; and (4) requiring that an item have “substantial value\\nindependent of the services of any individual” rather than “of individual\\npersons.” Tax Equity & Fiscal Responsibility Act of 1982, Pub. L. 97-\\n248, § 213, 96 Stat. 324 (Sept. 3, 1982).\\n\\n 14\\n Another part of the quoted Senate Report states that the\\nCommittee viewed the bill as combatting the practice of transferring\\nintangibles “created, developed or acquired in the United States” to\\nforeign entities to generate income tax-free. S. Rep. 97-494, at 158–59\\n(1982), as reprinted in 1982 U.S.C.C.A.N. 781, 921–22. It seems from\\n\\f26 AMAZON.COM V. CIR\\n\\n The Tax Reform Act of 1986 amended section 482 to\\nincorporate the “intangible property” definition from section\\n936: “In the case of any transfer (or license) of intangible\\nproperty (within the meaning of section 936(h)(3)(B)), the\\nincome with respect to such transfer or license shall be\\ncommensurate with the income attributable to the\\nintangible.” Pub. L. 99–514, § 1231, 100 Stat. 2085 (Oct.\\n22, 1986).\\n\\n When it amended section 482, Congress requested that\\nthe IRS conduct “a comprehensive study of intercompany\\npricing rules” and report on whether the existing regulations\\n“could be modified in any respect.” H.R. Rep. No. 99-841,\\nat II-637 (1986). The resulting IRS report, known as the\\n“White Paper,” laid the groundwork for what would\\nultimately become the 1994/1995 regulations. A Study of\\nIntercompany Pricing Under Section 482 of the Code, I.R.S.\\nNotice 88-123, 1988-2 C.B. 458.\\n\\n The White Paper discussed, among other things, cost\\nsharing arrangements. According to the White Paper, the\\npurpose of the buy-in requirement is for “a party to a cost\\nsharing arrangement that has contributed funds or incurred\\nrisks for development of intangibles at an earlier stage” to be\\n“compensated by the other participants.” Id. at 497. “[I]f\\nthere are intangibles that are not fully developed that relate\\nto the research to be conducted under the cost sharing\\narrangement, it is necessary to value them in order to\\ndetermine an appropriate buy-in payment.” Id.\\n\\n\\n\\n\\nthis language, the Committee didn’t contemplate intangibles that are not\\nindependently transferrable.\\n\\f AMAZON.COM V. CIR 27\\n\\n The White Paper proposed “three basic types of\\nintangibles” that would be subject to the buy-in requirement:\\n\\n • “preexisting intangibles at various stages of\\n development that will become subject to the\\n arrangement”;\\n\\n • “basic research not associated with any product”; and\\n\\n • “a going concern value associated with a\\n participant’s research facilities and capabilities that\\n will be utilized.”\\n\\nId. Although the White Paper proposed including going\\nconcern value of a research facility in the buy-in, after\\nreceiving opposition in public comments, Treasury proposed\\nnew regulations that essentially retained the definition of\\n“intangible” from before without referencing going concern\\nvalue or any other residual-business asset. See 57 Fed. Reg.\\n3571, 3579 (Jan. 30, 1992).\\n\\n In 1993, Treasury issued revised temporary and\\nproposed regulations that defined an “intangible” as “any\\ncommercially transferable interest” in the intangibles listed\\nin § 936(h)(3)(B) that had “substantial value independent of\\nthe services of any individual.” 58 Fed. Reg. 5263, 5287\\n(Jan. 21, 1993). Treasury also requested comment on\\n“whether the definition of intangible property . . . should be\\nexpanded to include items not normally considered to be\\nitems of intellectual property, such as work force in place,\\ngoodwill or going concern value.” 58 Fed. Reg. 5310, 5312\\n(Jan. 21, 1993). Amazon cites opposition comments\\nsubmitted in response to Treasury’s request for comment.\\n\\n In 1994, Treasury issued final regulations (the ones\\napplicable here), which reflected a minor reworking of the\\n\\f28 AMAZON.COM V. CIR\\n\\ndefinition. 59 Fed. Reg. 34971, 35016 (Jul. 8, 1994).\\nTreasury explained that the revised definition omitted the\\n“commercially transferrable” language that appeared in the\\ntemporary regulations “because it was superfluous: if the\\nproperty was not commercially transferrable, then it could\\nnot have been transferred in a controlled transaction.” Id.\\nat 34983. Treasury also explained that the revision\\n“clarified” that the phrase “other similar items” in the\\ndefinition “refer[s] to items that derive their value from\\nintellectual content or other intangible properties rather than\\nphysical attributes.” Id. In 1995, Treasury issued final\\nregulations governing cost sharing arrangements, including\\n§ 1.482‑7A(g)’s “buy-in” requirement.\\n\\n The drafting history of the transfer pricing regulations\\ndoes not support the Commissioner’s argument that the\\ndefinition of an “intangible” covered residual-business\\nassets. The only references in the drafting history to any\\nresidual-business assets suggest that such items were\\nexcluded from the definition of intangible assets. The IRS’s\\n1988 White Paper proposed including “going concern value”\\nof a research facility in the buy-in, but Treasury’s 1994/1995\\nregulations kept essentially the same definition as before\\nwithout referring to “going concern value” or any other\\nresidual-business asset. See 57 Fed. Reg. at 3579.\\n\\n Two key statements by Treasury in the drafting history\\nrender the Commissioner’s current position untenable. First,\\nin 1993, Treasury confirmed that the then-existing definition\\nof “intangible” did not include residual-business assets when\\nit asked for comments on whether the definition of\\nintangibles “should be expanded to include items not\\nnormally considered to be items of intellectual property,\\nsuch as work force in place, goodwill or going concern\\nvalue.” 58 Fed. Reg. at 5312 (emphasis added). Second, a\\n\\f AMAZON.COM V. CIR 29\\n\\nyear later after opting against such an expansion, and instead\\nretaining the same essential definition from before\\n(including the same list of 28 items), Treasury explained that\\nthe final (1994) rule merely “clarified” when an item would\\nbe deemed similar to the 28 items listed in the definition.\\n59 Fed. Reg. at 34983.\\n\\n The Commissioner is thus forced to argue that what\\nTreasury explicitly confirmed would not be considered an\\n“intangible” without a substantive “expan[sion]” of the\\ndefinition was implicitly added to the definition through a\\nnon-specific “clarifi[cation].” The Commissioner’s\\nargument stretches “clarification” beyond its commonly\\nunderstood meaning of merely clearing up what was\\npreviously ambiguous or otherwise restating a standard\\nconsistent with what was previously intended. Cf. Motorola,\\nInc. v. Fed. Exp. Corp., 308 F.3d 995, 1007 (9th Cir. 2002)\\n(distinguishing between a “clarifying amendment” and “one\\nthat work[s] a substantive change” (emphasis omitted)). 15\\n\\n Another statement from the drafting history of § 1.482‑4\\nlends further support for Amazon’s position that\\n“intangible” has always been understood to be limited to\\nassets that are independently transferrable. Treasury’s\\ntemporary regulations in 1993 defined an “intangible” as\\n“any commercially transferable interest” in the intangibles\\nlisted in section 936(h)(3)(B) that had “substantial value\\nindependent of the services of any individual.” 58 Fed. Reg.\\nat 5287. When Treasury left out the “commercially\\n 15\\n Amazon and amici curiae also argue that if the Commissioner is\\ncorrect that the non-specific “clarifi[cation]” of § 1.482-4(b)’s catchall\\nsubstantively expanded the definition of an “intangible,” then\\nTreasury/IRS violated the Administrative Procedures Act. We need not\\naddress this argument because we reject the Commissioner’s post hoc\\ninterpretation of the changes to the regulatory definition.\\n\\f30 AMAZON.COM V. CIR\\n\\ntransferrable” language from the final regulations issued a\\nyear later, it explained that the omitted language was\\n“superfluous” because “if the property was not\\ncommercially transferrable, then it could not have been\\ntransferred in a controlled transaction.” 59 Fed. Reg. at\\n34983. This is consistent with the basic premise of the\\ntransfer pricing regulations, which contemplate a situation in\\nwhich particular assets are transferred from one entity to\\nanother.\\n\\n The Commissioner now contends that the 1994\\ndefinition was intended to embrace residual-business assets\\neven though such assets “cannot be transferred\\nindependently.” Yet the Commissioner fails to identify any\\ncontemporaneous statement by the agency that would\\n“display awareness” that it was changing its position on\\nwhether residual-business assets are included within the\\ndefinition of intangibles. See FCC v. Fox Television\\nStations, Inc., 556 U.S. 502, 515 (2009) (“[T]he requirement\\nthat an agency provide reasoned explanation for its action\\nwould ordinarily demand that it display awareness that\\nit is changing position. An agency may not, for example,\\ndepart from a prior policy sub silentio or simply disregard\\nrules that are still on the books.”).\\n\\n The Commissioner argues his position is supported by a\\ndifferent regulatory scheme governing penalties for\\ntaxpayers who substantially misstate the value of property\\non their tax returns. See 26 U.S.C. § 6662(e). In 1993,\\nTreasury proposed regulations “designed to encourage\\ntaxpayers to document their transfer pricing transactions and\\nto provide that documentation to the [IRS] upon request.”\\n58 Fed. Reg. 5304, 5304 (Jan. 21, 1993). To that end, the\\nproposed regulations provided rules for determining whether\\na taxpayer has substantially misstated the value of property.\\n\\f AMAZON.COM V. CIR 31\\n\\nThose rules defined “property” to include “intangible\\nproperty,” which in turn was defined to include “property\\nsuch as goodwill, covenants not to compete, leaseholds,\\npatents, contract rights, debts, and choices in action.” Id.\\nat 5306.\\n\\n The Commissioner’s attempt to bootstrap § 1.6662-5’s\\nreference to “goodwill” ignores that the proposed regulation\\nconcerning substantial valuation misstatements was issued\\non the same day Treasury separately confirmed that the\\ndefinition of “intangible” for purposes of the regulations\\nimplementing section 482 did not include goodwill or other\\nresidual-business assets. 58 Fed. Reg. at 5312. Considering\\nthe context, Treasury’s use of “goodwill” in § 1.6662-5 but\\nnot in § 1.482‑4(b) most likely evinces an intent not to\\ninclude goodwill or other residual-business assets within\\n§ 1.482‑4(b)’s definition of “intangible.”\\n\\n Amazon points to other Treasury regulations that define\\ncertain covered property by incorporating the definition of\\nintangible property under section 936(h)(3)(B) and then\\nadding goodwill and going concern value. Treas. Reg.\\n§§ 1.954-2(e)(3)(iv), 1.861-9T(h)(1)(ii). These provisions\\nadd little to the discussion, but the express inclusion of\\ngoodwill and going concern value in these regulations is\\nconsistent with Amazon’s position that the definition of\\n“intangible” in section 936(h)(3)(B) and § 1.482‑4(b) was\\nunderstood to exclude goodwill and going concern value.\\nThese regulations also show that Treasury “clearly knew\\nhow to write its regulations” to include goodwill and other\\nresidual-business assets. Karczewski v. DCH Mission Valley\\nLLC, 862 F.3d 1006, 1015–16 (9th Cir. 2017).\\n\\f32 AMAZON.COM V. CIR\\n\\n The drafting history of § 1.482‑4(b) strongly supports\\nAmazon’s position that Treasury limited the definition of\\n“intangible” to independently transferrable assets. 16\\n\\n D.\\n\\n The Commissioner next argues that the tax court should\\nhave deferred to the IRS’s interpretation of its own\\nregulations. Under certain circumstances, an agency’s\\ninterpretation of its own regulations “must be given\\n‘controlling weight’” if it is not “‘plainly erroneous or\\ninconsistent with the regulation.’” Stinson v. United States,\\n508 U.S. 36, 45 (1993) (quoting Bowles v. Seminole Rock &\\nSand Co., 325 U.S. 410, 414 (1945)); see also Auer v.\\nRobbins, 519 U.S. 452, 461–62 (1997). This is frequently\\nreferred to as Auer deference.\\n\\n The Supreme Court “has cabined Auer’s scope in varied\\nand critical ways.” Kisor, 139 S. Ct. at 2418. “First and\\nforemost, a court should not afford Auer deference unless the\\n 16\\n In 2017, Congress amended the definition of “intangible property”\\nin section 936(h)(3)(B). That amendment added “goodwill, going\\nconcern value, or workforce in place” to the list of specific items\\nincluded in the definition of “intangible property.” Tax Cuts & Jobs Act\\nof 2017, Pub. L. 115-97, § 14221(a), 131 Stat. 2054, 2218 (2017).\\nCharacterizing the change as merely a “clarification,” the Commissioner\\nargues that the 2017 amendment supports his proffered interpretation of\\nthe 1994/1995 regulations. But the Commissioner’s post-hoc label of\\nCongress’s amendment is not controlling, see Beaver v. Tarsadia Hotels,\\n816 F.3d 1170, 1186 (9th Cir. 2016) (“Post-hoc labeling as a\\n‘clarification’ by bill supporters of what otherwise appears to be a change\\n. . . is not controlling . . . .”), and it’s also not supported by Congress’s\\nown words. Congress stated that the amendment should not be\\n“construed to create any inference” as to the definition of intangibles for\\ntaxable years occurring before the amendment’s effective date. 131 Stat.\\nat 2219. Congress said nothing to indicate that the amendment was\\nmeant only as a clarification.\\n\\f AMAZON.COM V. CIR 33\\n\\nregulation is genuinely ambiguous.” Id. at 2415. Genuine\\nambiguity is not determined by examination of the\\nregulatory text alone. Instead, “before concluding that a rule\\nis genuinely ambiguous, a court must exhaust all the\\n‘traditional tools’ of construction,” “‘carefully\\nconsider[ing]’ the text, structure, history, and purpose of a\\nregulation, in all the ways it would if it had no agency to fall\\nback on.” Id. (first quoting Chevron, 467 U.S. at 843 n.9;\\nthen quoting Pauley v. BethEnergy Mines, Inc., 501 U.S.\\n680, 707 (1991) (Scalia, J., dissenting))).\\n\\n The text of the regulatory definition of “intangible,” the\\ndefinition’s place within the transfer pricing regulations\\ngenerally, and the rulemaking history leave little room for\\nthe Commissioner’s proffered meaning. But even if there\\nwere genuine ambiguity, there is a separate reason Auer\\ndeference is not warranted here.\\n\\n “[N]ot every reasonable agency reading of a genuinely\\nambiguous rule should receive Auer deference.” Kisor,\\n139 S. Ct. at 2416. Instead, “a court must make an\\nindependent inquiry into whether the character and context\\nof the agency interpretation entitles it to controlling weight.”\\nId. (citing Christopher v. SmithKline Beecham Corp.,\\n567 U.S. 142, 155 (2012)). For example, courts will not\\ndefer to an agency’s interpretation when doing so “would\\nseriously undermine the principle that agencies should\\nprovide regulated parties fair warning of the conduct [a\\nregulation] prohibits or requires.” Barboza v. Cal. Ass’n of\\nProf’l Firefighters, 799 F.3d 1257, 1267 (9th Cir. 2015)\\n(alteration in original) (quoting Christopher, 567 U.S.\\nat 156). This exception accounts for the “risk that agencies\\nwill promulgate vague and open-ended regulations that they\\ncan later interpret as they see fit, thereby ‘frustrat[ing] the\\nnotice and predictability purposes of rulemaking.’”\\n\\f34 AMAZON.COM V. CIR\\n\\nChristopher, 567 U.S. at 158 (quoting Talk Am., Inc. v.\\nMichigan Bell Tel. Co., 564 U.S. 50, 68 (2011) (alteration in\\noriginal) (Scalia, J., concurring)).\\n\\n Christopher and Barboza thus teach that the timing of an\\nagency’s first announcement of its interpretation may be\\ndispositive on whether the agency’s view will be given Auer\\ndeference. Here, the Commissioner does not identify a\\nspecific document (e.g., policy manual or court brief)\\ndefinitively expressing the agency’s view of its regulations.\\nIt thus appears that the Commissioner’s court briefs in this\\ncase present Treasury’s “first announce[ment of] its view,”\\nsee Christopher, 567 U.S. at 153, that the definition of\\nintangible in § 1.482-4(b) embraces residual-business assets.\\nThe exception to Auer deference from Christopher and\\nBarboza therefore applies. “Where an agency announces its\\ninterpretation for the first time in an enforcement\\nproceeding, and has not previously taken any action to\\nenforce that interpretation, ‘the potential for unfair surprise\\nis acute.’” Barboza, 799 F.3d at 1267 (quoting Christopher,\\n567 U.S. at 158). Even if first arising before the current\\nlitigation, a new interpretation is owed no deference if it\\nwould “create[] ‘unfair surprise’ to regulated parties.”\\nKisor, 139 S. Ct. at 2417–18 (quoting Long Island Care at\\nHome, Ltd. v. Coke, 551 U.S. 158, 170 (2007)). No\\nstatement from Treasury in the drafting history of the\\n1994/1995 regulations expresses the position the\\nCommissioner advances now. Indeed, as discussed above,\\nTreasury’s contemporaneous explanations of the regulations\\nare to the contrary. Amazon and other taxpayers were thus\\nnot given fair warning of the Commissioner’s current\\ninterpretation of the regulatory definition of an “intangible.”\\nThat interpretation is not entitled to deference.\\n\\f AMAZON.COM V. CIR 35\\n\\n IV.\\n\\n The Commissioner’s calculation of AEHT’s buy-in\\nunder the cost sharing arrangement included residual-\\nbusiness assets as part of Amazon’s pre-existing intangibles.\\nThe language of the (now-superseded) regulatory definition\\nof an “intangible” is ambiguous and could be construed as\\nincluding residual-business assets. But the drafting history\\nof the regulations and other indicators of Treasury’s\\ncontemporaneous intent strongly favor Amazon’s proffered\\nmeaning—that intangibles were limited to independently\\ntransferrable assets. Treasury appears to have changed its\\nposition on the meaning of the regulation after Amazon and\\nAEHT entered into their cost sharing arrangement. We share\\nthe sentiment reflected in the concurring opinion in Xilinx:\\n\\n Indeed, I am troubled by the complex,\\n theoretical nature of many of the\\n Commissioner’s arguments trying to\\n reconcile the two regulations. Not only does\\n this make it difficult for the court to navigate\\n the regulatory framework, it shows that\\n taxpayers have not been given clear, fair\\n notice of how the regulations will affect\\n them.\\n\\nXilinx, 598 F.3d at 1198 (Fisher, J., concurring). We\\ntherefore agree with the tax court that the former regulatory\\ndefinition of an “intangible” does not include residual-\\nbusiness assets.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426343/\", \"author_raw\": \"CALLAHAN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge2":{"kind":"string","value":"CONSUELO M CALLAHAN"},"judge3":{"kind":"string","value":"MORGAN CHRISTEN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649090/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: amazon.com Inc. & Subsidiaries v. Cir\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-16\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nAMAZON.COM, INC. & SUBSIDIARIES, No. 17-72922\n Petitioner-Appellee,\n Tax Ct. No.\n v. 31197-12\n\nCOMMISSIONER OF INTERNAL\nREVENUE, OPINION\n Respondent-Appellant.\n\n Appeal from a Decision of the\n United States Tax Court\n\n Argued and Submitted April 12, 2019\n Seattle, Washington\n\n Filed August 16, 2019\n\n Before: William A. Fletcher, Consuelo M. Callahan,\n and Morgan Christen, Circuit Judges.\n\n Opinion by Judge Callahan\n\f2 AMAZON.COM V. CIR\n\n SUMMARY *\n\n\n Tax\n\n The panel affirmed the Tax Court’s decision on a petition\nfor redetermination of federal income tax deficiencies, in an\nappeal involving the regulatory definition of intangible\nassets and the method of their valuation in a cost-sharing\narrangement.\n\n In the course of restructuring its European businesses in\na way that would shift a substantial amount of income from\nU.S.-based entities to the European subsidiaries, appellee\nAmazon.com, Inc. entered into a cost sharing arrangement\nin which a holding company for the European subsidiaries\nmade a “buy-in” payment for Amazon’s assets that met the\nregulatory definition of an “intangible.” See 26 U.S.C.\n§ 482. Tax regulations required that the buy-in payment\nreflect the fair market value of Amazon’s pre-existing\nintangibles. After the Commissioner of Internal Revenue\nconcluded that the buy-in payment had not been determined\nat arm’s length in accordance with the transfer pricing\nregulations, the Internal Revenue Service performed its own\ncalculation, and Amazon filed a petition in the Tax Court\nchallenging that valuation.\n\n At issue is the correct method for valuing the pre-\nexisting intangibles under the then-applicable transfer\npricing regulations. The Commissioner sought to include all\nintangible assets of value, including “residual-business\nassets” such as Amazon’s culture of innovcation, the value\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f AMAZON.COM V. CIR 3\n\nof workforce in place, going concern value, goodwill, and\ngrowth options. The panel concluded that the definition of\n“intangible” does not include residual-business assets, and\nthat the definition is limited to independently transferrable\nassets.\n\n\n COUNSEL\n\nJudith A. Hagley (argued), Gilbert S. Rothenberg, and\nArthur T. Catterall, Attorneys, Tax Division; Richard E.\nZuckerman, Principal Deputy Assistant Attorney General;\nTravis A. Greaves, Deputy Assistant Attorney General; Tax\nDivision, United States Department of Justice, Washington,\nD.C.; for Respondent-Appellant.\n\nCarter G. Phillips (argued), Joseph R. Guerra, and Matthew\nD. Lerner, Sidley Austin LLP, Washington, D.C.; David R.\nCarpenter, Sidley Austin LLP, Los Angeles, California; for\nPetitioner-Appellee.\n\nChristopher J. Walker, The Ohio State University Moritz\nCollege of Law, Columbus, Ohio; Steven P. Lehotsky, U.S.\nChamber Litigation Center, Washington, D.C.; for Amicus\nCuriae The Chamber of Commerce of the United States of\nAmerica.\n\nElizabeth J. Stevens, Caplin & Drysdale, Chartered,\nWashington, D.C., for Amici Curiae H. David Rosenbloom\nand John P. Steines, Jr.\n\nAlice E. Loughran, Charles G. Cole, Michael C. Durst, and\nRobert J. Kovacev, Steptoe & Johnson LLP, Washington\nD.C., for Amici Curiae Semiconductor Industry Association,\nInformation Technology Industry Council, National Foreign\n\f4 AMAZON.COM V. CIR\n\nTrade Council, Software Finance and Tax Executives\nCouncil, and TechNet.\n\nA. Duane Webber, George M. Clarke, and Phillip J. Taylor,\nBaker & McKenzie LLP, Washington, D.C.; Mark A. Oates,\nSusan E. Ryba, and Cameron C. Reilly, Baker & McKenzie\nLLP, Chicago, Illinois; Scott H. Frewing, Baker &\nMcKenzie LLP, Palo Alto, California; for Amici Curiae\nSilicon Valley Tax Directors Group, Agilent Technologies,\nInc., Cisco Systems, Inc., Dell Technologies Inc., Dolby\nLaboratories, Inc., Expedia Group, Inc., FireEye, Inc.,\nGenesys Int’l Corp. Ltd., Informatica, Inc., NetApp, Inc.,\nPalo Alto Networks, Inc., Surveymonkey Inc., and\nVMware, Inc.\n\n\n OPINION\n\nCALLAHAN, Circuit Judge:\n\n Appellee, Amazon.com, Inc., is a U.S.-based online\nretailer with highly profitable intangible assets. In 2005 and\n2006, Amazon restructured its European businesses in a way\nthat would shift a substantial amount of its income from\nU.S.-based entities to newly created European subsidiaries.\nBecause the restructuring would allow the European entities\nto generate income using Amazon’s pre-existing intangible\nassets developed in the United States, the tax code and\ncorresponding regulations required that the European\nentities compensate Amazon for the use of assets that meet\nthe regulatory definition of an “intangible.”\n\n The compensation was provided through a cost sharing\narrangement, whereby Amazon and a holding company for\nthe European subsidiaries would be treated as co-owners of\n\f AMAZON.COM V. CIR 5\n\nthe intangibles. Under the arrangement, the holding\ncompany was required to make a “buy-in” payment for the\npre-existing intangibles Amazon contributed to the\narrangement and to make cost sharing payments going\nforward for its share of future research and development\n(R&D) efforts. The buy-in payment was taxable income to\nAmazon, and the holding company’s cost sharing payments\nwould reduce Amazon’s U.S. tax deductions for R&D costs.\n\n To guard against manipulation by jointly controlled\nentities, the regulations require that the buy-in payment\nreflect the fair market value of the pre-existing intangibles\nmade available under a cost sharing arrangement. Amazon\ninitially reported a buy-in payment of about $255 million.\nAppellant, the Commissioner of Internal Revenue,\nconcluded that the buy-in payment had not been determined\nat arm’s length in accordance with the transfer pricing\nregulations, so the IRS performed its own calculation,\nvaluing the buy-in at about $3.6 billion. Amazon filed a\npetition in the United States Tax Court challenging the IRS’s\nvaluation.\n\n In the tax court proceedings, Amazon and the\nCommissioner offered competing methods for valuing\nAmazon’s pre-existing intangibles. There was a key\ndifference between the parties’ respective approaches.\nAmazon’s methodology isolated and valued only the\nspecific intangible assets that it transferred to the European\nholding company under the cost sharing arrangement,\nincluding website technology, trademarks, and customer\nlists. The Commissioner’s methodology essentially valued\nthe entire European business, minus pre-existing tangible\nassets. That method necessarily swept into the calculation\nall contributions of value, including those that are more\nnebulous and inseparable from the business itself, like the\n\f6 AMAZON.COM V. CIR\n\nvalue of employees’ experience, education, and training\n(known as “workforce in place”), going concern value,\ngoodwill, and other unique business attributes and\nexpectancies (which the parties refer to as “growth\noptions”). The tax court sided primarily with Amazon, and\nthe Commissioner appealed.\n\n This case requires us to interpret the meaning of an\n“intangible” in the applicable (but now outdated) transfer\npricing regulations. 1 The case turns on whether, as the\nCommissioner argues, the regulatory definition is broad\nenough to include all intangible assets of value, even the\nmore nebulous ones that the Commissioner refers to as\n“residual-business assets” (i.e., Amazon’s culture of\ninnovation, the value of workforce in place, going concern\nvalue, goodwill, and growth options). We conclude that the\ndefinition does not include residual-business assets.\nAlthough the language of the definition is ambiguous, the\ndrafting history of the regulations shows that “intangible”\nwas understood to be limited to independently transferrable\nassets. We thus affirm.\n\n\n\n\n 1\n This case is governed by regulations promulgated in 1994 and\n1995. In 2009, more than three years after the tax years at issue here, the\nDepartment of Treasury issued temporary regulations broadening the\nscope of contributions for which compensation must be made as part of\nthe buy-in payment. See 74 Fed. Reg. 340 (Jan. 5, 2009). In 2017,\nCongress amended the definition of “intangible property” in 26 U.S.C.\n§ 936(h)(3)(B) (which is incorporated by reference in 26 U.S.C. § 482).\nTax Cuts and Jobs Act of 2017, Pub. L. 115-97, § 14221(a), 131 Stat.\n2054, 2218 (2017). If this case were governed by the 2009 regulations\nor by the 2017 statutory amendment, there is no doubt the\nCommissioner’s position would be correct.\n\f AMAZON.COM V. CIR 7\n\n I.\n\n Before summarizing Amazon’s corporate restructuring\nand the procedural history of this case, we first provide an\noverview of the statutory and regulatory framework for\ntransfer pricing.\n\n A.\n\n When a taxpayer sells or licenses its property, including\nintangible assets, to another entity, the purchase price or\nlicense royalty is taxable income. Often, such transactions\noccur between entities “owned or controlled directly or\nindirectly by the same interests.” 26 U.S.C. § 482. 2 The\nparties in a controlled transaction are in a position to\npotentially manipulate the terms of the transaction to\nminimize taxable income artificially. But section 482 gives\nthe Department of the Treasury the power to reallocate the\ndollar figures of such controlled transactions if necessary “to\nprevent evasion of taxes or clearly to reflect the income of”\na taxpayer. Id.; see also Comm’r v. First Sec. Bank of Utah,\nN.A., 405 U.S. 394, 400 (1972) (“[Section] 482 is designed\nto prevent ‘artificial shifting, milking, or distorting of the\ntrue net incomes of commonly controlled enterprises.’”\n(quoting B. Bittker & J. Eustice, Federal Income Taxation of\nCorporations and Shareholders p. 15–21 (3d ed. 1971)).\n\n The implementing regulations state that “[t]he purpose\nof section 482 is to ensure that taxpayers clearly reflect\nincome attributable to controlled transactions, and to prevent\nthe avoidance of taxes with respect to such transactions.”\n\n 2\n Unless otherwise indicated, references to section 482 of the\nInternal Revenue Code and the implementing regulations are to the\nversions in effect during the tax years at issue here (2005 and 2006).\n\f8 AMAZON.COM V. CIR\n\nTreas. Reg. § 1.482‑1(a)(1). A reallocation under\nsection 482 and the implementing regulations is intended to\n“place[] a controlled taxpayer on a tax parity with an\nuncontrolled taxpayer by determining the true taxable\nincome of the controlled taxpayer.” Id. The true taxable\nincome is determined as if the parties to the controlled\ntransaction had conducted their affairs in the manner of\nunrelated parties “dealing at arm’s length.” Treas. Reg.\n§ 1.482‑1(b)(1). The “arm’s length” standard is met “if the\nresults of the transaction are consistent with the results that\nwould have been realized if uncontrolled taxpayers had\nengaged in the same transaction under the same\ncircumstances.” Id.\n\n The key regulations were promulgated in 1994 and 1995.\nThe regulations divide property into two categories: tangible\nproperty (Treas. Reg. § 1.482‑3) and intangible property\n(Treas. Reg. § 1.482‑4). This case concerns intangible\nproperty. The regulations provide:\n\n (b) Definition of intangible. For purposes of\n section 482, an intangible is an asset that\n comprises any of the following items and has\n substantial value independent of the services\n of any individual—\n\n (1) Patents, inventions, formulae,\n processes, designs, patterns, or know-\n how;\n\n (2) Copyrights and literary, musical, or\n artistic compositions;\n\n (3) Trademarks, trade names, or brand\n names;\n\f AMAZON.COM V. CIR 9\n\n (4) Franchises, licenses, or contracts;\n\n (5) Methods, programs, systems,\n procedures, campaigns, surveys, studies,\n forecasts, estimates, customer lists, or\n technical data; and\n\n (6) Other similar items. For purposes of\n section 482, an item is considered similar\n to those listed in paragraph (b)(1) through\n (5) of this section if it derives its value not\n from its physical attributes but from its\n intellectual content or other intangible\n properties.\n\nTreas. Reg. § 1.482‑4(b).\n\n A controlled taxpayer may make its intangibles available\nto a foreign affiliate by entering a licensing agreement, under\nwhich the foreign affiliate’s royalty fee is taxable income to\nthe controlled taxpayer. See generally Treas. Reg. §§ 1.482-\n1, 1.482-4. The royalty fee must, of course, reflect the\n“arm’s length” value of the license. As new intangibles\ncontinue to be developed, the value of the license changes\nand becomes open to dispute for tax purposes on an ongoing\nbasis.\n\n As an alternative to licensing, the regulations authorize\njointly controlled entities to enter a cost sharing\narrangement, under which they become co-owners of\nintangibles developed as a result of the entities’ joint R&D\n\f10 AMAZON.COM V. CIR\n\nefforts. See generally Treas. Reg. § 1.482‑7A. 3 This\narrangement provides the taxpayer the benefit of certainty\nbecause, assuming the arrangement satisfies the\nrequirements of the regulations, new intangibles need not be\nvalued as they are developed.\n\n Under the regulations, a subsidiary that enters a cost\nsharing arrangement with its parent must make two distinct\npayments. First, the subsidiary must make an arm’s length\n“buy-in” payment reflecting the value of the pre-existing\nintangibles the parent contributes to the arrangement:\n\n If a controlled participant makes pre-existing\n intangible property in which it owns an\n interest available to other controlled\n participants for purposes of research in the\n intangible development area under a\n qualified cost sharing arrangement, then each\n such other controlled participant must make\n a buy-in payment to the owner.\n\nTreas. Reg. § 1.482‑7A(g)(2). The buy-in payment is\ntaxable income for the entity contributing the intangibles—\nhere, Amazon.\n\n Second, the subsidiary must pay a share of the intangible\ndevelopment costs (or R&D) “equal to its share of\nreasonably anticipated benefits attributable to such\ndevelopment.” Treas. Reg. § 1.482‑7A(a)(2). If the\ncontrolled taxpayer incurs the lion’s share of the intangible\ndevelopment costs, the subsidiary is required to make cost\n\n 3\n We follow the parties’ convention of referring to the regulations\non cost sharing arrangements by reference to the numbering as\nredesignated in 2009 (i.e., Treas. Reg. § 1.482‑7A).\n\f AMAZON.COM V. CIR 11\n\nsharing payments to reflect its share of the anticipated\nbenefits. The subsidiary’s cost sharing payments serve to\nreduce the deductions the controlled taxpayer can take for\nR&D costs (thereby increasing tax liability).\n\n B.\n\n Amazon is an online retailer that began operating in the\nUnited States in 1995. In the late 1990s and early 2000s,\nAmazon expanded operations into France, Germany, and the\nUnited Kingdom. Amazon’s business in Europe had a siloed\nstructure, with separate European subsidiaries independently\noperating and managing the business, each subsidiary\nhaving its own website, fulfillment centers, and customer\nbase. The European subsidiaries licensed the right to use\nAmazon’s website technology, customer information, and\nmarketing intangibles (including trademarks and domain\nnames).\n\n To address operational inefficiencies in Europe, in the\nearly 2000s Amazon investigated options for creating a\ncentralized European headquarters. Amazon ultimately\nlocated its new headquarters in Luxembourg, which offered\na central location, the lowest value-added tax rate in Europe,\nand a relatively low corporate tax rate. Beginning in 2004,\nAmazon undertook a series of transactions to implement its\nplan for centralizing business operations in Europe.\n\n Amazon formed Amazon Europe Holding Technologies\nSCS (“AEHT”) and transferred to AEHT the pre-existing\nEuropean subsidiaries, their operating assets, and their pre-\nexisting intangible rights (i.e., those developed in Europe).\nAmazon and AEHT also entered a cost sharing\n\f12 AMAZON.COM V. CIR\n\narrangement—the transaction most pertinent to this appeal. 4\nFor the arrangement to be a “qualified cost sharing\narrangement” under the transfer pricing regulations outlined\nabove, AEHT needed to pay Amazon for the pre-existing\nintangibles Amazon contributed to the arrangement. To\ndetermine the amount of this “buy-in,” Amazon hired a tax\nfirm, which concluded the pre-existing intangibles were\nworth $217 million. 5\n\n Under the cost sharing arrangement, AEHT also makes\ncost sharing payments to Amazon for its share of ongoing\nintangible development costs. Amazon reported cost\nsharing payments from AEHT of about $116 million for\n2005 and about $77 million for 2006.\n\n The IRS rejected Amazon’s calculation of AEHT’s buy-\nin, concluding that Amazon grossly undervalued the\nintangibles Amazon made available to AEHT. Applying a\ndiscounted-cash-flow valuation methodology, the IRS\ndetermined a buy-in payment of $3.6 billion.6 Amazon filed\na petition in the tax court challenging the IRS’s valuation.\nAmazon argued that the cost sharing arrangement covered\n\n 4\n The following transactions were also part of the restructuring:\n(1) Amazon granted AEHT a license to use Amazon’s existing\ntechnology-related intellectual property; (2) Amazon assigned to AEHT\ncustomer data and certain marketing intangibles (including trademarks,\nwebsite content, and domain names) relating to the European business;\n(3) Amazon transferred the stock of the European subsidiaries to AEHT;\n(4) Amazon transferred other business assets (other than intellectual\nproperty) to AEHT; and (5) the pre-existing European subsidiaries\nlicensed to AEHT intellectual property titled in their names.\n 5\n The tax firm set the buy-in price at $254.5 million, to be paid over\na seven-year period (resulting in a present value of $217 million).\n\n 6\n The IRS’s calculation was later reduced to $3.468 billion.\n\f AMAZON.COM V. CIR 13\n\nthree distinct groups of transferred assets—website\ntechnology, marketing intangibles, and European-customer\ninformation—that must be valued separately using a\nmethodology referred to in the transfer pricing regulations as\nthe comparable uncontrolled transaction method.\n\n After a six-week trial that included testimony and written\nreports of thirty expert witnesses, the tax court concluded\nthat the Commissioner abused his discretion in determining\nthat the discounted cash flow methodology supplied the best\nmethod for determining an arm’s length buy-in payment and\nin determining that the required payment is $3.468 billion.\nAmazon.Com, Inc. v. Comm’r, 148 T.C. 108, 150 (2017).\nRelying on the rationale of its prior decision in Veritas\nSoftware Corp. v. Commissioner, 133 T.C. 297 (2009),\nnonacq., 2010-49 I.R.B (2010), action on dec., 2010-05\n(Nov. 12, 2010), the tax court reasoned that the\nCommissioner’s valuation methodology “is based in essence\non an ‘akin to a sale’ theory” and “necessarily sweeps into\n[the] calculation assets that were not transferred under the\n[cost sharing arrangement] and assets that were not\ncompensable ‘intangibles’ to begin with.” Amazon.Com,\n148 T.C. at 156–57. The tax court concluded that “[a]n\nenterprise valuation of a business,” like the one conducted\nby the Commissioner’s primary expert, “includes many\nitems of value that are not ‘intangibles’” under the cost\nsharing regulations. Id. at 157. Such items include\n“workforce in place, going concern value, goodwill, and\nwhat trial witnesses described as ‘growth options’ and\ncorporate ‘resources’ or ‘opportunities.’” Id. The tax court\nreasoned that such items are “[u]nlike the ‘intangibles’ listed\n\f14 AMAZON.COM V. CIR\n\nin the statutory and regulatory definitions” in that they\n“cannot be bought and sold independently.” Id. 7\n\n The tax court adopted the comparable uncontrolled\ntransaction method as the best way to value the buy-in\npayment because that method allows for the intangibles at\nissue to be isolated and separately valued. Id. at 164.\nAlthough the tax court agreed with Amazon’s general\nvaluation approach, it disagreed with certain aspects of\nAmazon’s implementation of that method. Id. After making\nadjustments, the tax court calculated the value of the buy-in\npayment to be about $779 million. The Commissioner\ntimely appealed, challenging the tax court’s rejection of his\nexpert’s discounted cash flow methodology. 8\n\n II.\n\n The tax court had jurisdiction over this action under\n26 U.S.C. § 6213(a). We have jurisdiction under 26 U.S.C.\n§ 7482(a)(1), (b)(1).\n\n “[W]e review the tax court’s conclusions of law de novo\nand its factual findings for clear error.” MK Hillside\n\n 7\n The tax court made several other legal conclusions and factual\nfindings not relevant to the Commissioner’s appeal.\n 8\n Although we affirm the tax court, nothing in our decision should\nbe construed as an outright rejection of any particular valuation\nmethodology. Here, the parties’ respective valuation methods did not\ndiffer only in how they valued Amazon’s assets; they differed in what\nassets they valued. The selection of the “best method” here, see Treas.\nReg. § 1.482-1(c), thus turns on an interpretation of the regulatory\ndefinition of an “intangible.” Nothing in our opinion should be\nconstrued as favoring one valuation method over another in\ncircumstances not present here.\n\f AMAZON.COM V. CIR 15\n\nPartners v. Comm’r, 826 F.3d 1200, 1203 (9th Cir. 2016)\n(quoting DHL Corp. & Subsidiaries v. Comm’r, 285 F.3d\n1210, 1216 (9th Cir. 2002)).\n\n III.\n\n The dispositive issue in this case is whether, under the\n1994/1995 regulations, the “buy-in” required for “pre-\nexisting intangible property” must include compensation for\nresidual-business assets. 9 To answer this legal question, we\nconsider the regulatory definition of an “intangible,” the\noverall transfer pricing regulatory framework, the\nrulemaking history of the regulations, and whether the\nCommissioner’s position is entitled to deference under Auer\nv. Robbins, 519 U.S. 452 (1997). We agree with the tax court\nthat the definition of an “intangible” in § 1.482-4(b) was not\nintended to embrace residual-business assets.\n\n A.\n\n The Commissioner argues that Amazon’s valuable\nresidual-business assets meet “§ 1.482‑4(b)’s definition of\nintangibles.” “Regulations are interpreted according to the\nsame rules as statutes, applying traditional rules of\nconstruction.” Minnick v. Comm’r, 796 F.3d 1156, 1159\n(9th Cir. 2015); see Kisor v. Wilkie, 139 S. Ct. 2400, 2415\n(2019) (recognizing that “all the ‘traditional tools’ of\nconstruction” are the same for both statutes and regulations\n\n 9\n The Commissioner also challenges some of the tax court’s other\nreasons for rejecting the valuation conducted by the Commissioner’s\nexpert. But in each of his other challenges, the Commissioner’s\nargument assumes he is correct on his primary contention that the\ndefinition of “intangible” in the 1994/1995 regulations embraces\nresidual-business assets. Because we reject the Commissioner’s view on\nthe primary issue, we need not address his other contentions.\n\f16 AMAZON.COM V. CIR\n\n(quoting Chevron, U.S.A., Inc. v. Nat. Res. Def. Council,\nInc., 467 U.S. 837, 843 n.9 (1984))). Our “legal toolkit”\nincludes careful examination of “the text, structure, history,\nand purpose of a regulation.” Kisor, 139 S. Ct. at 2415.\n\n We begin with the language of § 1.482-4(b). If the\nregulation is unambiguous, its plain meaning governs. Safe\nAir For Everyone v. EPA, 488 F.3d 1088, 1097 (9th Cir.\n2007).\n\n The regulation defines an “intangible” as an asset that\nboth “has substantial value independent of the services of\nany individual” and is one of the items listed in subsection\n(b)(1)–(6). Treas. Reg. § 1.482‑4(b). Each of the 28 specific\nitems in subsection (b) is independently transferrable—none\nis a residual-business asset. The Commissioner thus relies\non the catchall provision for “[o]ther similar items.” Treas.\nReg. § 1.482‑4(b)(6). “[A]n item is considered similar” to\nthe other items in the subsection “if it derives its value not\nfrom its physical attributes but from its intellectual content\nor other intangible properties.” Id.\n\n Reading the catchall provision together with the\nintroductory language of the definition, residual-business\nassets are intangibles if they (1) have substantial value\nindependent of the services of any individual and (2) derive\ntheir value from intellectual content or other intangible\nproperties. The Commissioner argues both elements are\nsatisfied. First, he argues that Amazon’s growth options\n“derive their value from intangible, rather than physical,\nattributes.” He then cites testimony from Amazon’s expert\n(Dr. Bradford Cornell) that Amazon’s growth options are\nprimarily attributable to its culture of innovation. Second,\nhe argues that the value of Amazon’s growth options is\nindependent of the services of any individual because they\n\f AMAZON.COM V. CIR 17\n\nare “part of the culture of Amazon to be able to have creative\nideas bubble up in their organization and actually use them.”\n\n Amazon argues that the Commissioner’s interpretation\nof the catchall provision is too sweeping for several reasons.\nAmazon’s central argument is that to qualify as an\n“intangible” under the regulation, an item must be capable\nof being bought and sold independently of the business—\nand residual-business assets are inseparable from the\nbusiness. The tax court agreed. See Amazon.Com, 148 T.C.\nat 157 (concluding that, unlike the specific intangibles listed\nin the regulation, “workforce in place, going concern value,\ngoodwill, and what trial witnesses described as ‘growth\noptions’ and corporate ‘resources’ or ‘opportunities’ . . .\ncannot be bought and sold independently”).\n\n Amazon offers several arguments in support of its\nposition that assets that are inseparable from the business do\nnot meet the regulatory definition of an “intangible.”\nQuoting Arcadia v. Ohio Power Co., 498 U.S. 73, 78 (1990),\nAmazon argues that the Commissioner’s interpretation\nrenders the “enumeration of specific subjects entirely\nsuperfluous—in effect adding to that detailed list ‘or\nanything else.’” Amazon invokes the canon ejusdem\ngeneris—of the same kind or class—and asserts that “[t]he\ncommon attribute of all 28 specified items is that they can be\nsold independently.” Amazon argues this canon should be\napplied to prevent the catchall clause from swallowing the\npreceding language of the regulation.\n\n Amazon’s focus on the commonality of the 28 specified\nitems has some force. After all, if all 28 listed items share a\ncommon attribute, why would anyone understand a catchall\nfor “similar items” to include a non-listed item that doesn’t\nshare that attribute? Amazon’s commonality argument\nfalters, however, because the catchall provision did not\n\f18 AMAZON.COM V. CIR\n\nsimply say “[o]ther similar items.” Instead, the catchall\nelaborated by explaining how an item is determined to be\n“similar” to the other items: “if it derives its value not from\nits physical attributes but from its intellectual content or\nother intangible properties.” Treas. Reg. § 1.482-4(b)(6).\nThis explanation of what is “similar” leaves open the\npossibility of a non-listed item being included in the\ndefinition even if it doesn’t share the attribute of being\nseparately transferrable.\n\n Amazon also argues that the regulation’s requirement\nthat an intangible have “substantial value independent of the\nservices of any individual,” Treas. Reg. § 1.482‑4(b),\nsupports its position. The tax court in Veritas agreed. In that\ncase, the tax court rejected the Commissioner’s argument\nthat a foreign subsidiary’s buy-in under a cost sharing\narrangement must include the value of the subsidiary’s\naccess to the U.S. corporate parent’s R&D and marketing\nteams—i.e., workforce in place. Veritas, 133 T.C. at 323.\nAlthough the court found insufficient evidence that the cost\nsharing arrangement made such a transfer, it also stated that\n“[e]ven if such evidence existed, these items would not be\ntaken into account in calculating the requisite buy-in\npayment because they do not have ‘substantial value\nindependent of the services of any individual’ and thus do\nnot meet the” definition of “intangible.” Id. at 323 n.31\n(quoting Treas. Reg. § 1.482‑4(b)). The court reasoned that\nthe value of access to the parent’s “R&D and marketing\nteams is based primarily on the services of individuals (i.e.,\nthe work, knowledge, and skills of team members).” Id. The\nCommissioner clearly disagrees with Veritas on this point,\nbut he doesn’t explain why the tax court’s analysis is wrong.\n\n Analysis of the regulatory text alone does not\ndefinitively resolve the question here. The definition of an\n\f AMAZON.COM V. CIR 19\n\n“intangible” is susceptible to, but does not compel, an\ninterpretation that embraces residual-business assets. The\nproblem is that residual-business assets, such as “growth\noptions” and a “culture of innovation,” are amorphous, and\nit’s not self-evident whether such assets have “substantial\nvalue independent of the services of any individual.” See\nTreas. Reg. § 1.482‑4(b). Amazon raises legitimate\nconcerns about the regulation’s catchall being stretched too\nfar, and those concerns likely bear on which party has the\nmore reasonable view of the regulatory definition. It\nnonetheless remains that the definition of “intangible” could\nbe construed as covering residual-business assets if the\nlanguage of § 1.482‑4(b) is viewed in isolation. 10\n\n B.\n\n But we are required to look at the regulatory scheme “as\na whole,” viewing the regulatory “definition in the context\nof the entire [transfer pricing] regulations.” Alaska Trojan\nP’ship v. Gutierrez, 425 F.3d 620, 628 (9th Cir. 2005). The\nCommissioner argues that other regulations governing\ntransfer pricing support his view that the regulatory\ndefinition of an “intangible” embraces residual-business\n\n 10\n The Commissioner also argues that the definition of an\n“intangible” includes residual-business assets because an uncontrolled\nparty would pay for access to those assets in an arm’s length transaction.\nHe cites the testimony of Amazon’s expert that parties dealing at arm’s\nlength “[d]efinitely” pay for “growth options” because “[n]o company is\ngoing to give away something of value without compensation.” The\nCommissioner’s argument misses the mark. Under the regulations, the\narm’s length standard governs the valuation of intangibles; it doesn’t\nanswer whether an item is an intangible. The definition of an\n“intangible” is provided in § 1.482‑4(b). The Commissioner points to no\nlanguage in the statute or regulations suggesting that the definition of\nwhat constitutes an intangible is determined by asking whether an\nuncontrolled party would pay for it.\n\f20 AMAZON.COM V. CIR\n\nassets. He cites two sections in the regulations, §§ 1.482‑7A\nand 1.482‑1.\n\n Section 1.482‑7A specifies the requirements of a\nqualified cost sharing arrangement and the methods for\ndetermining the taxable income resulting from such an\narrangement. The Commissioner cites the subsection that\nimposes the “buy-in” requirement where “[a] controlled\nparticipant . . . makes intangible property available to a\nqualified cost sharing arrangement.” Treas. Reg.\n§ 1.482‑7A(g)(1). The Commissioner argues that this\nprovision mandates that residual-business assets be paid for\n“if they are made ‘available’ to the cost-sharing participants”\neven though such assets generally cannot be transferred\nindependently from the business.\n\n The Commissioner’s argument based on\n§ 1.482‑7A(g)(1) presupposes the very point he attempts to\nprove—that residual-business assets are “intangible\nproperty” within the meaning of the regulations. The\nprovision requiring a “buy-in” does not expand but instead\nincorporates the meaning of an “intangible” given in\n§ 1.482-4(b). See Treas. Reg. § 1.482‑7A(g)(2) (“The buy-\nin payment by each such other controlled participant is the\narm’s length charge for the use of the intangible . . . .”\n(emphasis added)). The “makes . . . available” language thus\nprovides no meaningful insight into the regulatory definition\nof an “intangible.” 11\n\n\n 11\n Considered in context of the regulatory framework, it appears that\nthe purpose of the “makes . . . available” phrasing in § 1.482‑7A(g) is to\naccount for taxpayers who might seek to avoid the requirements of the\ntransfer pricing regulations by not formally transferring their intangible\nassets to a related party and instead informally making available the\n\f AMAZON.COM V. CIR 21\n\n The Commissioner also finds support for his position in\nthe regulations’ preamble stating that “[t]he purpose of\nsection 482 is to ensure that taxpayers clearly reflect income\nattributable to controlled transactions and to prevent the\navoidance of taxes with respect to such transactions.” Treas.\nReg. § 1.482‑1(a)(1). Relying on Xilinx, Inc. v.\nCommissioner, 598 F.3d 1191 (9th Cir. 2010), the\nCommissioner argues that “anything of value that is made\navailable between related parties must be paid for” in the\nbuy-in, regardless of whether it is defined as an intangible.\n\n In Xilinx, we considered whether the parties to a cost\nsharing arrangement “must include the value of certain stock\noption compensation one participant gives to its employees\nin the pool of costs to be shared.” Id. at 1192. We found\ntwo provisions of the regulations in conflict. We first cited\n§ 1.482‑1(b)(1)’s statement that “the standard to be applied\nin every case is that of a taxpayer dealing at arm’s length\nwith an uncontrolled taxpayer.” Id. at 1195 (quoting Treas.\nReg. § 1.482‑1(b)(1)). It was undisputed on appeal that\nunrelated parties would not have shared the employee stock\noption costs. Id. at 1194. Another provision, however,\nrequired that “controlled parties in a cost sharing agreement\n. . . share all ‘costs . . . related to the intangible development\narea,’” which by definition would include the employee\nstock options. Id. at 1196. In resolving this conflict, the\nXilinx majority affirmed the tax court’s conclusion that the\nstock options need not be paid for as part of the buy-in. Id.\n\n\n\n\nassets. The regulations thus treat such taxpayers “as having transferred\ninterests in such [intangible] property.” Treas. Reg. § 1.482‑7A(g)(1).\n\f22 AMAZON.COM V. CIR\n\nat 1196–97. The primary opinion reasoned that “[p]urpose\nis paramount.” Id. at 1196. 12\n\n Relying on the logic of the primary opinion in Xilinx, the\nCommissioner asserts that “it is undisputed that a company\nentering into the same transaction under the same\ncircumstances with an unrelated party would have required\ncompensation.” To evaluate this claim, it’s important to be\nclear about what exactly is meant by the phrase “the same\ntransaction under the same circumstances.” The\nCommissioner relies on deposition testimony from one of\nAmazon’s experts that parties dealing at arm’s length would\npay for growth options. But the expert explained at trial the\ndifference between an investor or purchaser of the entire\nbusiness (who would pay for the full value of the business)\nand a partner (who would not). The question becomes\nwhether a cost sharing arrangement is akin to the sale of a\nbusiness or like a partnership in certain assets or aspects of\nthe business. The Commissioner assumes, but does not\nexplain why, the transfer of intangible assets under\n\n\n\n 12\n Caution should be taken before relying on the rationale of the\nprimary opinion in Xilinx. One panel member concurred—thus\nproviding a majority vote for the disposition—but wrote separately “to\nexplain [his] particular reasons for rejecting the Commissioner’s\nposition.” Id. at 1197 (Fisher, J., concurring). Although the reasoning\nof the concurring opinion overlaps somewhat with the reasoning of the\nprimary opinion, the concurrence does not seem to adopt the “[p]urpose\nis paramount” logic. Instead, based on the language of the regulations,\nthe legislative and regulatory history, international tax treaties, and the\nunderstanding of the business community and tax professionals (i.e.,\namici curiae), the concurrence concluded that the taxpayer’s\n“understanding of the regulations is the more reasonable even if the\nCommissioner’s current interpretation may be theoretically plausible.”\nId. at 1198.\n\f AMAZON.COM V. CIR 23\n\nAmazon’s cost sharing arrangement with AEHT should be\ntreated the same as the sale of the business.\n\n The Commissioner’s reliance on Xilinx thus suffers the\nsame defect as his “made available” argument based on\n§ 1.482-7A(g)—he assumes the very conclusion he’s aiming\nto prove. Although the regulatory provisions the\nCommissioner cites are consistent with his position, they do\nnot provide independent support and they are likewise\nconsistent with Amazon’s view.\n\n If the cost sharing regulations as a whole tip the scale\neither direction, they tend to favor Amazon on the issue\npresented here. The regulations describe a cost sharing\narrangement as an agreement “to share the costs of\ndevelopment of one or more intangibles.” Treas. Reg.\n§ 1.482‑7A(a)(1). The written document memorializing the\narrangement must, among other things, describe both “the\nscope of the research and development to be undertaken,\nincluding the intangible or class of intangibles intended to be\ndeveloped.” Treas. Reg. § 1.482‑7A(b)(4)(iii). The writing\nmust also describe “each participant’s interest in any covered\nintangibles,” which is “any intangible property that is\ndeveloped as a result of the research and development\nundertaken under the cost sharing arrangement.” Treas.\nReg. § 1.482‑7A(b)(4)(iv). By identifying intangibles as\nbeing the product of R&D efforts, the regulations seem to\ncontemplate a meaning of “intangible” that excludes items\nlike goodwill and going concern value, which “are generated\nby earning income, not by incurring deductions.” Staff of J.\nComm. on Taxation, 98th Cong., General Explanation of the\nRevenue Provisions of the Deficit Reduction Act of 1984,\n428 (Comm. Print 1984).\n\f24 AMAZON.COM V. CIR\n\n The overall regulatory scheme doesn’t definitively\nresolve the issue, but it favors Amazon more than the\nCommissioner.\n\n C.\n\n We next turn to the drafting history of the regulatory\ndefinition. Both parties claim support for their respective\npositions in the historical development of the regulations.\n\n Treasury first defined intangible property for purposes of\nsection 482 in regulations adopted in 1968:\n\n [I]ntangible property shall consist of the\n items described in subdivision (ii) of this\n subparagraph, provided that such items have\n substantial value independent of the services\n of individual persons.\n\n (ii) The items referred to in subdivision (i) of\n this subparagraph are as follows:\n\n (a) Patents, inventions, formulas,\n processes, designs, patterns, and other\n similar items;\n\n (b) Copyrights, literary, musical, or\n artistic compositions, and other similar\n items;\n\n (c) Trademarks, trade names, brand\n names, and other similar items;\n\n (d) Franchises, licenses, contracts, and\n other similar items;\n\f AMAZON.COM V. CIR 25\n\n (e) Methods, programs, systems,\n procedures, campaigns, surveys, studies,\n forecasts, estimates, customer lists,\n technical data, and other similar items.\n\n33 Fed. Reg. 5848, 5854 (Apr. 16, 1968).\n\n Fourteen years later, Congress enacted the Tax Equity\nand Fiscal Responsibility Act of 1982, adopting a definition\nof intangible property in section 936 of the Internal Revenue\nCode that was nearly identical to the one in the regulations\nimplementing section 482. 13 The Commissioner plucks a\nphrase from a Senate Report for the 1982 Act suggesting that\nthe statute “defines intangible assets broadly.” With more\ncontext, the quoted sentence from the Senate Report states\nthat “[t]he bill defines intangible assets broadly to include”\nthe 28 specifically-listed items “and other items similar to\nany of those listed, so long as the item has substantial value\nindependent of the services of individual persons.” S. Rep.\n97-494, at 161 (1982), as reprinted in 1982 U.S.C.C.A.N.\n781, 924 (emphasis added). 14\n\n\n 13\n The new statutory definition differed from the regulatory one in\nfour, relatively minor respects: (1) making all enumerated items singular;\n(2) adding “know-how” to the first category; (3) inserting “any similar\nitem” as a stand-alone category in lieu of “other similar items” in each\ncategory; and (4) requiring that an item have “substantial value\nindependent of the services of any individual” rather than “of individual\npersons.” Tax Equity & Fiscal Responsibility Act of 1982, Pub. L. 97-\n248, § 213, 96 Stat. 324 (Sept. 3, 1982).\n\n 14\n Another part of the quoted Senate Report states that the\nCommittee viewed the bill as combatting the practice of transferring\nintangibles “created, developed or acquired in the United States” to\nforeign entities to generate income tax-free. S. Rep. 97-494, at 158–59\n(1982), as reprinted in 1982 U.S.C.C.A.N. 781, 921–22. It seems from\n\f26 AMAZON.COM V. CIR\n\n The Tax Reform Act of 1986 amended section 482 to\nincorporate the “intangible property” definition from section\n936: “In the case of any transfer (or license) of intangible\nproperty (within the meaning of section 936(h)(3)(B)), the\nincome with respect to such transfer or license shall be\ncommensurate with the income attributable to the\nintangible.” Pub. L. 99–514, § 1231, 100 Stat. 2085 (Oct.\n22, 1986).\n\n When it amended section 482, Congress requested that\nthe IRS conduct “a comprehensive study of intercompany\npricing rules” and report on whether the existing regulations\n“could be modified in any respect.” H.R. Rep. No. 99-841,\nat II-637 (1986). The resulting IRS report, known as the\n“White Paper,” laid the groundwork for what would\nultimately become the 1994/1995 regulations. A Study of\nIntercompany Pricing Under Section 482 of the Code, I.R.S.\nNotice 88-123, 1988-2 C.B. 458.\n\n The White Paper discussed, among other things, cost\nsharing arrangements. According to the White Paper, the\npurpose of the buy-in requirement is for “a party to a cost\nsharing arrangement that has contributed funds or incurred\nrisks for development of intangibles at an earlier stage” to be\n“compensated by the other participants.” Id. at 497. “[I]f\nthere are intangibles that are not fully developed that relate\nto the research to be conducted under the cost sharing\narrangement, it is necessary to value them in order to\ndetermine an appropriate buy-in payment.” Id.\n\n\n\n\nthis language, the Committee didn’t contemplate intangibles that are not\nindependently transferrable.\n\f AMAZON.COM V. CIR 27\n\n The White Paper proposed “three basic types of\nintangibles” that would be subject to the buy-in requirement:\n\n • “preexisting intangibles at various stages of\n development that will become subject to the\n arrangement”;\n\n • “basic research not associated with any product”; and\n\n • “a going concern value associated with a\n participant’s research facilities and capabilities that\n will be utilized.”\n\nId. Although the White Paper proposed including going\nconcern value of a research facility in the buy-in, after\nreceiving opposition in public comments, Treasury proposed\nnew regulations that essentially retained the definition of\n“intangible” from before without referencing going concern\nvalue or any other residual-business asset. See 57 Fed. Reg.\n3571, 3579 (Jan. 30, 1992).\n\n In 1993, Treasury issued revised temporary and\nproposed regulations that defined an “intangible” as “any\ncommercially transferable interest” in the intangibles listed\nin § 936(h)(3)(B) that had “substantial value independent of\nthe services of any individual.” 58 Fed. Reg. 5263, 5287\n(Jan. 21, 1993). Treasury also requested comment on\n“whether the definition of intangible property . . . should be\nexpanded to include items not normally considered to be\nitems of intellectual property, such as work force in place,\ngoodwill or going concern value.” 58 Fed. Reg. 5310, 5312\n(Jan. 21, 1993). Amazon cites opposition comments\nsubmitted in response to Treasury’s request for comment.\n\n In 1994, Treasury issued final regulations (the ones\napplicable here), which reflected a minor reworking of the\n\f28 AMAZON.COM V. CIR\n\ndefinition. 59 Fed. Reg. 34971, 35016 (Jul. 8, 1994).\nTreasury explained that the revised definition omitted the\n“commercially transferrable” language that appeared in the\ntemporary regulations “because it was superfluous: if the\nproperty was not commercially transferrable, then it could\nnot have been transferred in a controlled transaction.” Id.\nat 34983. Treasury also explained that the revision\n“clarified” that the phrase “other similar items” in the\ndefinition “refer[s] to items that derive their value from\nintellectual content or other intangible properties rather than\nphysical attributes.” Id. In 1995, Treasury issued final\nregulations governing cost sharing arrangements, including\n§ 1.482‑7A(g)’s “buy-in” requirement.\n\n The drafting history of the transfer pricing regulations\ndoes not support the Commissioner’s argument that the\ndefinition of an “intangible” covered residual-business\nassets. The only references in the drafting history to any\nresidual-business assets suggest that such items were\nexcluded from the definition of intangible assets. The IRS’s\n1988 White Paper proposed including “going concern value”\nof a research facility in the buy-in, but Treasury’s 1994/1995\nregulations kept essentially the same definition as before\nwithout referring to “going concern value” or any other\nresidual-business asset. See 57 Fed. Reg. at 3579.\n\n Two key statements by Treasury in the drafting history\nrender the Commissioner’s current position untenable. First,\nin 1993, Treasury confirmed that the then-existing definition\nof “intangible” did not include residual-business assets when\nit asked for comments on whether the definition of\nintangibles “should be expanded to include items not\nnormally considered to be items of intellectual property,\nsuch as work force in place, goodwill or going concern\nvalue.” 58 Fed. Reg. at 5312 (emphasis added). Second, a\n\f AMAZON.COM V. CIR 29\n\nyear later after opting against such an expansion, and instead\nretaining the same essential definition from before\n(including the same list of 28 items), Treasury explained that\nthe final (1994) rule merely “clarified” when an item would\nbe deemed similar to the 28 items listed in the definition.\n59 Fed. Reg. at 34983.\n\n The Commissioner is thus forced to argue that what\nTreasury explicitly confirmed would not be considered an\n“intangible” without a substantive “expan[sion]” of the\ndefinition was implicitly added to the definition through a\nnon-specific “clarifi[cation].” The Commissioner’s\nargument stretches “clarification” beyond its commonly\nunderstood meaning of merely clearing up what was\npreviously ambiguous or otherwise restating a standard\nconsistent with what was previously intended. Cf. Motorola,\nInc. v. Fed. Exp. Corp., 308 F.3d 995, 1007 (9th Cir. 2002)\n(distinguishing between a “clarifying amendment” and “one\nthat work[s] a substantive change” (emphasis omitted)). 15\n\n Another statement from the drafting history of § 1.482‑4\nlends further support for Amazon’s position that\n“intangible” has always been understood to be limited to\nassets that are independently transferrable. Treasury’s\ntemporary regulations in 1993 defined an “intangible” as\n“any commercially transferable interest” in the intangibles\nlisted in section 936(h)(3)(B) that had “substantial value\nindependent of the services of any individual.” 58 Fed. Reg.\nat 5287. When Treasury left out the “commercially\n 15\n Amazon and amici curiae also argue that if the Commissioner is\ncorrect that the non-specific “clarifi[cation]” of § 1.482-4(b)’s catchall\nsubstantively expanded the definition of an “intangible,” then\nTreasury/IRS violated the Administrative Procedures Act. We need not\naddress this argument because we reject the Commissioner’s post hoc\ninterpretation of the changes to the regulatory definition.\n\f30 AMAZON.COM V. CIR\n\ntransferrable” language from the final regulations issued a\nyear later, it explained that the omitted language was\n“superfluous” because “if the property was not\ncommercially transferrable, then it could not have been\ntransferred in a controlled transaction.” 59 Fed. Reg. at\n34983. This is consistent with the basic premise of the\ntransfer pricing regulations, which contemplate a situation in\nwhich particular assets are transferred from one entity to\nanother.\n\n The Commissioner now contends that the 1994\ndefinition was intended to embrace residual-business assets\neven though such assets “cannot be transferred\nindependently.” Yet the Commissioner fails to identify any\ncontemporaneous statement by the agency that would\n“display awareness” that it was changing its position on\nwhether residual-business assets are included within the\ndefinition of intangibles. See FCC v. Fox Television\nStations, Inc., 556 U.S. 502, 515 (2009) (“[T]he requirement\nthat an agency provide reasoned explanation for its action\nwould ordinarily demand that it display awareness that\nit is changing position. An agency may not, for example,\ndepart from a prior policy sub silentio or simply disregard\nrules that are still on the books.”).\n\n The Commissioner argues his position is supported by a\ndifferent regulatory scheme governing penalties for\ntaxpayers who substantially misstate the value of property\non their tax returns. See 26 U.S.C. § 6662(e). In 1993,\nTreasury proposed regulations “designed to encourage\ntaxpayers to document their transfer pricing transactions and\nto provide that documentation to the [IRS] upon request.”\n58 Fed. Reg. 5304, 5304 (Jan. 21, 1993). To that end, the\nproposed regulations provided rules for determining whether\na taxpayer has substantially misstated the value of property.\n\f AMAZON.COM V. CIR 31\n\nThose rules defined “property” to include “intangible\nproperty,” which in turn was defined to include “property\nsuch as goodwill, covenants not to compete, leaseholds,\npatents, contract rights, debts, and choices in action.” Id.\nat 5306.\n\n The Commissioner’s attempt to bootstrap § 1.6662-5’s\nreference to “goodwill” ignores that the proposed regulation\nconcerning substantial valuation misstatements was issued\non the same day Treasury separately confirmed that the\ndefinition of “intangible” for purposes of the regulations\nimplementing section 482 did not include goodwill or other\nresidual-business assets. 58 Fed. Reg. at 5312. Considering\nthe context, Treasury’s use of “goodwill” in § 1.6662-5 but\nnot in § 1.482‑4(b) most likely evinces an intent not to\ninclude goodwill or other residual-business assets within\n§ 1.482‑4(b)’s definition of “intangible.”\n\n Amazon points to other Treasury regulations that define\ncertain covered property by incorporating the definition of\nintangible property under section 936(h)(3)(B) and then\nadding goodwill and going concern value. Treas. Reg.\n§§ 1.954-2(e)(3)(iv), 1.861-9T(h)(1)(ii). These provisions\nadd little to the discussion, but the express inclusion of\ngoodwill and going concern value in these regulations is\nconsistent with Amazon’s position that the definition of\n“intangible” in section 936(h)(3)(B) and § 1.482‑4(b) was\nunderstood to exclude goodwill and going concern value.\nThese regulations also show that Treasury “clearly knew\nhow to write its regulations” to include goodwill and other\nresidual-business assets. Karczewski v. DCH Mission Valley\nLLC, 862 F.3d 1006, 1015–16 (9th Cir. 2017).\n\f32 AMAZON.COM V. CIR\n\n The drafting history of § 1.482‑4(b) strongly supports\nAmazon’s position that Treasury limited the definition of\n“intangible” to independently transferrable assets. 16\n\n D.\n\n The Commissioner next argues that the tax court should\nhave deferred to the IRS’s interpretation of its own\nregulations. Under certain circumstances, an agency’s\ninterpretation of its own regulations “must be given\n‘controlling weight’” if it is not “‘plainly erroneous or\ninconsistent with the regulation.’” Stinson v. United States,\n508 U.S. 36, 45 (1993) (quoting Bowles v. Seminole Rock &\nSand Co., 325 U.S. 410, 414 (1945)); see also Auer v.\nRobbins, 519 U.S. 452, 461–62 (1997). This is frequently\nreferred to as Auer deference.\n\n The Supreme Court “has cabined Auer’s scope in varied\nand critical ways.” Kisor, 139 S. Ct. at 2418. “First and\nforemost, a court should not afford Auer deference unless the\n 16\n In 2017, Congress amended the definition of “intangible property”\nin section 936(h)(3)(B). That amendment added “goodwill, going\nconcern value, or workforce in place” to the list of specific items\nincluded in the definition of “intangible property.” Tax Cuts & Jobs Act\nof 2017, Pub. L. 115-97, § 14221(a), 131 Stat. 2054, 2218 (2017).\nCharacterizing the change as merely a “clarification,” the Commissioner\nargues that the 2017 amendment supports his proffered interpretation of\nthe 1994/1995 regulations. But the Commissioner’s post-hoc label of\nCongress’s amendment is not controlling, see Beaver v. Tarsadia Hotels,\n816 F.3d 1170, 1186 (9th Cir. 2016) (“Post-hoc labeling as a\n‘clarification’ by bill supporters of what otherwise appears to be a change\n. . . is not controlling . . . .”), and it’s also not supported by Congress’s\nown words. Congress stated that the amendment should not be\n“construed to create any inference” as to the definition of intangibles for\ntaxable years occurring before the amendment’s effective date. 131 Stat.\nat 2219. Congress said nothing to indicate that the amendment was\nmeant only as a clarification.\n\f AMAZON.COM V. CIR 33\n\nregulation is genuinely ambiguous.” Id. at 2415. Genuine\nambiguity is not determined by examination of the\nregulatory text alone. Instead, “before concluding that a rule\nis genuinely ambiguous, a court must exhaust all the\n‘traditional tools’ of construction,” “‘carefully\nconsider[ing]’ the text, structure, history, and purpose of a\nregulation, in all the ways it would if it had no agency to fall\nback on.” Id. (first quoting Chevron, 467 U.S. at 843 n.9;\nthen quoting Pauley v. BethEnergy Mines, Inc., 501 U.S.\n680, 707 (1991) (Scalia, J., dissenting))).\n\n The text of the regulatory definition of “intangible,” the\ndefinition’s place within the transfer pricing regulations\ngenerally, and the rulemaking history leave little room for\nthe Commissioner’s proffered meaning. But even if there\nwere genuine ambiguity, there is a separate reason Auer\ndeference is not warranted here.\n\n “[N]ot every reasonable agency reading of a genuinely\nambiguous rule should receive Auer deference.” Kisor,\n139 S. Ct. at 2416. Instead, “a court must make an\nindependent inquiry into whether the character and context\nof the agency interpretation entitles it to controlling weight.”\nId. (citing Christopher v. SmithKline Beecham Corp.,\n567 U.S. 142, 155 (2012)). For example, courts will not\ndefer to an agency’s interpretation when doing so “would\nseriously undermine the principle that agencies should\nprovide regulated parties fair warning of the conduct [a\nregulation] prohibits or requires.” Barboza v. Cal. Ass’n of\nProf’l Firefighters, 799 F.3d 1257, 1267 (9th Cir. 2015)\n(alteration in original) (quoting Christopher, 567 U.S.\nat 156). This exception accounts for the “risk that agencies\nwill promulgate vague and open-ended regulations that they\ncan later interpret as they see fit, thereby ‘frustrat[ing] the\nnotice and predictability purposes of rulemaking.’”\n\f34 AMAZON.COM V. CIR\n\nChristopher, 567 U.S. at 158 (quoting Talk Am., Inc. v.\nMichigan Bell Tel. Co., 564 U.S. 50, 68 (2011) (alteration in\noriginal) (Scalia, J., concurring)).\n\n Christopher and Barboza thus teach that the timing of an\nagency’s first announcement of its interpretation may be\ndispositive on whether the agency’s view will be given Auer\ndeference. Here, the Commissioner does not identify a\nspecific document (e.g., policy manual or court brief)\ndefinitively expressing the agency’s view of its regulations.\nIt thus appears that the Commissioner’s court briefs in this\ncase present Treasury’s “first announce[ment of] its view,”\nsee Christopher, 567 U.S. at 153, that the definition of\nintangible in § 1.482-4(b) embraces residual-business assets.\nThe exception to Auer deference from Christopher and\nBarboza therefore applies. “Where an agency announces its\ninterpretation for the first time in an enforcement\nproceeding, and has not previously taken any action to\nenforce that interpretation, ‘the potential for unfair surprise\nis acute.’” Barboza, 799 F.3d at 1267 (quoting Christopher,\n567 U.S. at 158). Even if first arising before the current\nlitigation, a new interpretation is owed no deference if it\nwould “create[] ‘unfair surprise’ to regulated parties.”\nKisor, 139 S. Ct. at 2417–18 (quoting Long Island Care at\nHome, Ltd. v. Coke, 551 U.S. 158, 170 (2007)). No\nstatement from Treasury in the drafting history of the\n1994/1995 regulations expresses the position the\nCommissioner advances now. Indeed, as discussed above,\nTreasury’s contemporaneous explanations of the regulations\nare to the contrary. Amazon and other taxpayers were thus\nnot given fair warning of the Commissioner’s current\ninterpretation of the regulatory definition of an “intangible.”\nThat interpretation is not entitled to deference.\n\f AMAZON.COM V. CIR 35\n\n IV.\n\n The Commissioner’s calculation of AEHT’s buy-in\nunder the cost sharing arrangement included residual-\nbusiness assets as part of Amazon’s pre-existing intangibles.\nThe language of the (now-superseded) regulatory definition\nof an “intangible” is ambiguous and could be construed as\nincluding residual-business assets. But the drafting history\nof the regulations and other indicators of Treasury’s\ncontemporaneous intent strongly favor Amazon’s proffered\nmeaning—that intangibles were limited to independently\ntransferrable assets. Treasury appears to have changed its\nposition on the meaning of the regulation after Amazon and\nAEHT entered into their cost sharing arrangement. We share\nthe sentiment reflected in the concurring opinion in Xilinx:\n\n Indeed, I am troubled by the complex,\n theoretical nature of many of the\n Commissioner’s arguments trying to\n reconcile the two regulations. Not only does\n this make it difficult for the court to navigate\n the regulatory framework, it shows that\n taxpayers have not been given clear, fair\n notice of how the regulations will affect\n them.\n\nXilinx, 598 F.3d at 1198 (Fisher, J., concurring). We\ntherefore agree with the tax court that the former regulatory\ndefinition of an “intangible” does not include residual-\nbusiness assets.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: amazon.com Inc. & Subsidiaries v. Cir\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-16\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nAMAZON.COM, INC. & SUBSIDIARIES, No. 17-72922\\n Petitioner-Appellee,\\n Tax Ct. No.\\n v. 31197-12\\n\\nCOMMISSIONER OF INTERNAL\\nREVENUE, OPINION\\n Respondent-Appellant.\\n\\n Appeal from a Decision of the\\n United States Tax Court\\n\\n Argued and Submitted April 12, 2019\\n Seattle, Washington\\n\\n Filed August 16, 2019\\n\\n Before: William A. Fletcher, Consuelo M. Callahan,\\n and Morgan Christen, Circuit Judges.\\n\\n Opinion by Judge Callahan\\n\\f2 AMAZON.COM V. CIR\\n\\n SUMMARY *\\n\\n\\n Tax\\n\\n The panel affirmed the Tax Court’s decision on a petition\\nfor redetermination of federal income tax deficiencies, in an\\nappeal involving the regulatory definition of intangible\\nassets and the method of their valuation in a cost-sharing\\narrangement.\\n\\n In the course of restructuring its European businesses in\\na way that would shift a substantial amount of income from\\nU.S.-based entities to the European subsidiaries, appellee\\nAmazon.com, Inc. entered into a cost sharing arrangement\\nin which a holding company for the European subsidiaries\\nmade a “buy-in” payment for Amazon’s assets that met the\\nregulatory definition of an “intangible.” See 26 U.S.C.\\n§ 482. Tax regulations required that the buy-in payment\\nreflect the fair market value of Amazon’s pre-existing\\nintangibles. After the Commissioner of Internal Revenue\\nconcluded that the buy-in payment had not been determined\\nat arm’s length in accordance with the transfer pricing\\nregulations, the Internal Revenue Service performed its own\\ncalculation, and Amazon filed a petition in the Tax Court\\nchallenging that valuation.\\n\\n At issue is the correct method for valuing the pre-\\nexisting intangibles under the then-applicable transfer\\npricing regulations. The Commissioner sought to include all\\nintangible assets of value, including “residual-business\\nassets” such as Amazon’s culture of innovcation, the value\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f AMAZON.COM V. CIR 3\\n\\nof workforce in place, going concern value, goodwill, and\\ngrowth options. The panel concluded that the definition of\\n“intangible” does not include residual-business assets, and\\nthat the definition is limited to independently transferrable\\nassets.\\n\\n\\n COUNSEL\\n\\nJudith A. Hagley (argued), Gilbert S. Rothenberg, and\\nArthur T. Catterall, Attorneys, Tax Division; Richard E.\\nZuckerman, Principal Deputy Assistant Attorney General;\\nTravis A. Greaves, Deputy Assistant Attorney General; Tax\\nDivision, United States Department of Justice, Washington,\\nD.C.; for Respondent-Appellant.\\n\\nCarter G. Phillips (argued), Joseph R. Guerra, and Matthew\\nD. Lerner, Sidley Austin LLP, Washington, D.C.; David R.\\nCarpenter, Sidley Austin LLP, Los Angeles, California; for\\nPetitioner-Appellee.\\n\\nChristopher J. Walker, The Ohio State University Moritz\\nCollege of Law, Columbus, Ohio; Steven P. Lehotsky, U.S.\\nChamber Litigation Center, Washington, D.C.; for Amicus\\nCuriae The Chamber of Commerce of the United States of\\nAmerica.\\n\\nElizabeth J. Stevens, Caplin & Drysdale, Chartered,\\nWashington, D.C., for Amici Curiae H. David Rosenbloom\\nand John P. Steines, Jr.\\n\\nAlice E. Loughran, Charles G. Cole, Michael C. Durst, and\\nRobert J. Kovacev, Steptoe & Johnson LLP, Washington\\nD.C., for Amici Curiae Semiconductor Industry Association,\\nInformation Technology Industry Council, National Foreign\\n\\f4 AMAZON.COM V. CIR\\n\\nTrade Council, Software Finance and Tax Executives\\nCouncil, and TechNet.\\n\\nA. Duane Webber, George M. Clarke, and Phillip J. Taylor,\\nBaker & McKenzie LLP, Washington, D.C.; Mark A. Oates,\\nSusan E. Ryba, and Cameron C. Reilly, Baker & McKenzie\\nLLP, Chicago, Illinois; Scott H. Frewing, Baker &\\nMcKenzie LLP, Palo Alto, California; for Amici Curiae\\nSilicon Valley Tax Directors Group, Agilent Technologies,\\nInc., Cisco Systems, Inc., Dell Technologies Inc., Dolby\\nLaboratories, Inc., Expedia Group, Inc., FireEye, Inc.,\\nGenesys Int’l Corp. Ltd., Informatica, Inc., NetApp, Inc.,\\nPalo Alto Networks, Inc., Surveymonkey Inc., and\\nVMware, Inc.\\n\\n\\n OPINION\\n\\nCALLAHAN, Circuit Judge:\\n\\n Appellee, Amazon.com, Inc., is a U.S.-based online\\nretailer with highly profitable intangible assets. In 2005 and\\n2006, Amazon restructured its European businesses in a way\\nthat would shift a substantial amount of its income from\\nU.S.-based entities to newly created European subsidiaries.\\nBecause the restructuring would allow the European entities\\nto generate income using Amazon’s pre-existing intangible\\nassets developed in the United States, the tax code and\\ncorresponding regulations required that the European\\nentities compensate Amazon for the use of assets that meet\\nthe regulatory definition of an “intangible.”\\n\\n The compensation was provided through a cost sharing\\narrangement, whereby Amazon and a holding company for\\nthe European subsidiaries would be treated as co-owners of\\n\\f AMAZON.COM V. CIR 5\\n\\nthe intangibles. Under the arrangement, the holding\\ncompany was required to make a “buy-in” payment for the\\npre-existing intangibles Amazon contributed to the\\narrangement and to make cost sharing payments going\\nforward for its share of future research and development\\n(R&D) efforts. The buy-in payment was taxable income to\\nAmazon, and the holding company’s cost sharing payments\\nwould reduce Amazon’s U.S. tax deductions for R&D costs.\\n\\n To guard against manipulation by jointly controlled\\nentities, the regulations require that the buy-in payment\\nreflect the fair market value of the pre-existing intangibles\\nmade available under a cost sharing arrangement. Amazon\\ninitially reported a buy-in payment of about $255 million.\\nAppellant, the Commissioner of Internal Revenue,\\nconcluded that the buy-in payment had not been determined\\nat arm’s length in accordance with the transfer pricing\\nregulations, so the IRS performed its own calculation,\\nvaluing the buy-in at about $3.6 billion. Amazon filed a\\npetition in the United States Tax Court challenging the IRS’s\\nvaluation.\\n\\n In the tax court proceedings, Amazon and the\\nCommissioner offered competing methods for valuing\\nAmazon’s pre-existing intangibles. There was a key\\ndifference between the parties’ respective approaches.\\nAmazon’s methodology isolated and valued only the\\nspecific intangible assets that it transferred to the European\\nholding company under the cost sharing arrangement,\\nincluding website technology, trademarks, and customer\\nlists. The Commissioner’s methodology essentially valued\\nthe entire European business, minus pre-existing tangible\\nassets. That method necessarily swept into the calculation\\nall contributions of value, including those that are more\\nnebulous and inseparable from the business itself, like the\\n\\f6 AMAZON.COM V. CIR\\n\\nvalue of employees’ experience, education, and training\\n(known as “workforce in place”), going concern value,\\ngoodwill, and other unique business attributes and\\nexpectancies (which the parties refer to as “growth\\noptions”). The tax court sided primarily with Amazon, and\\nthe Commissioner appealed.\\n\\n This case requires us to interpret the meaning of an\\n“intangible” in the applicable (but now outdated) transfer\\npricing regulations. 1 The case turns on whether, as the\\nCommissioner argues, the regulatory definition is broad\\nenough to include all intangible assets of value, even the\\nmore nebulous ones that the Commissioner refers to as\\n“residual-business assets” (i.e., Amazon’s culture of\\ninnovation, the value of workforce in place, going concern\\nvalue, goodwill, and growth options). We conclude that the\\ndefinition does not include residual-business assets.\\nAlthough the language of the definition is ambiguous, the\\ndrafting history of the regulations shows that “intangible”\\nwas understood to be limited to independently transferrable\\nassets. We thus affirm.\\n\\n\\n\\n\\n 1\\n This case is governed by regulations promulgated in 1994 and\\n1995. In 2009, more than three years after the tax years at issue here, the\\nDepartment of Treasury issued temporary regulations broadening the\\nscope of contributions for which compensation must be made as part of\\nthe buy-in payment. See 74 Fed. Reg. 340 (Jan. 5, 2009). In 2017,\\nCongress amended the definition of “intangible property” in 26 U.S.C.\\n§ 936(h)(3)(B) (which is incorporated by reference in 26 U.S.C. § 482).\\nTax Cuts and Jobs Act of 2017, Pub. L. 115-97, § 14221(a), 131 Stat.\\n2054, 2218 (2017). If this case were governed by the 2009 regulations\\nor by the 2017 statutory amendment, there is no doubt the\\nCommissioner’s position would be correct.\\n\\f AMAZON.COM V. CIR 7\\n\\n I.\\n\\n Before summarizing Amazon’s corporate restructuring\\nand the procedural history of this case, we first provide an\\noverview of the statutory and regulatory framework for\\ntransfer pricing.\\n\\n A.\\n\\n When a taxpayer sells or licenses its property, including\\nintangible assets, to another entity, the purchase price or\\nlicense royalty is taxable income. Often, such transactions\\noccur between entities “owned or controlled directly or\\nindirectly by the same interests.” 26 U.S.C. § 482. 2 The\\nparties in a controlled transaction are in a position to\\npotentially manipulate the terms of the transaction to\\nminimize taxable income artificially. But section 482 gives\\nthe Department of the Treasury the power to reallocate the\\ndollar figures of such controlled transactions if necessary “to\\nprevent evasion of taxes or clearly to reflect the income of”\\na taxpayer. Id.; see also Comm’r v. First Sec. Bank of Utah,\\nN.A., 405 U.S. 394, 400 (1972) (“[Section] 482 is designed\\nto prevent ‘artificial shifting, milking, or distorting of the\\ntrue net incomes of commonly controlled enterprises.’”\\n(quoting B. Bittker & J. Eustice, Federal Income Taxation of\\nCorporations and Shareholders p. 15–21 (3d ed. 1971)).\\n\\n The implementing regulations state that “[t]he purpose\\nof section 482 is to ensure that taxpayers clearly reflect\\nincome attributable to controlled transactions, and to prevent\\nthe avoidance of taxes with respect to such transactions.”\\n\\n 2\\n Unless otherwise indicated, references to section 482 of the\\nInternal Revenue Code and the implementing regulations are to the\\nversions in effect during the tax years at issue here (2005 and 2006).\\n\\f8 AMAZON.COM V. CIR\\n\\nTreas. Reg. § 1.482‑1(a)(1). A reallocation under\\nsection 482 and the implementing regulations is intended to\\n“place[] a controlled taxpayer on a tax parity with an\\nuncontrolled taxpayer by determining the true taxable\\nincome of the controlled taxpayer.” Id. The true taxable\\nincome is determined as if the parties to the controlled\\ntransaction had conducted their affairs in the manner of\\nunrelated parties “dealing at arm’s length.” Treas. Reg.\\n§ 1.482‑1(b)(1). The “arm’s length” standard is met “if the\\nresults of the transaction are consistent with the results that\\nwould have been realized if uncontrolled taxpayers had\\nengaged in the same transaction under the same\\ncircumstances.” Id.\\n\\n The key regulations were promulgated in 1994 and 1995.\\nThe regulations divide property into two categories: tangible\\nproperty (Treas. Reg. § 1.482‑3) and intangible property\\n(Treas. Reg. § 1.482‑4). This case concerns intangible\\nproperty. The regulations provide:\\n\\n (b) Definition of intangible. For purposes of\\n section 482, an intangible is an asset that\\n comprises any of the following items and has\\n substantial value independent of the services\\n of any individual—\\n\\n (1) Patents, inventions, formulae,\\n processes, designs, patterns, or know-\\n how;\\n\\n (2) Copyrights and literary, musical, or\\n artistic compositions;\\n\\n (3) Trademarks, trade names, or brand\\n names;\\n\\f AMAZON.COM V. CIR 9\\n\\n (4) Franchises, licenses, or contracts;\\n\\n (5) Methods, programs, systems,\\n procedures, campaigns, surveys, studies,\\n forecasts, estimates, customer lists, or\\n technical data; and\\n\\n (6) Other similar items. For purposes of\\n section 482, an item is considered similar\\n to those listed in paragraph (b)(1) through\\n (5) of this section if it derives its value not\\n from its physical attributes but from its\\n intellectual content or other intangible\\n properties.\\n\\nTreas. Reg. § 1.482‑4(b).\\n\\n A controlled taxpayer may make its intangibles available\\nto a foreign affiliate by entering a licensing agreement, under\\nwhich the foreign affiliate’s royalty fee is taxable income to\\nthe controlled taxpayer. See generally Treas. Reg. §§ 1.482-\\n1, 1.482-4. The royalty fee must, of course, reflect the\\n“arm’s length” value of the license. As new intangibles\\ncontinue to be developed, the value of the license changes\\nand becomes open to dispute for tax purposes on an ongoing\\nbasis.\\n\\n As an alternative to licensing, the regulations authorize\\njointly controlled entities to enter a cost sharing\\narrangement, under which they become co-owners of\\nintangibles developed as a result of the entities’ joint R&D\\n\\f10 AMAZON.COM V. CIR\\n\\nefforts. See generally Treas. Reg. § 1.482‑7A. 3 This\\narrangement provides the taxpayer the benefit of certainty\\nbecause, assuming the arrangement satisfies the\\nrequirements of the regulations, new intangibles need not be\\nvalued as they are developed.\\n\\n Under the regulations, a subsidiary that enters a cost\\nsharing arrangement with its parent must make two distinct\\npayments. First, the subsidiary must make an arm’s length\\n“buy-in” payment reflecting the value of the pre-existing\\nintangibles the parent contributes to the arrangement:\\n\\n If a controlled participant makes pre-existing\\n intangible property in which it owns an\\n interest available to other controlled\\n participants for purposes of research in the\\n intangible development area under a\\n qualified cost sharing arrangement, then each\\n such other controlled participant must make\\n a buy-in payment to the owner.\\n\\nTreas. Reg. § 1.482‑7A(g)(2). The buy-in payment is\\ntaxable income for the entity contributing the intangibles—\\nhere, Amazon.\\n\\n Second, the subsidiary must pay a share of the intangible\\ndevelopment costs (or R&D) “equal to its share of\\nreasonably anticipated benefits attributable to such\\ndevelopment.” Treas. Reg. § 1.482‑7A(a)(2). If the\\ncontrolled taxpayer incurs the lion’s share of the intangible\\ndevelopment costs, the subsidiary is required to make cost\\n\\n 3\\n We follow the parties’ convention of referring to the regulations\\non cost sharing arrangements by reference to the numbering as\\nredesignated in 2009 (i.e., Treas. Reg. § 1.482‑7A).\\n\\f AMAZON.COM V. CIR 11\\n\\nsharing payments to reflect its share of the anticipated\\nbenefits. The subsidiary’s cost sharing payments serve to\\nreduce the deductions the controlled taxpayer can take for\\nR&D costs (thereby increasing tax liability).\\n\\n B.\\n\\n Amazon is an online retailer that began operating in the\\nUnited States in 1995. In the late 1990s and early 2000s,\\nAmazon expanded operations into France, Germany, and the\\nUnited Kingdom. Amazon’s business in Europe had a siloed\\nstructure, with separate European subsidiaries independently\\noperating and managing the business, each subsidiary\\nhaving its own website, fulfillment centers, and customer\\nbase. The European subsidiaries licensed the right to use\\nAmazon’s website technology, customer information, and\\nmarketing intangibles (including trademarks and domain\\nnames).\\n\\n To address operational inefficiencies in Europe, in the\\nearly 2000s Amazon investigated options for creating a\\ncentralized European headquarters. Amazon ultimately\\nlocated its new headquarters in Luxembourg, which offered\\na central location, the lowest value-added tax rate in Europe,\\nand a relatively low corporate tax rate. Beginning in 2004,\\nAmazon undertook a series of transactions to implement its\\nplan for centralizing business operations in Europe.\\n\\n Amazon formed Amazon Europe Holding Technologies\\nSCS (“AEHT”) and transferred to AEHT the pre-existing\\nEuropean subsidiaries, their operating assets, and their pre-\\nexisting intangible rights (i.e., those developed in Europe).\\nAmazon and AEHT also entered a cost sharing\\n\\f12 AMAZON.COM V. CIR\\n\\narrangement—the transaction most pertinent to this appeal. 4\\nFor the arrangement to be a “qualified cost sharing\\narrangement” under the transfer pricing regulations outlined\\nabove, AEHT needed to pay Amazon for the pre-existing\\nintangibles Amazon contributed to the arrangement. To\\ndetermine the amount of this “buy-in,” Amazon hired a tax\\nfirm, which concluded the pre-existing intangibles were\\nworth $217 million. 5\\n\\n Under the cost sharing arrangement, AEHT also makes\\ncost sharing payments to Amazon for its share of ongoing\\nintangible development costs. Amazon reported cost\\nsharing payments from AEHT of about $116 million for\\n2005 and about $77 million for 2006.\\n\\n The IRS rejected Amazon’s calculation of AEHT’s buy-\\nin, concluding that Amazon grossly undervalued the\\nintangibles Amazon made available to AEHT. Applying a\\ndiscounted-cash-flow valuation methodology, the IRS\\ndetermined a buy-in payment of $3.6 billion.6 Amazon filed\\na petition in the tax court challenging the IRS’s valuation.\\nAmazon argued that the cost sharing arrangement covered\\n\\n 4\\n The following transactions were also part of the restructuring:\\n(1) Amazon granted AEHT a license to use Amazon’s existing\\ntechnology-related intellectual property; (2) Amazon assigned to AEHT\\ncustomer data and certain marketing intangibles (including trademarks,\\nwebsite content, and domain names) relating to the European business;\\n(3) Amazon transferred the stock of the European subsidiaries to AEHT;\\n(4) Amazon transferred other business assets (other than intellectual\\nproperty) to AEHT; and (5) the pre-existing European subsidiaries\\nlicensed to AEHT intellectual property titled in their names.\\n 5\\n The tax firm set the buy-in price at $254.5 million, to be paid over\\na seven-year period (resulting in a present value of $217 million).\\n\\n 6\\n The IRS’s calculation was later reduced to $3.468 billion.\\n\\f AMAZON.COM V. CIR 13\\n\\nthree distinct groups of transferred assets—website\\ntechnology, marketing intangibles, and European-customer\\ninformation—that must be valued separately using a\\nmethodology referred to in the transfer pricing regulations as\\nthe comparable uncontrolled transaction method.\\n\\n After a six-week trial that included testimony and written\\nreports of thirty expert witnesses, the tax court concluded\\nthat the Commissioner abused his discretion in determining\\nthat the discounted cash flow methodology supplied the best\\nmethod for determining an arm’s length buy-in payment and\\nin determining that the required payment is $3.468 billion.\\nAmazon.Com, Inc. v. Comm’r, 148 T.C. 108, 150 (2017).\\nRelying on the rationale of its prior decision in Veritas\\nSoftware Corp. v. Commissioner, 133 T.C. 297 (2009),\\nnonacq., 2010-49 I.R.B (2010), action on dec., 2010-05\\n(Nov. 12, 2010), the tax court reasoned that the\\nCommissioner’s valuation methodology “is based in essence\\non an ‘akin to a sale’ theory” and “necessarily sweeps into\\n[the] calculation assets that were not transferred under the\\n[cost sharing arrangement] and assets that were not\\ncompensable ‘intangibles’ to begin with.” Amazon.Com,\\n148 T.C. at 156–57. The tax court concluded that “[a]n\\nenterprise valuation of a business,” like the one conducted\\nby the Commissioner’s primary expert, “includes many\\nitems of value that are not ‘intangibles’” under the cost\\nsharing regulations. Id. at 157. Such items include\\n“workforce in place, going concern value, goodwill, and\\nwhat trial witnesses described as ‘growth options’ and\\ncorporate ‘resources’ or ‘opportunities.’” Id. The tax court\\nreasoned that such items are “[u]nlike the ‘intangibles’ listed\\n\\f14 AMAZON.COM V. CIR\\n\\nin the statutory and regulatory definitions” in that they\\n“cannot be bought and sold independently.” Id. 7\\n\\n The tax court adopted the comparable uncontrolled\\ntransaction method as the best way to value the buy-in\\npayment because that method allows for the intangibles at\\nissue to be isolated and separately valued. Id. at 164.\\nAlthough the tax court agreed with Amazon’s general\\nvaluation approach, it disagreed with certain aspects of\\nAmazon’s implementation of that method. Id. After making\\nadjustments, the tax court calculated the value of the buy-in\\npayment to be about $779 million. The Commissioner\\ntimely appealed, challenging the tax court’s rejection of his\\nexpert’s discounted cash flow methodology. 8\\n\\n II.\\n\\n The tax court had jurisdiction over this action under\\n26 U.S.C. § 6213(a). We have jurisdiction under 26 U.S.C.\\n§ 7482(a)(1), (b)(1).\\n\\n “[W]e review the tax court’s conclusions of law de novo\\nand its factual findings for clear error.” MK Hillside\\n\\n 7\\n The tax court made several other legal conclusions and factual\\nfindings not relevant to the Commissioner’s appeal.\\n 8\\n Although we affirm the tax court, nothing in our decision should\\nbe construed as an outright rejection of any particular valuation\\nmethodology. Here, the parties’ respective valuation methods did not\\ndiffer only in how they valued Amazon’s assets; they differed in what\\nassets they valued. The selection of the “best method” here, see Treas.\\nReg. § 1.482-1(c), thus turns on an interpretation of the regulatory\\ndefinition of an “intangible.” Nothing in our opinion should be\\nconstrued as favoring one valuation method over another in\\ncircumstances not present here.\\n\\f AMAZON.COM V. CIR 15\\n\\nPartners v. Comm’r, 826 F.3d 1200, 1203 (9th Cir. 2016)\\n(quoting DHL Corp. & Subsidiaries v. Comm’r, 285 F.3d\\n1210, 1216 (9th Cir. 2002)).\\n\\n III.\\n\\n The dispositive issue in this case is whether, under the\\n1994/1995 regulations, the “buy-in” required for “pre-\\nexisting intangible property” must include compensation for\\nresidual-business assets. 9 To answer this legal question, we\\nconsider the regulatory definition of an “intangible,” the\\noverall transfer pricing regulatory framework, the\\nrulemaking history of the regulations, and whether the\\nCommissioner’s position is entitled to deference under Auer\\nv. Robbins, 519 U.S. 452 (1997). We agree with the tax court\\nthat the definition of an “intangible” in § 1.482-4(b) was not\\nintended to embrace residual-business assets.\\n\\n A.\\n\\n The Commissioner argues that Amazon’s valuable\\nresidual-business assets meet “§ 1.482‑4(b)’s definition of\\nintangibles.” “Regulations are interpreted according to the\\nsame rules as statutes, applying traditional rules of\\nconstruction.” Minnick v. Comm’r, 796 F.3d 1156, 1159\\n(9th Cir. 2015); see Kisor v. Wilkie, 139 S. Ct. 2400, 2415\\n(2019) (recognizing that “all the ‘traditional tools’ of\\nconstruction” are the same for both statutes and regulations\\n\\n 9\\n The Commissioner also challenges some of the tax court’s other\\nreasons for rejecting the valuation conducted by the Commissioner’s\\nexpert. But in each of his other challenges, the Commissioner’s\\nargument assumes he is correct on his primary contention that the\\ndefinition of “intangible” in the 1994/1995 regulations embraces\\nresidual-business assets. Because we reject the Commissioner’s view on\\nthe primary issue, we need not address his other contentions.\\n\\f16 AMAZON.COM V. CIR\\n\\n(quoting Chevron, U.S.A., Inc. v. Nat. Res. Def. Council,\\nInc., 467 U.S. 837, 843 n.9 (1984))). Our “legal toolkit”\\nincludes careful examination of “the text, structure, history,\\nand purpose of a regulation.” Kisor, 139 S. Ct. at 2415.\\n\\n We begin with the language of § 1.482-4(b). If the\\nregulation is unambiguous, its plain meaning governs. Safe\\nAir For Everyone v. EPA, 488 F.3d 1088, 1097 (9th Cir.\\n2007).\\n\\n The regulation defines an “intangible” as an asset that\\nboth “has substantial value independent of the services of\\nany individual” and is one of the items listed in subsection\\n(b)(1)–(6). Treas. Reg. § 1.482‑4(b). Each of the 28 specific\\nitems in subsection (b) is independently transferrable—none\\nis a residual-business asset. The Commissioner thus relies\\non the catchall provision for “[o]ther similar items.” Treas.\\nReg. § 1.482‑4(b)(6). “[A]n item is considered similar” to\\nthe other items in the subsection “if it derives its value not\\nfrom its physical attributes but from its intellectual content\\nor other intangible properties.” Id.\\n\\n Reading the catchall provision together with the\\nintroductory language of the definition, residual-business\\nassets are intangibles if they (1) have substantial value\\nindependent of the services of any individual and (2) derive\\ntheir value from intellectual content or other intangible\\nproperties. The Commissioner argues both elements are\\nsatisfied. First, he argues that Amazon’s growth options\\n“derive their value from intangible, rather than physical,\\nattributes.” He then cites testimony from Amazon’s expert\\n(Dr. Bradford Cornell) that Amazon’s growth options are\\nprimarily attributable to its culture of innovation. Second,\\nhe argues that the value of Amazon’s growth options is\\nindependent of the services of any individual because they\\n\\f AMAZON.COM V. CIR 17\\n\\nare “part of the culture of Amazon to be able to have creative\\nideas bubble up in their organization and actually use them.”\\n\\n Amazon argues that the Commissioner’s interpretation\\nof the catchall provision is too sweeping for several reasons.\\nAmazon’s central argument is that to qualify as an\\n“intangible” under the regulation, an item must be capable\\nof being bought and sold independently of the business—\\nand residual-business assets are inseparable from the\\nbusiness. The tax court agreed. See Amazon.Com, 148 T.C.\\nat 157 (concluding that, unlike the specific intangibles listed\\nin the regulation, “workforce in place, going concern value,\\ngoodwill, and what trial witnesses described as ‘growth\\noptions’ and corporate ‘resources’ or ‘opportunities’ . . .\\ncannot be bought and sold independently”).\\n\\n Amazon offers several arguments in support of its\\nposition that assets that are inseparable from the business do\\nnot meet the regulatory definition of an “intangible.”\\nQuoting Arcadia v. Ohio Power Co., 498 U.S. 73, 78 (1990),\\nAmazon argues that the Commissioner’s interpretation\\nrenders the “enumeration of specific subjects entirely\\nsuperfluous—in effect adding to that detailed list ‘or\\nanything else.’” Amazon invokes the canon ejusdem\\ngeneris—of the same kind or class—and asserts that “[t]he\\ncommon attribute of all 28 specified items is that they can be\\nsold independently.” Amazon argues this canon should be\\napplied to prevent the catchall clause from swallowing the\\npreceding language of the regulation.\\n\\n Amazon’s focus on the commonality of the 28 specified\\nitems has some force. After all, if all 28 listed items share a\\ncommon attribute, why would anyone understand a catchall\\nfor “similar items” to include a non-listed item that doesn’t\\nshare that attribute? Amazon’s commonality argument\\nfalters, however, because the catchall provision did not\\n\\f18 AMAZON.COM V. CIR\\n\\nsimply say “[o]ther similar items.” Instead, the catchall\\nelaborated by explaining how an item is determined to be\\n“similar” to the other items: “if it derives its value not from\\nits physical attributes but from its intellectual content or\\nother intangible properties.” Treas. Reg. § 1.482-4(b)(6).\\nThis explanation of what is “similar” leaves open the\\npossibility of a non-listed item being included in the\\ndefinition even if it doesn’t share the attribute of being\\nseparately transferrable.\\n\\n Amazon also argues that the regulation’s requirement\\nthat an intangible have “substantial value independent of the\\nservices of any individual,” Treas. Reg. § 1.482‑4(b),\\nsupports its position. The tax court in Veritas agreed. In that\\ncase, the tax court rejected the Commissioner’s argument\\nthat a foreign subsidiary’s buy-in under a cost sharing\\narrangement must include the value of the subsidiary’s\\naccess to the U.S. corporate parent’s R&D and marketing\\nteams—i.e., workforce in place. Veritas, 133 T.C. at 323.\\nAlthough the court found insufficient evidence that the cost\\nsharing arrangement made such a transfer, it also stated that\\n“[e]ven if such evidence existed, these items would not be\\ntaken into account in calculating the requisite buy-in\\npayment because they do not have ‘substantial value\\nindependent of the services of any individual’ and thus do\\nnot meet the” definition of “intangible.” Id. at 323 n.31\\n(quoting Treas. Reg. § 1.482‑4(b)). The court reasoned that\\nthe value of access to the parent’s “R&D and marketing\\nteams is based primarily on the services of individuals (i.e.,\\nthe work, knowledge, and skills of team members).” Id. The\\nCommissioner clearly disagrees with Veritas on this point,\\nbut he doesn’t explain why the tax court’s analysis is wrong.\\n\\n Analysis of the regulatory text alone does not\\ndefinitively resolve the question here. The definition of an\\n\\f AMAZON.COM V. CIR 19\\n\\n“intangible” is susceptible to, but does not compel, an\\ninterpretation that embraces residual-business assets. The\\nproblem is that residual-business assets, such as “growth\\noptions” and a “culture of innovation,” are amorphous, and\\nit’s not self-evident whether such assets have “substantial\\nvalue independent of the services of any individual.” See\\nTreas. Reg. § 1.482‑4(b). Amazon raises legitimate\\nconcerns about the regulation’s catchall being stretched too\\nfar, and those concerns likely bear on which party has the\\nmore reasonable view of the regulatory definition. It\\nnonetheless remains that the definition of “intangible” could\\nbe construed as covering residual-business assets if the\\nlanguage of § 1.482‑4(b) is viewed in isolation. 10\\n\\n B.\\n\\n But we are required to look at the regulatory scheme “as\\na whole,” viewing the regulatory “definition in the context\\nof the entire [transfer pricing] regulations.” Alaska Trojan\\nP’ship v. Gutierrez, 425 F.3d 620, 628 (9th Cir. 2005). The\\nCommissioner argues that other regulations governing\\ntransfer pricing support his view that the regulatory\\ndefinition of an “intangible” embraces residual-business\\n\\n 10\\n The Commissioner also argues that the definition of an\\n“intangible” includes residual-business assets because an uncontrolled\\nparty would pay for access to those assets in an arm’s length transaction.\\nHe cites the testimony of Amazon’s expert that parties dealing at arm’s\\nlength “[d]efinitely” pay for “growth options” because “[n]o company is\\ngoing to give away something of value without compensation.” The\\nCommissioner’s argument misses the mark. Under the regulations, the\\narm’s length standard governs the valuation of intangibles; it doesn’t\\nanswer whether an item is an intangible. The definition of an\\n“intangible” is provided in § 1.482‑4(b). The Commissioner points to no\\nlanguage in the statute or regulations suggesting that the definition of\\nwhat constitutes an intangible is determined by asking whether an\\nuncontrolled party would pay for it.\\n\\f20 AMAZON.COM V. CIR\\n\\nassets. He cites two sections in the regulations, §§ 1.482‑7A\\nand 1.482‑1.\\n\\n Section 1.482‑7A specifies the requirements of a\\nqualified cost sharing arrangement and the methods for\\ndetermining the taxable income resulting from such an\\narrangement. The Commissioner cites the subsection that\\nimposes the “buy-in” requirement where “[a] controlled\\nparticipant . . . makes intangible property available to a\\nqualified cost sharing arrangement.” Treas. Reg.\\n§ 1.482‑7A(g)(1). The Commissioner argues that this\\nprovision mandates that residual-business assets be paid for\\n“if they are made ‘available’ to the cost-sharing participants”\\neven though such assets generally cannot be transferred\\nindependently from the business.\\n\\n The Commissioner’s argument based on\\n§ 1.482‑7A(g)(1) presupposes the very point he attempts to\\nprove—that residual-business assets are “intangible\\nproperty” within the meaning of the regulations. The\\nprovision requiring a “buy-in” does not expand but instead\\nincorporates the meaning of an “intangible” given in\\n§ 1.482-4(b). See Treas. Reg. § 1.482‑7A(g)(2) (“The buy-\\nin payment by each such other controlled participant is the\\narm’s length charge for the use of the intangible . . . .”\\n(emphasis added)). The “makes . . . available” language thus\\nprovides no meaningful insight into the regulatory definition\\nof an “intangible.” 11\\n\\n\\n 11\\n Considered in context of the regulatory framework, it appears that\\nthe purpose of the “makes . . . available” phrasing in § 1.482‑7A(g) is to\\naccount for taxpayers who might seek to avoid the requirements of the\\ntransfer pricing regulations by not formally transferring their intangible\\nassets to a related party and instead informally making available the\\n\\f AMAZON.COM V. CIR 21\\n\\n The Commissioner also finds support for his position in\\nthe regulations’ preamble stating that “[t]he purpose of\\nsection 482 is to ensure that taxpayers clearly reflect income\\nattributable to controlled transactions and to prevent the\\navoidance of taxes with respect to such transactions.” Treas.\\nReg. § 1.482‑1(a)(1). Relying on Xilinx, Inc. v.\\nCommissioner, 598 F.3d 1191 (9th Cir. 2010), the\\nCommissioner argues that “anything of value that is made\\navailable between related parties must be paid for” in the\\nbuy-in, regardless of whether it is defined as an intangible.\\n\\n In Xilinx, we considered whether the parties to a cost\\nsharing arrangement “must include the value of certain stock\\noption compensation one participant gives to its employees\\nin the pool of costs to be shared.” Id. at 1192. We found\\ntwo provisions of the regulations in conflict. We first cited\\n§ 1.482‑1(b)(1)’s statement that “the standard to be applied\\nin every case is that of a taxpayer dealing at arm’s length\\nwith an uncontrolled taxpayer.” Id. at 1195 (quoting Treas.\\nReg. § 1.482‑1(b)(1)). It was undisputed on appeal that\\nunrelated parties would not have shared the employee stock\\noption costs. Id. at 1194. Another provision, however,\\nrequired that “controlled parties in a cost sharing agreement\\n. . . share all ‘costs . . . related to the intangible development\\narea,’” which by definition would include the employee\\nstock options. Id. at 1196. In resolving this conflict, the\\nXilinx majority affirmed the tax court’s conclusion that the\\nstock options need not be paid for as part of the buy-in. Id.\\n\\n\\n\\n\\nassets. The regulations thus treat such taxpayers “as having transferred\\ninterests in such [intangible] property.” Treas. Reg. § 1.482‑7A(g)(1).\\n\\f22 AMAZON.COM V. CIR\\n\\nat 1196–97. The primary opinion reasoned that “[p]urpose\\nis paramount.” Id. at 1196. 12\\n\\n Relying on the logic of the primary opinion in Xilinx, the\\nCommissioner asserts that “it is undisputed that a company\\nentering into the same transaction under the same\\ncircumstances with an unrelated party would have required\\ncompensation.” To evaluate this claim, it’s important to be\\nclear about what exactly is meant by the phrase “the same\\ntransaction under the same circumstances.” The\\nCommissioner relies on deposition testimony from one of\\nAmazon’s experts that parties dealing at arm’s length would\\npay for growth options. But the expert explained at trial the\\ndifference between an investor or purchaser of the entire\\nbusiness (who would pay for the full value of the business)\\nand a partner (who would not). The question becomes\\nwhether a cost sharing arrangement is akin to the sale of a\\nbusiness or like a partnership in certain assets or aspects of\\nthe business. The Commissioner assumes, but does not\\nexplain why, the transfer of intangible assets under\\n\\n\\n\\n 12\\n Caution should be taken before relying on the rationale of the\\nprimary opinion in Xilinx. One panel member concurred—thus\\nproviding a majority vote for the disposition—but wrote separately “to\\nexplain [his] particular reasons for rejecting the Commissioner’s\\nposition.” Id. at 1197 (Fisher, J., concurring). Although the reasoning\\nof the concurring opinion overlaps somewhat with the reasoning of the\\nprimary opinion, the concurrence does not seem to adopt the “[p]urpose\\nis paramount” logic. Instead, based on the language of the regulations,\\nthe legislative and regulatory history, international tax treaties, and the\\nunderstanding of the business community and tax professionals (i.e.,\\namici curiae), the concurrence concluded that the taxpayer’s\\n“understanding of the regulations is the more reasonable even if the\\nCommissioner’s current interpretation may be theoretically plausible.”\\nId. at 1198.\\n\\f AMAZON.COM V. CIR 23\\n\\nAmazon’s cost sharing arrangement with AEHT should be\\ntreated the same as the sale of the business.\\n\\n The Commissioner’s reliance on Xilinx thus suffers the\\nsame defect as his “made available” argument based on\\n§ 1.482-7A(g)—he assumes the very conclusion he’s aiming\\nto prove. Although the regulatory provisions the\\nCommissioner cites are consistent with his position, they do\\nnot provide independent support and they are likewise\\nconsistent with Amazon’s view.\\n\\n If the cost sharing regulations as a whole tip the scale\\neither direction, they tend to favor Amazon on the issue\\npresented here. The regulations describe a cost sharing\\narrangement as an agreement “to share the costs of\\ndevelopment of one or more intangibles.” Treas. Reg.\\n§ 1.482‑7A(a)(1). The written document memorializing the\\narrangement must, among other things, describe both “the\\nscope of the research and development to be undertaken,\\nincluding the intangible or class of intangibles intended to be\\ndeveloped.” Treas. Reg. § 1.482‑7A(b)(4)(iii). The writing\\nmust also describe “each participant’s interest in any covered\\nintangibles,” which is “any intangible property that is\\ndeveloped as a result of the research and development\\nundertaken under the cost sharing arrangement.” Treas.\\nReg. § 1.482‑7A(b)(4)(iv). By identifying intangibles as\\nbeing the product of R&D efforts, the regulations seem to\\ncontemplate a meaning of “intangible” that excludes items\\nlike goodwill and going concern value, which “are generated\\nby earning income, not by incurring deductions.” Staff of J.\\nComm. on Taxation, 98th Cong., General Explanation of the\\nRevenue Provisions of the Deficit Reduction Act of 1984,\\n428 (Comm. Print 1984).\\n\\f24 AMAZON.COM V. CIR\\n\\n The overall regulatory scheme doesn’t definitively\\nresolve the issue, but it favors Amazon more than the\\nCommissioner.\\n\\n C.\\n\\n We next turn to the drafting history of the regulatory\\ndefinition. Both parties claim support for their respective\\npositions in the historical development of the regulations.\\n\\n Treasury first defined intangible property for purposes of\\nsection 482 in regulations adopted in 1968:\\n\\n [I]ntangible property shall consist of the\\n items described in subdivision (ii) of this\\n subparagraph, provided that such items have\\n substantial value independent of the services\\n of individual persons.\\n\\n (ii) The items referred to in subdivision (i) of\\n this subparagraph are as follows:\\n\\n (a) Patents, inventions, formulas,\\n processes, designs, patterns, and other\\n similar items;\\n\\n (b) Copyrights, literary, musical, or\\n artistic compositions, and other similar\\n items;\\n\\n (c) Trademarks, trade names, brand\\n names, and other similar items;\\n\\n (d) Franchises, licenses, contracts, and\\n other similar items;\\n\\f AMAZON.COM V. CIR 25\\n\\n (e) Methods, programs, systems,\\n procedures, campaigns, surveys, studies,\\n forecasts, estimates, customer lists,\\n technical data, and other similar items.\\n\\n33 Fed. Reg. 5848, 5854 (Apr. 16, 1968).\\n\\n Fourteen years later, Congress enacted the Tax Equity\\nand Fiscal Responsibility Act of 1982, adopting a definition\\nof intangible property in section 936 of the Internal Revenue\\nCode that was nearly identical to the one in the regulations\\nimplementing section 482. 13 The Commissioner plucks a\\nphrase from a Senate Report for the 1982 Act suggesting that\\nthe statute “defines intangible assets broadly.” With more\\ncontext, the quoted sentence from the Senate Report states\\nthat “[t]he bill defines intangible assets broadly to include”\\nthe 28 specifically-listed items “and other items similar to\\nany of those listed, so long as the item has substantial value\\nindependent of the services of individual persons.” S. Rep.\\n97-494, at 161 (1982), as reprinted in 1982 U.S.C.C.A.N.\\n781, 924 (emphasis added). 14\\n\\n\\n 13\\n The new statutory definition differed from the regulatory one in\\nfour, relatively minor respects: (1) making all enumerated items singular;\\n(2) adding “know-how” to the first category; (3) inserting “any similar\\nitem” as a stand-alone category in lieu of “other similar items” in each\\ncategory; and (4) requiring that an item have “substantial value\\nindependent of the services of any individual” rather than “of individual\\npersons.” Tax Equity & Fiscal Responsibility Act of 1982, Pub. L. 97-\\n248, § 213, 96 Stat. 324 (Sept. 3, 1982).\\n\\n 14\\n Another part of the quoted Senate Report states that the\\nCommittee viewed the bill as combatting the practice of transferring\\nintangibles “created, developed or acquired in the United States” to\\nforeign entities to generate income tax-free. S. Rep. 97-494, at 158–59\\n(1982), as reprinted in 1982 U.S.C.C.A.N. 781, 921–22. It seems from\\n\\f26 AMAZON.COM V. CIR\\n\\n The Tax Reform Act of 1986 amended section 482 to\\nincorporate the “intangible property” definition from section\\n936: “In the case of any transfer (or license) of intangible\\nproperty (within the meaning of section 936(h)(3)(B)), the\\nincome with respect to such transfer or license shall be\\ncommensurate with the income attributable to the\\nintangible.” Pub. L. 99–514, § 1231, 100 Stat. 2085 (Oct.\\n22, 1986).\\n\\n When it amended section 482, Congress requested that\\nthe IRS conduct “a comprehensive study of intercompany\\npricing rules” and report on whether the existing regulations\\n“could be modified in any respect.” H.R. Rep. No. 99-841,\\nat II-637 (1986). The resulting IRS report, known as the\\n“White Paper,” laid the groundwork for what would\\nultimately become the 1994/1995 regulations. A Study of\\nIntercompany Pricing Under Section 482 of the Code, I.R.S.\\nNotice 88-123, 1988-2 C.B. 458.\\n\\n The White Paper discussed, among other things, cost\\nsharing arrangements. According to the White Paper, the\\npurpose of the buy-in requirement is for “a party to a cost\\nsharing arrangement that has contributed funds or incurred\\nrisks for development of intangibles at an earlier stage” to be\\n“compensated by the other participants.” Id. at 497. “[I]f\\nthere are intangibles that are not fully developed that relate\\nto the research to be conducted under the cost sharing\\narrangement, it is necessary to value them in order to\\ndetermine an appropriate buy-in payment.” Id.\\n\\n\\n\\n\\nthis language, the Committee didn’t contemplate intangibles that are not\\nindependently transferrable.\\n\\f AMAZON.COM V. CIR 27\\n\\n The White Paper proposed “three basic types of\\nintangibles” that would be subject to the buy-in requirement:\\n\\n • “preexisting intangibles at various stages of\\n development that will become subject to the\\n arrangement”;\\n\\n • “basic research not associated with any product”; and\\n\\n • “a going concern value associated with a\\n participant’s research facilities and capabilities that\\n will be utilized.”\\n\\nId. Although the White Paper proposed including going\\nconcern value of a research facility in the buy-in, after\\nreceiving opposition in public comments, Treasury proposed\\nnew regulations that essentially retained the definition of\\n“intangible” from before without referencing going concern\\nvalue or any other residual-business asset. See 57 Fed. Reg.\\n3571, 3579 (Jan. 30, 1992).\\n\\n In 1993, Treasury issued revised temporary and\\nproposed regulations that defined an “intangible” as “any\\ncommercially transferable interest” in the intangibles listed\\nin § 936(h)(3)(B) that had “substantial value independent of\\nthe services of any individual.” 58 Fed. Reg. 5263, 5287\\n(Jan. 21, 1993). Treasury also requested comment on\\n“whether the definition of intangible property . . . should be\\nexpanded to include items not normally considered to be\\nitems of intellectual property, such as work force in place,\\ngoodwill or going concern value.” 58 Fed. Reg. 5310, 5312\\n(Jan. 21, 1993). Amazon cites opposition comments\\nsubmitted in response to Treasury’s request for comment.\\n\\n In 1994, Treasury issued final regulations (the ones\\napplicable here), which reflected a minor reworking of the\\n\\f28 AMAZON.COM V. CIR\\n\\ndefinition. 59 Fed. Reg. 34971, 35016 (Jul. 8, 1994).\\nTreasury explained that the revised definition omitted the\\n“commercially transferrable” language that appeared in the\\ntemporary regulations “because it was superfluous: if the\\nproperty was not commercially transferrable, then it could\\nnot have been transferred in a controlled transaction.” Id.\\nat 34983. Treasury also explained that the revision\\n“clarified” that the phrase “other similar items” in the\\ndefinition “refer[s] to items that derive their value from\\nintellectual content or other intangible properties rather than\\nphysical attributes.” Id. In 1995, Treasury issued final\\nregulations governing cost sharing arrangements, including\\n§ 1.482‑7A(g)’s “buy-in” requirement.\\n\\n The drafting history of the transfer pricing regulations\\ndoes not support the Commissioner’s argument that the\\ndefinition of an “intangible” covered residual-business\\nassets. The only references in the drafting history to any\\nresidual-business assets suggest that such items were\\nexcluded from the definition of intangible assets. The IRS’s\\n1988 White Paper proposed including “going concern value”\\nof a research facility in the buy-in, but Treasury’s 1994/1995\\nregulations kept essentially the same definition as before\\nwithout referring to “going concern value” or any other\\nresidual-business asset. See 57 Fed. Reg. at 3579.\\n\\n Two key statements by Treasury in the drafting history\\nrender the Commissioner’s current position untenable. First,\\nin 1993, Treasury confirmed that the then-existing definition\\nof “intangible” did not include residual-business assets when\\nit asked for comments on whether the definition of\\nintangibles “should be expanded to include items not\\nnormally considered to be items of intellectual property,\\nsuch as work force in place, goodwill or going concern\\nvalue.” 58 Fed. Reg. at 5312 (emphasis added). Second, a\\n\\f AMAZON.COM V. CIR 29\\n\\nyear later after opting against such an expansion, and instead\\nretaining the same essential definition from before\\n(including the same list of 28 items), Treasury explained that\\nthe final (1994) rule merely “clarified” when an item would\\nbe deemed similar to the 28 items listed in the definition.\\n59 Fed. Reg. at 34983.\\n\\n The Commissioner is thus forced to argue that what\\nTreasury explicitly confirmed would not be considered an\\n“intangible” without a substantive “expan[sion]” of the\\ndefinition was implicitly added to the definition through a\\nnon-specific “clarifi[cation].” The Commissioner’s\\nargument stretches “clarification” beyond its commonly\\nunderstood meaning of merely clearing up what was\\npreviously ambiguous or otherwise restating a standard\\nconsistent with what was previously intended. Cf. Motorola,\\nInc. v. Fed. Exp. Corp., 308 F.3d 995, 1007 (9th Cir. 2002)\\n(distinguishing between a “clarifying amendment” and “one\\nthat work[s] a substantive change” (emphasis omitted)). 15\\n\\n Another statement from the drafting history of § 1.482‑4\\nlends further support for Amazon’s position that\\n“intangible” has always been understood to be limited to\\nassets that are independently transferrable. Treasury’s\\ntemporary regulations in 1993 defined an “intangible” as\\n“any commercially transferable interest” in the intangibles\\nlisted in section 936(h)(3)(B) that had “substantial value\\nindependent of the services of any individual.” 58 Fed. Reg.\\nat 5287. When Treasury left out the “commercially\\n 15\\n Amazon and amici curiae also argue that if the Commissioner is\\ncorrect that the non-specific “clarifi[cation]” of § 1.482-4(b)’s catchall\\nsubstantively expanded the definition of an “intangible,” then\\nTreasury/IRS violated the Administrative Procedures Act. We need not\\naddress this argument because we reject the Commissioner’s post hoc\\ninterpretation of the changes to the regulatory definition.\\n\\f30 AMAZON.COM V. CIR\\n\\ntransferrable” language from the final regulations issued a\\nyear later, it explained that the omitted language was\\n“superfluous” because “if the property was not\\ncommercially transferrable, then it could not have been\\ntransferred in a controlled transaction.” 59 Fed. Reg. at\\n34983. This is consistent with the basic premise of the\\ntransfer pricing regulations, which contemplate a situation in\\nwhich particular assets are transferred from one entity to\\nanother.\\n\\n The Commissioner now contends that the 1994\\ndefinition was intended to embrace residual-business assets\\neven though such assets “cannot be transferred\\nindependently.” Yet the Commissioner fails to identify any\\ncontemporaneous statement by the agency that would\\n“display awareness” that it was changing its position on\\nwhether residual-business assets are included within the\\ndefinition of intangibles. See FCC v. Fox Television\\nStations, Inc., 556 U.S. 502, 515 (2009) (“[T]he requirement\\nthat an agency provide reasoned explanation for its action\\nwould ordinarily demand that it display awareness that\\nit is changing position. An agency may not, for example,\\ndepart from a prior policy sub silentio or simply disregard\\nrules that are still on the books.”).\\n\\n The Commissioner argues his position is supported by a\\ndifferent regulatory scheme governing penalties for\\ntaxpayers who substantially misstate the value of property\\non their tax returns. See 26 U.S.C. § 6662(e). In 1993,\\nTreasury proposed regulations “designed to encourage\\ntaxpayers to document their transfer pricing transactions and\\nto provide that documentation to the [IRS] upon request.”\\n58 Fed. Reg. 5304, 5304 (Jan. 21, 1993). To that end, the\\nproposed regulations provided rules for determining whether\\na taxpayer has substantially misstated the value of property.\\n\\f AMAZON.COM V. CIR 31\\n\\nThose rules defined “property” to include “intangible\\nproperty,” which in turn was defined to include “property\\nsuch as goodwill, covenants not to compete, leaseholds,\\npatents, contract rights, debts, and choices in action.” Id.\\nat 5306.\\n\\n The Commissioner’s attempt to bootstrap § 1.6662-5’s\\nreference to “goodwill” ignores that the proposed regulation\\nconcerning substantial valuation misstatements was issued\\non the same day Treasury separately confirmed that the\\ndefinition of “intangible” for purposes of the regulations\\nimplementing section 482 did not include goodwill or other\\nresidual-business assets. 58 Fed. Reg. at 5312. Considering\\nthe context, Treasury’s use of “goodwill” in § 1.6662-5 but\\nnot in § 1.482‑4(b) most likely evinces an intent not to\\ninclude goodwill or other residual-business assets within\\n§ 1.482‑4(b)’s definition of “intangible.”\\n\\n Amazon points to other Treasury regulations that define\\ncertain covered property by incorporating the definition of\\nintangible property under section 936(h)(3)(B) and then\\nadding goodwill and going concern value. Treas. Reg.\\n§§ 1.954-2(e)(3)(iv), 1.861-9T(h)(1)(ii). These provisions\\nadd little to the discussion, but the express inclusion of\\ngoodwill and going concern value in these regulations is\\nconsistent with Amazon’s position that the definition of\\n“intangible” in section 936(h)(3)(B) and § 1.482‑4(b) was\\nunderstood to exclude goodwill and going concern value.\\nThese regulations also show that Treasury “clearly knew\\nhow to write its regulations” to include goodwill and other\\nresidual-business assets. Karczewski v. DCH Mission Valley\\nLLC, 862 F.3d 1006, 1015–16 (9th Cir. 2017).\\n\\f32 AMAZON.COM V. CIR\\n\\n The drafting history of § 1.482‑4(b) strongly supports\\nAmazon’s position that Treasury limited the definition of\\n“intangible” to independently transferrable assets. 16\\n\\n D.\\n\\n The Commissioner next argues that the tax court should\\nhave deferred to the IRS’s interpretation of its own\\nregulations. Under certain circumstances, an agency’s\\ninterpretation of its own regulations “must be given\\n‘controlling weight’” if it is not “‘plainly erroneous or\\ninconsistent with the regulation.’” Stinson v. United States,\\n508 U.S. 36, 45 (1993) (quoting Bowles v. Seminole Rock &\\nSand Co., 325 U.S. 410, 414 (1945)); see also Auer v.\\nRobbins, 519 U.S. 452, 461–62 (1997). This is frequently\\nreferred to as Auer deference.\\n\\n The Supreme Court “has cabined Auer’s scope in varied\\nand critical ways.” Kisor, 139 S. Ct. at 2418. “First and\\nforemost, a court should not afford Auer deference unless the\\n 16\\n In 2017, Congress amended the definition of “intangible property”\\nin section 936(h)(3)(B). That amendment added “goodwill, going\\nconcern value, or workforce in place” to the list of specific items\\nincluded in the definition of “intangible property.” Tax Cuts & Jobs Act\\nof 2017, Pub. L. 115-97, § 14221(a), 131 Stat. 2054, 2218 (2017).\\nCharacterizing the change as merely a “clarification,” the Commissioner\\nargues that the 2017 amendment supports his proffered interpretation of\\nthe 1994/1995 regulations. But the Commissioner’s post-hoc label of\\nCongress’s amendment is not controlling, see Beaver v. Tarsadia Hotels,\\n816 F.3d 1170, 1186 (9th Cir. 2016) (“Post-hoc labeling as a\\n‘clarification’ by bill supporters of what otherwise appears to be a change\\n. . . is not controlling . . . .”), and it’s also not supported by Congress’s\\nown words. Congress stated that the amendment should not be\\n“construed to create any inference” as to the definition of intangibles for\\ntaxable years occurring before the amendment’s effective date. 131 Stat.\\nat 2219. Congress said nothing to indicate that the amendment was\\nmeant only as a clarification.\\n\\f AMAZON.COM V. CIR 33\\n\\nregulation is genuinely ambiguous.” Id. at 2415. Genuine\\nambiguity is not determined by examination of the\\nregulatory text alone. Instead, “before concluding that a rule\\nis genuinely ambiguous, a court must exhaust all the\\n‘traditional tools’ of construction,” “‘carefully\\nconsider[ing]’ the text, structure, history, and purpose of a\\nregulation, in all the ways it would if it had no agency to fall\\nback on.” Id. (first quoting Chevron, 467 U.S. at 843 n.9;\\nthen quoting Pauley v. BethEnergy Mines, Inc., 501 U.S.\\n680, 707 (1991) (Scalia, J., dissenting))).\\n\\n The text of the regulatory definition of “intangible,” the\\ndefinition’s place within the transfer pricing regulations\\ngenerally, and the rulemaking history leave little room for\\nthe Commissioner’s proffered meaning. But even if there\\nwere genuine ambiguity, there is a separate reason Auer\\ndeference is not warranted here.\\n\\n “[N]ot every reasonable agency reading of a genuinely\\nambiguous rule should receive Auer deference.” Kisor,\\n139 S. Ct. at 2416. Instead, “a court must make an\\nindependent inquiry into whether the character and context\\nof the agency interpretation entitles it to controlling weight.”\\nId. (citing Christopher v. SmithKline Beecham Corp.,\\n567 U.S. 142, 155 (2012)). For example, courts will not\\ndefer to an agency’s interpretation when doing so “would\\nseriously undermine the principle that agencies should\\nprovide regulated parties fair warning of the conduct [a\\nregulation] prohibits or requires.” Barboza v. Cal. Ass’n of\\nProf’l Firefighters, 799 F.3d 1257, 1267 (9th Cir. 2015)\\n(alteration in original) (quoting Christopher, 567 U.S.\\nat 156). This exception accounts for the “risk that agencies\\nwill promulgate vague and open-ended regulations that they\\ncan later interpret as they see fit, thereby ‘frustrat[ing] the\\nnotice and predictability purposes of rulemaking.’”\\n\\f34 AMAZON.COM V. CIR\\n\\nChristopher, 567 U.S. at 158 (quoting Talk Am., Inc. v.\\nMichigan Bell Tel. Co., 564 U.S. 50, 68 (2011) (alteration in\\noriginal) (Scalia, J., concurring)).\\n\\n Christopher and Barboza thus teach that the timing of an\\nagency’s first announcement of its interpretation may be\\ndispositive on whether the agency’s view will be given Auer\\ndeference. Here, the Commissioner does not identify a\\nspecific document (e.g., policy manual or court brief)\\ndefinitively expressing the agency’s view of its regulations.\\nIt thus appears that the Commissioner’s court briefs in this\\ncase present Treasury’s “first announce[ment of] its view,”\\nsee Christopher, 567 U.S. at 153, that the definition of\\nintangible in § 1.482-4(b) embraces residual-business assets.\\nThe exception to Auer deference from Christopher and\\nBarboza therefore applies. “Where an agency announces its\\ninterpretation for the first time in an enforcement\\nproceeding, and has not previously taken any action to\\nenforce that interpretation, ‘the potential for unfair surprise\\nis acute.’” Barboza, 799 F.3d at 1267 (quoting Christopher,\\n567 U.S. at 158). Even if first arising before the current\\nlitigation, a new interpretation is owed no deference if it\\nwould “create[] ‘unfair surprise’ to regulated parties.”\\nKisor, 139 S. Ct. at 2417–18 (quoting Long Island Care at\\nHome, Ltd. v. Coke, 551 U.S. 158, 170 (2007)). No\\nstatement from Treasury in the drafting history of the\\n1994/1995 regulations expresses the position the\\nCommissioner advances now. Indeed, as discussed above,\\nTreasury’s contemporaneous explanations of the regulations\\nare to the contrary. Amazon and other taxpayers were thus\\nnot given fair warning of the Commissioner’s current\\ninterpretation of the regulatory definition of an “intangible.”\\nThat interpretation is not entitled to deference.\\n\\f AMAZON.COM V. CIR 35\\n\\n IV.\\n\\n The Commissioner’s calculation of AEHT’s buy-in\\nunder the cost sharing arrangement included residual-\\nbusiness assets as part of Amazon’s pre-existing intangibles.\\nThe language of the (now-superseded) regulatory definition\\nof an “intangible” is ambiguous and could be construed as\\nincluding residual-business assets. But the drafting history\\nof the regulations and other indicators of Treasury’s\\ncontemporaneous intent strongly favor Amazon’s proffered\\nmeaning—that intangibles were limited to independently\\ntransferrable assets. Treasury appears to have changed its\\nposition on the meaning of the regulation after Amazon and\\nAEHT entered into their cost sharing arrangement. We share\\nthe sentiment reflected in the concurring opinion in Xilinx:\\n\\n Indeed, I am troubled by the complex,\\n theoretical nature of many of the\\n Commissioner’s arguments trying to\\n reconcile the two regulations. Not only does\\n this make it difficult for the court to navigate\\n the regulatory framework, it shows that\\n taxpayers have not been given clear, fair\\n notice of how the regulations will affect\\n them.\\n\\nXilinx, 598 F.3d at 1198 (Fisher, J., concurring). We\\ntherefore agree with the tax court that the former regulatory\\ndefinition of an “intangible” does not include residual-\\nbusiness assets.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4072,"cells":{"id":{"kind":"number","value":4649091,"string":"4,649,091"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Miguel Cano"},"decision_date":{"kind":"string","value":"2019-08-16"},"docket_number":{"kind":"string","value":"17-50151"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Susan P. Graber and Jay S. Bybee, Circuit Judges, and M. Douglas Harpool,* District Judge.\", \"parties\": \"\", \"opinions\": [{\"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n UNITED STATES OF AMERICA, No. 17-50151\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:16-cr-01770-BTM-1\\n\\n MIGUEL ANGEL CANO,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the Southern District of California\\n Barry Ted Moskowitz, District Judge, Presiding\\n\\n Argued and Submitted April 10, 2019\\n Pasadena, California\\n\\n Filed August 16, 2019\\n\\nBefore: Susan P. Graber and Jay S. Bybee, Circuit Judges,\\n and M. Douglas Harpool,* District Judge.\\n\\n Opinion by Judge Bybee\\n\\n\\n\\n\\n *\\n The Honorable M. Douglas Harpool, United States District Judge\\nfor the Western District of Missouri, sitting by designation\\n\\f2 UNITED STATES V. CANO\\n\\n SUMMARY**\\n\\n\\n Criminal Law\\n\\n The panel reversed the district court’s order denying the\\ndefendant’s motion to suppress evidence obtained from\\nwarrantless searches of his cell phone by Customs and Border\\nProtection officials, and vacated his conviction for importing\\ncocaine.\\n\\n Applying United States v. Cotterman, 709 F.3d 952 (9th\\nCir. 2013) (en banc), the panel held that manual cell phone\\nsearches may be conducted by border officials without\\nreasonable suspicion but that forensic cell phone searches\\nrequire reasonable suspicion. The panel clarified Cotterman\\nby holding that “reasonable suspicion” in this context means\\nthat officials must reasonably suspect that the cell phone\\ncontains digital contraband. The panel further concluded that\\ncell phone searches at the border, whether manual or forensic,\\nmust be limited in scope to whether the phone contains digital\\ncontraband; and that a broader search for evidence of a crime\\ncannot be justified by the purposes of the border search\\nexception to the Fourth Amendment warrant requirement.\\n\\n The panel held that to the extent that a Border Patrol\\nagent’s search of the defendant’s phone – which included the\\nrecording of phone numbers and text messages for further\\nprocessing – went beyond a verification that the phone lacked\\ndigital contraband, the search exceeded the proper scope of a\\nborder search and was unreasonable as a border search under\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. CANO 3\\n\\nthe Fourth Amendment. The panel held that although the\\nagents had reason to suspect the defendant’s phone would\\ncontain evidence leading to additional drugs, the record does\\nnot give rise to an objectively reasonable suspicion that the\\ndigital data in the phone contained contraband, and the border\\nsearch exception therefore did not authorize the agents to\\nconduct a warrantless forensic search of the defendant’s\\nphone. The panel held that the good faith exception to the\\nexclusionary rule does not apply because the border officials\\ndid not rely on binding appellate precedent specifically\\nauthorizing the cell phone searches at issue here.\\n\\n Rejecting the defendant’s contention that the government\\nviolated his rights under Brady v, Maryland, 373 U.S. 83\\n(1963), and Fed. R. Crim. P. 16, by failing to turn over certain\\ninformation he requested from the FBI and DEA in pursuit of\\nthis third-party defense, the panel found no evidence that the\\nprosecution had knowledge or possession of evidence\\nshowing that the defendant’s cousin or his cousin’s gang were\\ninvolved in drug trafficking at the Mexico-California border,\\nand held that the prosecutor should not be held to have\\n“access” to any information that an agency not involved in\\nthe investigation or prosecution of the case refuses to turn\\nover.\\n\\f4 UNITED STATES V. CANO\\n\\n COUNSEL\\n\\nHarini P. Raghupathi (argued), Federal Defenders of San\\nDiego, Inc., San Diego, California, for Defendant-Appellant.\\n\\nMark R. Rehe (argued), Assistant United States Attorney;\\nHelen H. Hong, Assistant United States Attorney, Chief,\\nAppellate Section, Criminal Division; Adam L. Braverman,\\nUnited States Attorney; United States Attorney’s Office, San\\nDiego, California; for Plaintiff-Appellee.\\n\\nSophia Cope and Adam Schwartz, Electronic Frontier\\nFoundation, San Francisco, California, for Amicus Curiae\\nElectronic Frontier Foundation.\\n\\n\\n OPINION\\n\\nBYBEE, Circuit Judge:\\n\\n Defendant-Appellant Miguel Cano was arrested for\\ncarrying cocaine as he attempted to cross into the United\\nStates from Mexico at the San Ysidro Port of Entry.\\nFollowing his arrest, a Customs and Border Protection\\nofficial seized Cano’s cell phone and searched it, first\\nmanually and then using software that accesses all text\\nmessages, contacts, call logs, media, and application data.\\nWhen Cano moved to suppress the evidence obtained from\\nthe warrantless searches of his cell phone, the district court\\nheld that the searches were valid under the border search\\nexception to the Fourth Amendment’s warrant requirement.\\n\\n Applying United States v. Cotterman, 709 F.3d 952 (9th\\nCir. 2013) (en banc), we conclude that manual cell phone\\n\\f UNITED STATES V. CANO 5\\n\\nsearches may be conducted by border officials without\\nreasonable suspicion but that forensic cell phone searches\\nrequire reasonable suspicion. We clarify Cotterman by\\nholding that “reasonable suspicion” in this context means that\\nofficials must reasonably suspect that the cell phone contains\\ndigital contraband. We further conclude that cell phone\\nsearches at the border, whether manual or forensic, must be\\nlimited in scope to a search for digital contraband. In this\\ncase, the officials violated the Fourth Amendment when their\\nwarrantless searches exceeded the permissible scope of a\\nborder search. Accordingly, we hold that most of the\\nevidence from the searches of Cano’s cell phone should have\\nbeen suppressed. We also conclude that Cano’s Brady claims\\nare unpersuasive. Because we vacate Cano’s conviction, we\\ndo not reach his claim of prosecutorial misconduct.\\n\\n We reverse the district court’s order denying Cano’s\\nmotion to suppress and vacate Cano’s conviction.\\n\\n I. THE BACKGROUND\\n\\nA. The Facts\\n\\n Defendant-Appellant Miguel Cano worked in the flooring\\nand carpet installation trade and lived with his wife and\\nchildren in the Mission Hills community north of Los\\nAngeles. In the summer of 2016, however, Cano moved from\\nLos Angeles to Tijuana, Mexico, where he stayed with his\\ncousin Jose Medina. While staying with Medina, Cano\\ncrossed the border into the United States six times, sometimes\\nremaining in the United States for less than thirty minutes.\\nOn two of those trips, Cano was referred to secondary\\ninspection, but no contraband was found.\\n\\f6 UNITED STATES V. CANO\\n\\n On July 25, 2016, Cano arrived at the San Ysidro Port of\\nEntry from Tijuana. In primary inspection, Cano stated that\\n“he was living in Mexico, working in San Diego, but going to\\nLA on that day.” Pursuant to a random Customs and Border\\nProtection (CBP) computer referral, Cano was referred to\\nsecondary inspection, where a narcotic-detecting dog alerted\\nto the vehicle’s spare tire. A CBP official removed the spare\\ntire from the undercarriage of the truck and discovered 14\\nvacuum-sealed packages inside, containing 14.03 kilograms\\n(30.93 pounds) of cocaine.\\n\\n Cano was arrested, and a CBP official administratively\\nseized his cell phone. The CBP officials called Homeland\\nSecurity Investigations (HSI), which dispatched Agents\\nPetonak and Medrano to investigate. After arriving, Agent\\nPetonak “briefly” and manually reviewed Cano’s cell phone,\\nnoticing a “lengthy call log” but no text messages. Agent\\nPetonak later stated that the purpose of this manual search\\nwas “two-pronged”: “to find some brief investigative leads in\\nthe current case,” and “to see if there’s evidence of other\\nthings coming across the border.”\\n\\n Agent Petonak proceeded to question Cano, who waived\\nhis Miranda rights and agreed to talk. During that interview,\\nCano denied any knowledge of the cocaine. Cano stated that\\nhe had moved to Tijuana to look for work in nearby San\\nDiego, because work was slow in Los Angeles. He also said\\nhe had crossed the border every day for the previous three\\nweeks looking for work. He told Agent Petonak that he was\\nheaded to a carpet store in Chula Vista that day to seek work.\\nWhen pressed, Cano was not able to provide the name or\\naddress of the store, claiming that he intended to look it up on\\nGoogle after crossing the border. Cano also explained that he\\ndid not have his flooring tools with him in his pickup truck so\\n\\f UNITED STATES V. CANO 7\\n\\nas to avoid problems with border crossings; Cano intended to\\ndrive to Los Angeles to retrieve his tools if he located work\\nin San Diego.\\n\\n During the interrogation, Agent Petonak specifically\\nasked Cano about the lack of text messages on his cell phone.\\nCano responded that his cousin had advised him to delete his\\ntext messages “just in case” he got pulled over in Mexico and\\npolice were to check his cell phone. Cano stated that he\\nerased his messages to avoid “any problems” with the\\nMexican police.\\n\\n While Agent Petonak questioned Cano, Agent Medrano\\nconducted a second manual search of the cell phone. Agent\\nMedrano browsed the call log and wrote down some of the\\nphone numbers on a piece of paper. He also noticed two\\nmessages that arrived after Cano had reached the border, and\\nhe took a photograph of the messages. The first message\\nstated, “Good morning,” and the second message stated,\\n“Primo, are you coming to the house?” Agent Medrano gave\\nall of this information—the recorded list of calls and the\\nphotograph—to Agent Petonak.\\n\\n Finally, Agent Medrano conducted a “logical download”\\nof the phone using Cellebrite software. A Cellebrite search\\nenables the user to access text messages, contacts, call logs,\\nmedia, and application data on a cell phone and to select\\nwhich types of data to download. It does not, however, allow\\nthe user to access data stored within third-party applications.\\nAgent Medrano typically does not select the option to\\ndownload photographs.\\n\\n After Agent Petonak interviewed Cano, he reviewed the\\nresults of the Cellebrite download of Cano’s phone by Agent\\n\\f8 UNITED STATES V. CANO\\n\\nMedrano. The Cellebrite results revealed that Cano had sent\\nno text messages, and it listed all the calls made by Cano.\\nAgent Petonak later concluded that none of the phone\\nnumbers in the call log corresponded to carpeting stores in\\nSan Diego.\\n\\nB. The Proceedings\\n\\n Cano was indicted for importing cocaine. Before trial,\\nCano moved to suppress any evidence obtained from Agents\\nPetonak and Medrano’s warrantless searches of his cell phone\\nat the border. The district court denied Cano’s motion, ruling\\nthat the manual searches and the Cellebrite search of Cano’s\\nphone were valid border searches. During trial, the\\ngovernment introduced evidence that resulted from the\\nmanual searches of the phone and from Agent Medrano’s\\nCellebrite download of the phone.1\\n\\n In preparation for trial, Cano indicated his intent to\\npresent a third-party culpability defense claiming that his\\ncousin, Jose Medina, was responsible for placing the drugs in\\nCano’s spare tire without Cano’s knowledge. Cano proffered\\nevidence that Medina had a key to Cano’s car and had driven\\nit shortly before Cano’s attempted border crossing, that\\n\\n\\n 1\\n Some—but not all—of the evidence was available through\\nalternative channels. For example, the government introduced a call log,\\nunchallenged by Cano, that the government received from Cano’s phone\\ncompany. Similarly, the government later obtained a warrant to search the\\nphone, and an agent conducted further searches. Because the government\\nintroduced at trial much evidence pre-dating those events, and because the\\ngovernment has not argued that any Fourth Amendment error was\\nharmless, those later events do not affect our Fourth Amendment analysis\\nof the warrantless searches. United States v. Rodriguez, 880 F.3d 1151,\\n1163 (9th Cir. 2018)\\n\\f UNITED STATES V. CANO 9\\n\\nMedina had a criminal record including a conviction for\\ncocaine possession, that Medina was a member of a Chicago-\\nbased gang called the Latin Kings, and that the Latin Kings\\nsold cocaine within the United States and were involved with\\na cartel that trafficked drugs across the border.\\n\\n Following Cano’s implication of Medina, the government\\ncontacted Medina and promised him immunity and\\nimmigration papers in exchange for his cooperation. Medina\\ninitially denied being involved with drugs, but later contacted\\nthe government on his own and offered to help them with the\\n“biggest RICO case” and “drug seizures of 20 to 25\\nkilograms at a time.” All of this information was made\\navailable to Cano.\\n\\n As part of his defense, Cano sought additional discovery\\nfrom HSI, the Federal Bureau of Investigation (FBI), and the\\nDrug Enforcement Agency (DEA) regarding: (1) records\\nlinking Medina to drug sales, distribution, or trafficking; and\\n(2) records linking the Latin Kings to drug trafficking from\\nMexico to Southern California. The government opposed\\nCano’s discovery motion, arguing that the evidence was not\\nmaterial under Federal Rule of Criminal Procedure\\n16(a)(1)(E)(i) and that discovery should be limited to HSI, as\\nneither the DEA nor the FBI had participated in the\\ninvestigation of Cano. The district court originally overruled\\nboth objections, finding the evidence material under Rule 16\\nand exculpatory under Brady v. Maryland, 373 U.S. 83\\n(1963). The court also reasoned that, because HSI could\\ninquire of the DEA and FBI if it sought inculpatory evidence,\\nHSI had access to the files and was required to provide any\\nexculpatory evidence held by the DEA or FBI.\\n\\f10 UNITED STATES V. CANO\\n\\n In response to the court’s discovery order, HSI produced\\nMedina’s immigration file and his Bureau of Prisons record.\\nAgent Petonak also searched for Medina’s name in two\\ndifferent police clearinghouses, but neither returned any hits.2\\nBoth Agent Petonak and the United States Attorney’s Office\\n(USAO) subsequently requested information showing a link\\nbetween the Latin Kings and drug trafficking from Mexico\\nfrom the legal counsel of both the FBI and DEA. Both\\nagencies denied the requests without providing any\\nexplanation or any indication as to whether the requested\\ninformation existed.\\n\\n Following these attempts, the government moved for the\\ndistrict court to reconsider its discovery order and excuse it\\nfrom discovery relating to files held by the FBI and DEA.\\nThe district court granted the motion to reconsider, finding\\nthat the prosecutor did not have access to the evidence when\\nhe was “rebuffed” by agencies over which he had no control.\\n\\n The case proceeded to trial and Cano presented his third-\\nparty culpability defense. The first trial resulted in a hung\\njury and a mistrial. On retrial, Cano again relied on his third-\\nparty culpability defense. The second trial resulted in Cano’s\\nconviction. This appeal followed, in which Cano raises three\\nissues: (1) whether the warrantless searches of his cell phone\\nviolated the Fourth Amendment and whether the resulting\\nevidence should be suppressed; (2) whether the government’s\\nnon-disclosure of materials that may have been held by the\\nDEA and FBI violated his right to due process under Brady\\n\\n\\n 2\\n A police clearinghouse works for the purpose of “deconfliction” by\\nnotifying an agency if another agency has an investigation pending against\\nthe same person or item. The DEA and FBI participate in the two\\nclearinghouses searched by Agent Petonak.\\n\\f UNITED STATES V. CANO 11\\n\\nand Federal Rule of Criminal Procedure 16; and (3) whether\\nthe government raised an improper propensity inference in its\\nclosing argument. We address Cano’s first two arguments in\\nturn. Because we conclude that the district court erred in\\ndenying Cano’s motion to suppress, we vacate Cano’s\\nconviction and do not reach his claim of prosecutorial\\nmisconduct.\\n\\n II. THE WARRANTLESS SEARCH OF CANO’S\\n CELL PHONE\\n\\n The Fourth Amendment protects “[t]he right of the people\\nto be secure in their persons, houses, papers, and effects,\\nagainst unreasonable searches and seizures.” U.S. Const.\\namend. IV.3 Ordinarily, before conducting a search, police\\nmust obtain a warrant issued by a judicial officer based “upon\\nprobable cause, supported by Oath or affirmation, and\\nparticularly describing the place to be searched, and the\\npersons or things to be seized.” Id. Warrants are generally\\nrequired “unless ‘the exigencies of the situation’ make the\\nneeds of law enforcement so compelling that the warrantless\\nsearch is objectively reasonable under the Fourth\\nAmendment.” Mincey v. Arizona, 437 U.S. 385, 393–94\\n(1978) (citation omitted). Consequently, “searches conducted\\noutside the judicial process, without prior approval by judge\\nor magistrate, are per se unreasonable under the Fourth\\nAmendment—subject only to a few specifically established\\nand well-delineated exceptions.” Katz v. United States, 389\\nU.S. 347, 357 (1967) (footnote omitted). Such “specifically\\nestablished and well-delineated exceptions” include exigent\\n\\n\\n 3\\n We review de novo “the district court’s determination that [a]\\nwarrantless search . . . was a valid border search.” United States v.\\nCardona, 769 F.2d 625, 628 (9th Cir. 1985).\\n\\f12 UNITED STATES V. CANO\\n\\ncircumstances, searches incident to arrest, vehicle searches,\\nand border searches. See Arizona v. Gant, 556 U.S. 332, 343\\n(2009) (vehicle searches); Brigham City v. Stuart, 547 U.S.\\n398, 403 (2006) (exigent circumstances; listing other\\nexceptions, including warrantless entry to fight a fire, to\\nprevent the imminent destruction of evidence, or in “hot\\npursuit” of a fleeing suspect); United States v. Ramsey, 431\\nU.S. 606, 616 (1977) (border searches); Weeks v. United\\nStates, 232 U.S. 383, 392 (1914) (searches incident to arrest),\\noverruled in part on other grounds by Mapp v. Ohio, 367\\nU.S. 643 (1961).\\n\\n Exceptions to the warrant requirement are subject to two\\nimportant constraints. First, any search conducted under an\\nexception must be within the scope of the exception. Second,\\nsome searches, even when conducted within the scope of the\\nexception, are so intrusive that they require additional\\njustification, up to and including probable cause and a\\nwarrant.\\n\\n The first constraint is illustrated by the Supreme Court’s\\ndecision in Riley v. California, 573 U.S. 373 (2014), a case\\ninvolving the search incident to arrest exception. In Riley, the\\nCourt addressed “whether the police may, without a warrant,\\nsearch digital information on a cell phone seized from an\\nindividual who has been arrested”; in other words, whether\\ncell phones fell within the scope of the search incident to\\narrest exception. Id. at 378. The Court began by recognizing\\nthe increasing role in our lives of “minicomputers that also\\nhappen to have the capacity to be used as a telephone”;\\n“[m]odern cell phones, as a category, implicate privacy\\nconcerns far beyond those implicated by the search of a\\ncigarette pack, a wallet, or a purse.” Id. at 393.\\nAcknowledging that “it has been well accepted that [a search\\n\\f UNITED STATES V. CANO 13\\n\\nincident to lawful arrest] constitutes an exception to the\\nwarrant requirement,” id. at 382, the Court pointed out that\\nsuch searches serve two purposes: (1) to secure “the officer’s\\nsafety” and (2) to “prevent . . . concealment or destruction [of\\nevidence],” id. at 383 (citation omitted). The Court then\\nconsidered whether a cell phone search qualified as a search\\nincident to arrest by considering “whether application of the\\nsearch incident to arrest doctrine to [cell phones] would\\n‘untether the rule from the justifications underlying the . . .\\nexception.’” Id. at 386 (quoting Gant, 556 U.S. at 343).\\n\\n The Court concluded that neither purpose for the search\\nincident to arrest exception justified the search of a cell\\nphone. The Court rejected the government’s argument that\\nsearching a cell phone incident to arrest would “help ensure\\nofficer safety in . . . indirect ways, for example by alerting\\nofficers that confederates of the arrestee are headed to the\\nscene.” Id. at 387. The Court reasoned that the government’s\\nposition “would . . . represent a broadening” of the\\nexception’s foundational concern that “an arrestee himself\\nmight grab a weapon and use it against an officer.” Id. at\\n387–88. The Court observed that “once law enforcement\\nofficers have secured a cell phone, there is no longer any risk\\nthat the arrestee himself will be able to delete incriminating\\ndata from the phone,” id. at 388, and police have means to\\nensure that data cannot be wiped from the phone remotely,\\nid. at 390. The Court concluded “not that the information on\\na cell phone is immune from search; [but rather] that a\\nwarrant is generally required before such a search, even when\\na cell phone is seized incident to arrest.” Id. at 401.\\n\\n The second constraint on warrantless searches is\\nillustrated by the Court’s decision in United States v.\\nMontoya de Hernandez, 473 U.S. 531 (1985). Montoya was\\n\\f14 UNITED STATES V. CANO\\n\\nstopped at Los Angeles International Airport and referred to\\nsecondary inspection. Id. at 533. She had arrived from\\nBogota and was carrying $5,000 in cash. Id. She had no\\ncredit cards and no hotel reservations. Id. at 533–34.\\nBecause border officials suspected that Montoya may have\\nswallowed cocaine-filled balloons, Montoya was held in the\\ncustoms office and, after a magistrate judge issued an order,\\ntaken to a hospital for a rectal examination. Id. at 534–35.\\nOver the next four days, she passed 88 balloons containing\\ncocaine. Id. at 536. Montoya argued that the search she was\\nsubjected to, though a border search, was so intrusive that it\\ncould not be conducted without a high level of particularized\\nsuspicion. Id. at 536–37, 540. The Court balanced her\\nprivacy interests against the interests of the government at the\\nborder and concluded that, while routine searches may be\\nconducted at the border without any showing of suspicion, a\\nmore intrusive, nonroutine search must be supported by\\n“reasonable suspicion.” Id. at 537–41; see also United States\\nv. Flores-Montano, 541 U.S. 149, 152 (2004) (suggesting that\\nnonroutine searches are limited to “highly intrusive searches\\nof the person” involving “dignity and privacy interests”).\\n\\n Cano recognizes that he was subject to search at the\\nborder, but Cano and amicus Electronic Frontier Foundation\\n(“EFF”) raise two categorical challenges and one as-applied\\nchallenge to the searches conducted here. First, EFF argues\\nthat any warrantless search of a cell phone falls outside the\\nscope of the border search exception. Second, EFF argues\\nthat even if the search is within the scope of the border search\\nexception, a warrantless cell phone search is so intrusive that\\nit requires probable cause. We address these categorical\\nchallenges in Part II.A. Third, Cano asserts that, even if cell\\nphones are generally subject to search at the border, the\\nmanual and forensic searches of his cell phone exceeded the\\n\\f UNITED STATES V. CANO 15\\n\\n“well delineated” scope of the border search. We address this\\nas-applied question in Part II.B. Finally, the government\\nargues that even if the border search exceeded the limits of\\nthe Fourth Amendment, the search was conducted in good\\nfaith, and the evidence is admissible. We consider the good\\nfaith exception in Part II.C.\\n\\nA. Border Searches and Cell Phones\\n\\n “[B]order searches constitute a ‘historically recognized\\nexception to the Fourth Amendment’s general principle that\\na warrant be obtained.’” Cotterman, 709 F.3d at 957 (quoting\\nRamsey, 431 U.S. at 621). Indeed, border searches typically\\ndo not require any particularized suspicion, so long as they\\nare “routine inspections and searches of individuals or\\nconveyances seeking to cross our borders.” Almeida-Sanchez\\nv. United States, 413 U.S. 266, 272 (1973); see United States\\nv. Seljan, 547 F.3d 993, 999 (9th Cir. 2008) (en banc). Such\\nsearches are “reasonable simply by virtue of the fact they\\noccur at the border.” Ramsey, 431 U.S. at 616. The\\nexception is “rooted in ‘the long-standing right of the\\nsovereign to protect itself by stopping and examining persons\\nand property crossing into this country,’” Cotterman, 709\\nF.3d at 960 (quoting Ramsey, 431 U.S. at 616), to “prevent[]\\nthe entry of unwanted persons and effects,” id. (quoting\\nFlores-Montano, 541 U.S. at 152).\\n\\n The sovereign’s right to conduct suspicionless searches at\\nthe border “does not mean, however, that at the border\\n‘anything goes.’” Id. (quoting Seljan, 547 F.3d at 1000).\\nRather, the border search exception is a “narrow exception”\\nthat is limited in two important ways. Id. (citation omitted).\\nFirst, “[t]he authorizing statute limits the persons who may\\nlegally conduct a ‘border search’ to ‘persons authorized to\\n\\f16 UNITED STATES V. CANO\\n\\nboard or search vessels.’” United States v. Soto-Soto, 598\\nF.2d 545, 549 (9th Cir. 1979) (citing 19 U.S.C. § 482).4 This\\nincludes customs and immigration officials, but not general\\nlaw enforcement officers such as FBI agents. Id.; see United\\nStates v. Diamond, 471 F.2d 771, 773 (9th Cir. 1973) (stating\\nthat “customs agents are not general guardians of the public\\npeace”). Second, a border search must be conducted “in\\nenforcement of customs laws.” Soto-Soto, 598 F.2d at 549.\\nA border search must be conducted to “enforce importation\\nlaws,” and not for “general law enforcement purposes.” Id.\\n\\n\\n 4\\n Section 482 now reads in relevant part:\\n\\n Any of the officers or persons authorized to board\\n or search vessels may stop, search, and examine . . . any\\n vehicle, beast, or person, on which or whom he or they\\n shall suspect there is merchandise which is subject to\\n duty, or shall have been introduced into the United\\n States in any manner contrary to law . . . . [and may]\\n seize and secure the same for trial.\\n\\n19 U.S.C. § 482(a); see id. § 1467 (“[T]he appropriate customs officer for\\n[a] port or place of arrival may . . . enforce, cause inspection, examination,\\nand search to be made of the persons, baggage, and merchandise\\ndischarged or unladen from [an arriving] vessel . . . .”); id. § 1496 (“The\\nappropriate customs officer may cause an examination to be made of the\\nbaggage of any persons arriving in the United States in order to ascertain\\nwhat articles are contained therein and whether subject to duty, free of\\nduty, or prohibited . . . .”); id. § 1582 (“[A]ll persons coming into the\\nUnited States from foreign countries shall be liable to detention and search\\nby authorized officers or agents . . . .”).\\n\\n The Court has described § 482 as granting the executive “plenary\\nauthority to conduct routine searches and seizures at the border, without\\nprobable cause or a warrant.” Montoya de Hernandez, 473 U.S. at 537.\\nWe have held that the “outer limits of authority delegated by [§ 482 are]\\navailable only in border searches.” Corngold v. United States, 367 F.2d\\n1, 3 (9th Cir. 1966) (en banc).\\n\\f UNITED STATES V. CANO 17\\n\\nA general search cannot be “justif[ied] . . . on the mere basis\\nthat it occurred at the border.” Id. (affirming the suppression\\nof evidence where an FBI agent stopped and searched the\\nvehicle of an alien to determine whether the car had been\\nstolen).\\n\\n 1. Cell Phone Data as Contraband\\n\\n As we discussed briefly above, the Supreme Court has\\nidentified two principal purposes behind warrantless border\\nsearches: First, to identify “[t]ravellers . . . entitled to come\\nin” and, second, to verify their “belongings as effects which\\nmay be lawfully brought in.” Carroll v. United States, 267\\nU.S. 132, 154 (1925); see Ramsey, 431 U.S. at 620 (“The\\nborder-search exception is grounded in the recognized right\\nof the sovereign to control . . . who and what may enter the\\ncountry.”).\\n\\n EFF argues that applying the border search exception to\\na cell phone’s data would “untether” the exception from the\\npurposes underlying it. EFF contends that a border search\\nencompasses only a search for illegal persons and physical\\ncontraband located on the body of the applicant for\\nadmission or among his effects. Because digital data on a cell\\nphone cannot conceal objects such as drugs, guns, or\\nsmuggled persons, EFF asserts that digital cell phone\\nsearches are always beyond the scope of the border search\\nexception.\\n\\n We agree with EFF that the purpose of the border search\\nis to interdict contraband, but we disagree with its premise\\nthat cell phones cannot contain contraband. Although cell\\n\\f18 UNITED STATES V. CANO\\n\\nphone data cannot hide physical objects,5 the data can contain\\ndigital contraband. The best example is child pornography.\\nSee United States v. Molina-Isidoro, 884 F.3d 287, 295 n.3\\n(5th Cir. 2018) (Costa, J., specially concurring) (“One type of\\ncontraband that can be stored within the data of a cell\\nphone . . . is child pornography.”). And because cell phones\\nmay ultimately be released into the interior, even if the owner\\nhas been detained, the United States has a strong interest in\\npreventing the entry of such material. See, e.g., United States\\nv. Vergara, 884 F.3d 1309, 1311 (11th Cir.) (describing how\\nagents returned one of the defendant’s phones to a family\\nmember after defendant had been arrested for possessing\\nchild pornography on his other two phones), cert. denied, 139\\nS. Ct. 70 (2018). We find no basis for the proposition that the\\nborder search exception is limited to searching for physical\\ncontraband. At the very least, a cell phone that has photos\\nstored on it is the equivalent of photographs, magazines, and\\nbooks.6 See Riley, 573 U.S. at 394; Cotterman, 709 F.3d at\\n964. The contents may be digital when they are on the phone,\\nbut the physicality of the phone itself and the possibility that\\n\\n\\n 5\\n No one contests that a border official could, consistent with the\\nFourth Amendment, examine the physical body of a cell phone to see if\\nthe phone itself is contraband—because, for example, it is a pirated copy\\nof a patented U.S. phone—or if the phone itself presents a physical threat\\nto officers. See Riley, 573 U.S. at 387 (“Law enforcement officers remain\\nfree to examine the physical aspects of a phone to ensure that it will not\\nbe used as a weapon—say, to determine whether there is a razor blade\\nhidden between the phone and its case.”). The dispute here concerns only\\nwhether border officials may search the digital data contained within the\\nphone.\\n 6\\n We need not address here questions surrounding the use of “cloud\\ncomputing,” where the phone gives access to, but does not contain in its\\nown memory, digital data stored in the cloud. See Riley, 573 U.S. at\\n397–98; Cotterman, 709 F.3d at 965 & n.12.\\n\\f UNITED STATES V. CANO 19\\n\\nthe phone’s contents can be printed or shared electronically\\ngives border officials sufficient reason to inspect it at the\\nborder. We conclude that cell phones—including the phones’\\ndata—are subject to search at the border.\\n\\n 2. Forensic Cell Phone Searches as an Intrusive Search\\n\\n The second question we must address in response to\\namicus EFF is whether forensic searches of a cell phone are\\nso intrusive that they require reasonable suspicion or even\\nprobable cause. We answered this question in our en banc\\ndecision in Cotterman, but with respect to laptop computers.7\\nCotterman, 709 F.3d at 962–68. Cotterman was a United\\nStates citizen returning to the United States from Mexico. Id.\\nat 957. When he reached the port of entry, border officials\\nnoted that Cotterman had various convictions for sexual\\nconduct with children. Id. Concerned that Cotterman might\\nbe involved in child sex tourism, officials conducted a brief\\nsearch of his laptop computers and digital cameras and noted\\nthat the laptops had password-protected files. Id. at 958. The\\nofficials detained the computers for several days in order to\\nrun a comprehensive forensic search of the hard drive, which\\nrevealed hundreds of images of child pornography. Id. at\\n958–59. For us, “the legitimacy of the initial search of\\nCotterman’s electronic devices at the border [was] not in\\ndoubt,” id. at 960, “[t]he difficult question . . . [was] the\\nreasonableness, without a warrant, of the forensic\\n\\n\\n\\n 7\\n Although Cotterman referred to “electronic devices” generally, see\\n709 F.3d at 962–68, our holding was limited to the “examination of\\nCotterman’s computer,” id. at 968, and did not address cell phones. We\\nmentioned cell phones only once—in the first paragraph of the\\nintroduction describing the modern “digital world.” Id. at 956.\\n\\f20 UNITED STATES V. CANO\\n\\nexamination that comprehensively analyzed the hard drive of\\nthe computer,” id. at 961.\\n\\n We acknowledged the “substantial personal privacy\\ninterests” in “[e]lectronic devices . . . capable of storing\\nwarehouses full of information.” Id. at 964. At the same\\ntime, we recognized “the important security concerns that\\nprevail at the border” and the legitimacy of “[t]he effort to\\ninterdict child pornography.” Id. at 966. We held that a\\nroutine, manual search of files on a laptop computer—“a\\nquick look and unintrusive search”—is reasonable “even\\nwithout particularized suspicion,” but that officials must\\n“possess a particularized and objective basis for suspecting\\nthe person stopped of criminal activity” to engage in a\\nforensic examination, which is “essentially a computer strip\\nsearch.” Id. at 960–61, 966, 967 (citation omitted). We\\nconcluded that reasonable suspicion was “a modest, workable\\nstandard that is already applied in the extended border search,\\nTerry stop, and other contexts.” Id. at 966; see id. at 968\\n(defining reasonable suspicion as “a particularized and\\nobjective basis for suspecting the particular person stopped of\\ncriminal activity” (quoting United States v. Cortez, 449 U.S.\\n411, 417–18 (1981))).\\n\\n We think that Cotterman’s reasoning applies equally to\\ncell phones. In large measure, we anticipated the Supreme\\nCourt’s reasoning in Riley, 573 U.S. at 393–97, when we\\nrecognized in Cotterman that digital devices “contain the\\nmost intimate details of our lives” and “the uniquely sensitive\\nnature of data on electronic devices carries with it a\\nsignificant expectation of privacy,” Cotterman, 709 F.3d at\\n965–66; see Riley, 573 U.S. at 385, 393 (describing cell\\nphones as “a pervasive and insistent part of daily life” that,\\n“as a category, implicate privacy concerns far beyond those\\n\\f UNITED STATES V. CANO 21\\n\\nimplicated by the search of a cigarette pack, a wallet, or a\\npurse”). The Court’s view of cell phones in Riley so closely\\nresembles our own analysis of laptop computers in Cotterman\\nthat we find no basis to distinguish a forensic cell phone\\nsearch from a forensic laptop search.8\\n\\n Nor do we believe that Riley renders the Cotterman\\nstandard insufficiently protective. Riley, of course, held that\\n“a warrant is generally required” before searching a cell\\nphone, “even when a cell phone is seized incident to arrest.”\\n573 U.S. at 401. But here we deal with the border search\\nexception—not the search incident to arrest exception—and\\nthe difference in context is critical. In light of the\\ngovernment’s enhanced interest in protecting the “integrity of\\nthe border” and the individual’s decreased expectation of\\nprivacy, the Court has emphasized that “the Fourth\\nAmendment’s balance of reasonableness is qualitatively\\ndifferent at the international border than in the interior” and\\nis “struck much more favorably to the Government.”\\nMontoya de Hernandez, 473 U.S. at 538–40. As a result,\\npost-Riley, no court has required more than reasonable\\nsuspicion to justify even an intrusive border search. See\\nUnited States v. Wanjiku, 919 F.3d 472, 485 (7th Cir. 2019)\\n(“[N]o circuit court, before or after Riley, has required more\\nthan reasonable suspicion for a border search of cell phones\\nor electronically-stored data.”); Touset, 890 F.3d at 1234\\n\\n 8\\n We note that the Eleventh Circuit disagreed with Cotterman in\\nUnited States v. Touset, 890 F.3d 1227, 1234 (11th Cir. 2018). The court\\nheld that no level of suspicion was required to conduct a forensic search\\nof a cell phone. Id. at 1234–35. Nevertheless, the Touset court held, in\\nthe alternative, that the forensic search of various electronic devices seized\\nat the border were supported by reasonable suspicion. Id. at 1237. As\\nwith most cell phone search cases, in Touset border agents were looking\\nfor child pornography.\\n\\f22 UNITED STATES V. CANO\\n\\n(“Riley, which involved the search-incident-to-arrest\\nexception, does not apply to searches at the border.”);\\nMolina-Isidoro, 884 F.3d at 291 (“For border searches both\\nroutine and not, no case has required a warrant.”); id. at 293\\n(“The bottom line is that only two of the many federal cases\\naddressing border searches of electronic devices have ever\\nrequired any level of suspicion. They both required only\\nreasonable suspicion and that was for the more intrusive\\nforensic search.”); see also Kolsuz, 890 F.3d 133, 137 (4th\\nCir. 2018) (concluding that a “forensic examination of\\nKolsuz’s phone must be considered a nonroutine border\\nsearch, requiring some measure of individualized suspicion”\\nbut declining to decide whether the standard should be\\nreasonable suspicion or probable cause).\\n\\n Accordingly, we hold that manual searches of cell phones\\nat the border are reasonable without individualized suspicion,\\nwhereas the forensic examination of a cell phone requires a\\nshowing of reasonable suspicion. See Cotterman, 709 F.3d\\nat 968.\\n\\nB. The Searches of Cano’s Cell Phone and the Scope of the\\n Border Search Exception\\n\\n Having concluded that border officials may conduct\\nsuspicionless manual searches of cell phones, but must have\\nreasonable suspicion before they conduct a forensic search,\\nwe still must address the core of Cano’s argument: whether\\nthe manual and forensic searches of his cell phone were not\\nsearches for digital contraband, but searches for evidence of\\na crime, and thus exceeded the proper scope of a border\\nsearch.\\n\\f UNITED STATES V. CANO 23\\n\\n 1. The Border Exception and the Search for Contraband\\n\\n As a threshold matter, Cano argues that border searches\\nare limited in both purpose and scope to searches for\\ncontraband.9 In response, the government argues that\\nsearches for evidence that would aid in prosecuting past and\\npreventing future border-related crimes are tethered to the\\npurpose of the border search exception—namely, interdicting\\nforeign contraband—and thus fall within its scope.\\n\\n\\n\\n\\n 9\\n Cano emphasizes that the officials who arrested him were looking\\nfor evidence of a crime, not contraband that could be seized at the border,\\nand this renders the search unconstitutional. He points to Officers Petonak\\nand Medrano, who searched Cano’s cell phone, and who testified that their\\nsearches had a dual purpose: “to find some brief investigative leads in the\\ncurrent case” and “to see if there[] [was] evidence of other things coming\\nacross the border.” Because the agents acknowledged that they sought\\nevidence to use against Cano in building a criminal case, Cano argues that\\nthe court should treat the search as one conducted for “general law\\nenforcement purposes” rather than a border search.\\n\\n Cano’s focus on the officials’ subjective motivations is misplaced,\\nhowever. As the district court recognized, “courts have repeatedly held\\nthat the Fourth Amendment’s reasonableness analysis is ‘predominantly\\nan objective inquiry.’” See Whren v. United States, 517 U.S. 806, 813\\n(1996) (upholding a “pretextual” stop because “[s]ubjective intentions\\nplay no role in ordinary . . . Fourth Amendment analysis”). We have\\nupheld border searches of persons seeking entry even when those searches\\nwere conducted “at the behest” of DEA agents seeking criminal evidence.\\nSee United States v. Schoor, 597 F.2d 1303, 1305–06 (9th Cir. 1979)\\n(holding a border search reasonable where it was conducted “at the\\nbehest” of DEA agents and included a search for certain items of evidence\\nin addition to a search for contraband). Thus, the mere fact that Officers\\nPetonak and Medrano subjectively hoped to find “investigative leads”\\npertaining to the seized shipment of cocaine does not render their searches\\nof Cano’s phone beyond the border search exception.\\n\\f24 UNITED STATES V. CANO\\n\\n This is a close question, but we think Cano has the better\\nof the argument. There is a difference between a search for\\ncontraband and a search for evidence of border-related\\ncrimes, although the distinction may not be apparent.\\nCotterman helps us focus on the difference. There, border\\nofficials had been alerted that Cotterman had a criminal\\nrecord of sex abuse of minors and might be involved in “child\\nsex tourism.” Cotterman, 709 F.3d at 957. The officials\\nseized his laptop and subjected it to searches for child\\npornography, which they found. In Cotterman, the child\\npornography was contraband subject to seizure at the border.\\nAs contraband, the child pornography is also evidence of\\nvarious crimes, including possession of child pornography,\\n18 U.S.C. § 2252A(a)(5)(B), and importation of obscene\\nmaterial, 18 U.S.C. § 1462(a). But nothing in Cotterman\\nauthorized border officials to conduct a search for evidence\\nthat Cotterman was involved in sex-related crimes generally.\\n\\n Border officials are authorized to seize “merchandise\\nwhich . . . shall have been introduced into the United States\\nin any manner contrary to law.” 19 U.S.C. § 482(a)\\n(emphasis added). The photos on Cotterman’s laptop\\ncomputer were such merchandise. 18 U.S.C. § 2252(a). But\\nborder officials have no general authority to search for crime.\\nThis is true even if there is a possibility that such crimes may\\nbe perpetrated at the border in the future. So, for example, if\\nU.S. officials reasonably suspect that a person who has\\npresented himself at the border may be engaged in price\\nfixing, see 15 U.S.C. § 1, they may not conduct a forensic\\nsearch of his phone or laptop. Evidence of price fixing—\\ntexts or emails, for example—is not itself contraband whose\\nimportation is prohibited by law. Such emails may be\\nevidence of a crime, but they are not contraband, and there is\\nno law prohibiting the importation of mere evidence of crime.\\n\\f UNITED STATES V. CANO 25\\n\\n We recognize that our analysis is in tension with the\\nFourth Circuit’s decision in Kolsuz. Kolsuz was detained at\\nWashington Dulles International Airport when customs\\nagents discovered firearm parts in his luggage. Kolsuz, 890\\nF.3d at 138–39. Kolsuz was arrested and his cell phone\\nseized. Id. at 139. The agents subjected the phone to a\\nmonth-long forensic search, producing a 896-page report. Id.\\nKolsuz challenged the search, which the district court upheld\\nand the Fourth Circuit affirmed. Id. at 139–42. The court\\napproved the forensic search because the agents had “reason\\nto believe . . . that Kolsuz was attempting to export firearms\\nillegally” and that “their search would reveal not only\\nevidence of the export violation they already had detected,\\nbut also ‘information related to other ongoing attempts to\\nexport illegally various firearm parts.’” Id. at 143 (quoting\\nthe district court; citation omitted). According to the Fourth\\nCircuit, “[t]he justification behind the border search\\nexception is broad enough to accommodate not only the\\ndirect interception of contraband as it crosses the border, but\\nalso the prevention and disruption of ongoing efforts to\\nexport contraband illegally.” Id. (emphasis added).10\\n\\n We agree with much of the Fourth Circuit’s discussion of\\nfoundational principles, but we respectfully disagree with the\\nfinal step approving the search for further evidence that\\n\\n\\n 10\\n As support for this proposition, the Fourth Circuit cited two district\\ncourt cases originating within our circuit. Both of those cases addressed\\nfact-patterns almost identical to Cano’s, and in each case the district court\\nheld that the border-search exception was not limited to searching for\\ncontraband directly. See United States v. Mendez, 240 F. Supp. 3d 1005,\\n1007–08 (D. Ariz. 2017); United States v. Ramos, 190 F. Supp. 3d 992,\\n999 (S.D. Cal. 2016). In neither case was the issue appealed to our circuit.\\nThus, Cano’s case presents the first opportunity for us to consider the\\nmatter.\\n\\f26 UNITED STATES V. CANO\\n\\nKolsuz was smuggling weapons. Our disagreement focuses\\nprecisely on the critical question that we previously\\nidentified: Does the proper scope of a border search include\\nthe power to search for evidence of contraband that is not\\npresent at the border? Or, put differently, can border agents\\nconduct a warrantless search for evidence of past or future\\nborder-related crimes? We think that the answer must be\\n“no.” The “[d]etection of . . . contraband is the strongest\\nhistoric rationale for the border-search exception.” Molina-\\nIsidoro, 884 F.3d at 295 (Costa, J., specially concurring).\\nIndeed, “every border-search case the Supreme Court has\\ndecided involved searches to locate items being smuggled”\\nrather than evidence. Id. (emphasis added); see Montoya de\\nHernandez, 473 U.S. at 537 (the border search is “to prevent\\nthe introduction of contraband into this country”); United\\nStates v. 12 200-Foot Reels of Super 8mm. Film, 413 U.S.\\n123, 125 (1973) (border searches are “necessary to prevent\\nsmuggling and to prevent prohibited articles from entry”);\\nUnited States v. Thirty-Seven Photographs, 402 U.S. 363, 376\\n(1971) (“Customs officers characteristically inspect luggage\\nand their power to do so is not questioned in this case; it is an\\nold practice and is intimately associated with excluding\\nillegal articles from the country”). In fact, the Court has long\\n“draw[n] a sharp distinction between searches for contraband\\nand those for evidence that may reveal the importation of\\ncontraband.” Molina-Isidoro, 884 F.3d at 296 (Costa, J.,\", \"author\": \"BYBEE, Circuit Judge:\"}, {\"type\": \"concurrence\", \"author\": \"specially\", \"text\": \"specially concurring). The classic statement on the\\ndistinction between seizing goods at the border because their\\nimportation is prohibited and seizing goods at the border\\nbecause they may be useful in prosecuting crimes is found in\\nBoyd v. United States:\\n\\n Is a search and seizure, or, what is equivalent\\n thereto, a compulsory production of a man’s\\n\\f UNITED STATES V. CANO 27\\n\\n private papers, to be used in evidence against\\n him in a proceeding to forfeit his property for\\n alleged fraud against the revenue laws—is\\n such a proceeding for such a purpose an\\n “unreasonable search and seizure” within the\\n meaning of the fourth amendment of the\\n constitution? . . . . The search for and seizure\\n of stolen or forfeited goods, or goods liable to\\n duties and concealed to avoid the payment\\n thereof, are totally different things from a\\n search for and seizure of a man’s private\\n books and papers for the purpose of obtaining\\n information therein contained, or of using\\n them as evidence against him. The two things\\n differ toto coelo.\\n\\n116 U.S. 616, 622–23 (1886), overruled in part on other\\ngrounds by Warden, Md. Penitentiary v. Hayden, 387 U.S.\\n294 (1967); see also id. at 633 (stating that compelling a man\\nto produce the evidence against himself not only violates the\\nFifth Amendment, but makes the seizure of his “books and\\npapers” unreasonable under the Fourth Amendment).\\n\\n Although we continue to acknowledge that “[t]he\\nGovernment’s interest in preventing the entry of unwanted\\npersons and effects is at its zenith at the international border”\\nand that “the expectation of privacy is less at the border than\\nit is in the interior,” Flores-Montano, 541 U.S. at 152, 154,\\nwe hold that the border search exception authorizes\\nwarrantless searches of a cell phone only to determine\\nwhether the phone contains contraband. A broader search\\ncannot be “justified by the particular purposes served by the\\nexception.” Florida v. Royer, 460 U.S. 491, 500 (1983).\\n\\f28 UNITED STATES V. CANO\\n\\n 2. The Impact of a Limited Scope for Border Searches\\n\\n Our conclusion that the border search exception is\\nrestricted in scope to searches for contraband implicates two\\npractical limitations on warrantless border searches. First,\\nborder officials are limited to searching for contraband only;\\nthey may not search in a manner untethered to the search for\\ncontraband. The Supreme Court has repeatedly emphasized\\nthat “[t]he scope of the search must be ‘strictly tied to and\\njustified by’ the circumstances which rendered its initiation\\npermissible.” Terry v. Ohio, 392 U.S. 1, 19 (1968).\\n\\n The validity of the manual searches conducted by Agents\\nPetonak and Medrano at their inception is beyond dispute.\\nManual searches of a cell phone at the border can be\\nconducted without any suspicion whatsoever, see Cotterman,\\n709 F.3d at 960, and both agents were officers of HSI and\\nthus had authority to conduct border searches, Soto-Soto, 598\\nF.2d at 548–49. As the Supreme Court explained in Terry,\\nhowever, “a search which is reasonable at its inception may\\nviolate the Fourth Amendment by virtue of its intolerable\\nintensity and scope.” 392 U.S. at 18.\\n\\n Once Cano was arrested, Agent Petonak briefly searched\\nCano’s phone and observed that there were no text messages.\\nThe observation that the phone contained no text messages\\nfalls comfortably within the scope of a search for digital\\ncontraband. Child pornography may be sent via text\\nmessage, so the officers acted within the scope of a\\npermissible border search in accessing the phone’s text\\nmessages.\\n\\n Agent Medrano conducted a second manual search of the\\nphone log and text messages on Cano’s phone. Medrano,\\n\\f UNITED STATES V. CANO 29\\n\\nhowever, did more than thumb through the phone consistent\\nwith a search for contraband. He also recorded phone\\nnumbers found in the call log, and he photographed two\\nmessages received after Cano had reached the border. Those\\nactions have no connection whatsoever to digital contraband.\\nCriminals may hide contraband in unexpected places, so it\\nwas reasonable for the two HSI officers to open the phone’s\\ncall log to verify that the log contained a list of phone\\nnumbers and not surreptitious images or videos. But the\\nborder search exception does not justify Agent Medrano’s\\nrecording of the phone numbers and text messages for further\\nprocessing, because that action has no connection to ensuring\\nthat the phone lacks digital contraband. Accordingly, to the\\nextent that Agent Medrano’s search of Cano’s phone went\\nbeyond a verification that the phone lacked digital\\ncontraband, the search exceeded the proper scope of a border\\nsearch and was unreasonable as a border search under the\\nFourth Amendment.11\\n\\n\\n 11\\n The fact of Cano’s arrest does not affect our analysis. The border\\nsearch does not lose its identity as such once Cano was arrested. The\\nUnited States retains a strong interest in preventing contraband from\\nentering the United States, whether it is brought in inadvertently,\\nsmuggled, or admitted into the United States once its owner is arrested.\\nSee United States v. Ickes, 393 F.3d 501, 503–05 (4th Cir. 2005)\\n(upholding the post-arrest search of a laptop computer at the border where\\nthe officials had reason to suspect the computer carried child\\npornography); see also United States v. Bates, 526 F.2d 966, 967–68 (5th\\nCir. 1976) (per curiam) (upholding a search of the defendant’s vehicle\\nafter he had been arrested at the border for violating his bond in\\nconnection with a previous drug crime under both the search incident to\\narrest and the border search exception).\\n\\n The government has not argued that the forensic search of Cano’s\\nphone can be justified as a search incident to lawful arrest. Such an\\nargument is foreclosed by Riley. See Riley, 573 U.S. at 388–91. Nor has\\n\\f30 UNITED STATES V. CANO\\n\\n Second, because the border search exception is limited in\\nscope to searches for contraband, border officials may\\nconduct a forensic cell phone search only when they\\nreasonably suspect that the cell phone contains contraband.\\nWe have held that a “highly intrusive” search—such as a\\nforensic cell phone search—requires some level of\\nparticularized suspicion. Cotterman, 709 F.3d at 963, 968;\\nsee Flores-Montano, 541 U.S. at 152. But that just begs the\\nquestion: Particularized suspicion of what? Contraband? Or\\nevidence of future border-related crimes? Having concluded\\nabove that border searches are limited in scope to searches for\\ncontraband and do not encompass searches for evidence of\\npast or future border-related crimes, we think the answer here\\nis clear: to conduct a more intrusive, forensic cell phone\\nsearch border officials must reasonably suspect that the cell\\nphone to be searched itself contains contraband.\\n\\n Were we to rule otherwise, the government could conduct\\na full forensic search of every electronic device of anyone\\narrested at the border, for the probable cause required to\\njustify an arrest at the border will always satisfy the lesser\\nreasonable suspicion standard needed to justify a forensic\\nsearch. As the Court pointed out in Riley, modern cell phones\\nare “minicomputers” with “immense storage capacity.” 573\\nU.S. at 393. Such phones “carry a cache of sensitive personal\\ninformation”—“[t]he sum of an individual’s private life”—\\nsuch that a search of a cell phone may give the government\\n\\n\\nthe government argued that once Medrano saw the phone numbers in the\\ncall log and the text messages that he could record them consistent with\\nthe plain view exception. See United States v. Comprehensive Drug\\nTesting, 621 F.3d 1162, 1175–77 (9th Cir. 2010) (en banc) (per curiam),\\noverruled in part on other grounds as recognized by Demaree v.\\nPederson, 887 F.3d 870, 876 (9th Cir. 2018) (per curiam).\\n\\f UNITED STATES V. CANO 31\\n\\nnot only “sensitive records previously found in the home,”\\nbut a “broad array of private information never found in a\\nhome in any form—unless the phone is.” Id. at 393–97.\\nWere we to give the government unfettered access to cell\\nphones, we would enable the government to evade the\\nprotections laid out in Riley “on the mere basis that [the\\nsearches] occurred at the border.” Soto-Soto, 598 F.2d at 549.\\n\\n Moreover, in cases such as this, where the individual\\nsuspected of committing the border-related crime has already\\nbeen arrested, there is no reason why border officials cannot\\nobtain a warrant before conducting their forensic search. This\\n“is particularly true in light of ‘advances’ in technology that\\nnow permit ‘the more expeditious processing of warrant\\napplications.’” Birchfield v. North Dakota, 136 S. Ct. 2160,\\n2192 (2016) (quoting Missouri v. McNeely, 569 U.S. 141, 154\\n(2013)); see Riley, 573 U.S. at 401. Indeed, in most cases the\\ntime required to obtain a warrant would seem trivial\\ncompared to the hours, days, and weeks needed to complete\\na forensic electronic search. See, e.g., Wanjiku, 919 F.3d at\\n477 (noting that a forensic “preview” takes one to three\\nhours; the full examination “could take months”); Kolsuz, 890\\nF.3d at 139 (describing how the forensic search “lasted for a\\nfull month, and yielded an 896-page report”); Cotterman, 709\\nF.3d at 959 (describing how the first forensic search was\\nconducted over five days; additional evidence was found\\n“[o]ver the next few months”). We therefore conclude that\\nborder officials may conduct a forensic cell phone search\\nonly when they reasonably suspect that the cell phone to be\\nsearched itself contains contraband.\\n\\n Applied here, if the Cellebrite search of Cano’s cell phone\\nqualifies as a forensic search, the entire search was\\n\\f32 UNITED STATES V. CANO\\n\\nunreasonable under the Fourth Amendment.12 Although\\nAgents Petonak and Medrano had reason to suspect that\\nCano’s phone would contain evidence leading to additional\\ndrugs, the record does not give rise to any objectively\\nreasonable suspicion that the digital data in the phone\\ncontained contraband.13 Absent reasonable suspicion, the\\nborder search exception did not authorize the agents to\\nconduct a warrantless forensic search of Cano’s phone, and\\nevidence obtained through a forensic search should be\\nsuppressed.\\n\\n C. Good Faith Exception\\n\\n We next consider whether the evidence uncovered by the\\nsearches is nevertheless allowed by the good faith exception.\\nHaving held that the manual searches partially violated the\\nFourth Amendment and having held that, if the Cellebrite\\n\\n 12\\n Whether the Cellebrite search constitutes a forensic search is\\ndisputed. Because the district court passed on the issue without deciding\\nit, because neither party has briefed the question to us, and because we are\\nvacating Defendant’s conviction, we decline to reach the merits of the\\nparties’ dispute. See ASSE Int’l, Inc. v. Kerry, 803 F.3d 1059, 1079 (9th\\nCir. 2015).\\n 13\\n Indeed, the detection-of-contraband justification would rarely seem\\nto apply to an electronic search of a cell phone outside the context of child\\npornography. The courts of appeals have just begun to confront the\\ndifficult questions attending cell phone searches at the border. Most of the\\ncases have involved child pornography. See, e.g., Wanjiku, 919 F.3d 472;\\nTouset, 890 F.3d 1227; Molina-Isidoro, 884 F.3d 287; Vergara, 884 F.3d\\n1309; Cotterman, 709 F.3d 952. Among the courts of appeals, only the\\nFourth Circuit has addressed the question outside the context of\\npornography. Kolsuz, 890 F.3d 133 (exportation of firearms parts); see\\nalso United States v. Kim, 103 F. Supp. 3d 32 (D.D.C. 2015) (exports in\\nviolation of Iranian trade embargo); United States v. Saboonchi, 990 F.\\nSupp. 2d 536 (D. Md. 2014) (same).\\n\\f UNITED STATES V. CANO 33\\n\\nsearch of Cano’s phone was a forensic search, it violated the\\nFourth Amendment, we must determine whether the\\nappropriate remedy is suppression of the evidence. The\\nexclusionary rule is “a ‘prudential’ doctrine”; it is “‘not a\\npersonal constitutional right,’ nor is it designed to ‘redress the\\ninjury’ occasioned by an unconstitutional search.” Davis v.\\nUnited States, 564 U.S. 229, 236 (2011) (quoting Stone v.\\nPowell, 428 U.S. 465, 486 (1976)). Because “[e]xclusion\\nexacts a heavy toll on both the judicial system and society at\\nlarge,” we invoke the rule when we are confident that it will\\n“deter future Fourth Amendment violations.” Id. at 236–37.\\nThe exclusionary rule does not deter such violations “when\\nthe police conduct a search in objectively reasonable reliance\\non binding judicial precedent.” Id. at 239. We have said that\\nthe good faith exception applies only to searches where\\n“binding appellate precedent . . . ‘specifically authorizes’ the\\npolice’s search.” United States v. Lara, 815 F.3d 605, 613\\n(9th Cir. 2016) (quoting Davis, 564 U.S. at 232). It is not\\nsufficient for the question to be “unclear” or for the\\ngovernment’s position to be “plausibly . . . permissible.” Id.\\nat 613–14. At the same time, the “precedent [does not have]\\nto constitute a factual match with the circumstances of the\\nsearch in question for the good-faith exception to apply” so\\nas not to “make the good-faith exception a nullity.” United\\nStates v. Lustig, 830 F.3d 1075, 1082 (9th Cir. 2016).\\n\\n The government points to Cotterman as support for the\\ngood faith of the officials. We fail to see how border officials\\ncould believe that Cotterman was “binding appellate\\nprecedent” authorizing their search. Although we have\\nconcluded that Cotterman is still good law after Riley, the\\nofficials could not rely on Cotterman to justify a search for\\nevidence; Cotterman was a search for contraband that the\\ngovernment has a right to seize at the border. Here, the\\n\\f34 UNITED STATES V. CANO\\n\\nofficials’ search was objectively tied only to proving their\\ncase against Cano and finding evidence of future crimes.\\nSearching for evidence and searching for contraband are not\\nthe same thing.\\n\\n We understand that border officials might have thought\\nthat their actions were reasonable, and we recognize that\\nborder officials have to make in-the-moment decisions about\\nhow to conduct their business—whether or not they have\\nwritten guidance from the courts. But as we understand the\\nDavis rule, the good faith exception to the exclusionary rule\\napplies only when the officials have relied on “binding\\nappellate precedent.” See Lara, 815 F.3d at 613; see also\\nWanjiku, 919 F.3d at 485–86 (finding that agents had\\nreasonable suspicion to search the defendant’s cell phone,\\nlaptop, and portable hard drive for child pornography;\\nholding that, if probable cause was required, the officials\\nacted in good faith). This is a rapidly developing area, not an\\narea of settled law. Even if our decision in Cotterman\\nrendered the searches “plausibly . . . permissible,” it did not\\n“specifically authorize” the cell phone searches at issue here.\\nLara, 815 F.3d at 613–14.\\n\\n ***\\n\\n In sum, the manual searches and the Cellebrite search of\\nCano’s cell phone exceeded the scope of a valid border\\nsearch. Because the good faith exception does not apply,\\nmost of the evidence obtained from the searches of Cano’s\\ncell phone should have been suppressed. We thus reverse the\\ndistrict court’s order denying Cano’s motion to suppress, and\\nwe vacate Cano’s conviction. On any retrial, the district court\\nshould determine whether any additional evidence from the\\nwarrantless searches of Cano’s cell phone should be\\n\\f UNITED STATES V. CANO 35\\n\\nsuppressed, either because the Cellebrite search qualifies as\\na forensic search, which the government lacked reasonable\\nsuspicion to conduct, or because the evidence exceeds the\\nproper scope of a border search.\\n\\n III. DISCOVERY ISSUES\\n\\n Cano has also alleged that the government violated his\\nrights under both Brady and Federal Rule of Criminal\\nProcedure 16 when it failed to turn over certain information\\nthat Cano requested from the FBI and DEA. We address\\nCano’s discovery claims, as the issues may be relevant on any\\nretrial.\\n\\n Under Brady, the prosecution has an obligation, imposed\\nby the Due Process Clause, to produce “evidence favorable to\\nan accused upon request . . . where the evidence is material\\neither to guilt or to punishment.” 373 U.S. at 87.\\n“[E]vidence is material only if there is a reasonable\\nprobability that, had the evidence been disclosed to the\\ndefense, the result of the proceeding would have been\\ndifferent.” United States v. Bagley, 473 U.S. 667, 682\\n(1985).14\\n\\n Under Rule 16, the government must, upon request, turn\\nover any documents “within the government’s possession,\\ncustody, or control” that are “material to preparing the\\ndefense.” Fed. R. Crim. P. 16(a)(1)(E)(i). The defendant\\n“must make a threshold showing of materiality, which\\nrequires a presentation of facts which would tend to show that\\nthe Government is in possession of information helpful to the\\n\\n 14\\n We review de novo whether a Brady violation has occurred.\\nUnited States v. Stever, 603 F.3d 747, 752 (9th Cir. 2010).\\n\\f36 UNITED STATES V. CANO\\n\\ndefense.” United States v. Muniz-Jaquez, 718 F.3d 1180,\\n1183–84 (9th Cir. 2013) (quoting United States v. Stever, 603\\nF.3d 747, 752 (9th Cir. 2010)). Because “[i]nformation that\\nis not exculpatory or impeaching may still be relevant to\\ndeveloping a possible defense,” Rule 16 is “broader than\\nBrady.” Id. at 1183.15\\n\\n Under both Brady and Rule 16, the government “has no\\nobligation to produce information which it does not possess\\nor of which it is unaware.” Sanchez v. United States, 50 F.3d\\n1448, 1453 (9th Cir. 1995). It has an obligation to turn over\\nonly material, exculpatory or otherwise helpful to the defense,\\nthat it has in its possession.16 “Possession” is not limited to\\nwhat the prosecutor personally knows. Browning v. Baker,\\n875 F.3d 444, 460 (9th Cir. 2017), cert. denied, 138 S. Ct.\\n2608 (2018); United States v. Bryan, 868 F.2d 1032, 1036\\n(9th Cir. 1989). Because prosecutors are in a “unique\\nposition to obtain information known to other agents of the\\ngovernment,” they have an obligation to “disclos[e] what\\n[they] do[] not know but could have learned.” Carriger v.\\nStewart, 132 F.3d 463, 480 (9th Cir. 1997) (en banc); see also\\nKyles v. Whitley, 514 U.S. 419, 437 (1995) (describing how\\n\\n\\n 15\\n Although discovery rulings are generally reviewed for abuse of\\ndiscretion, Stever, 603 F.3d at 752, we review a district court’s\\ninterpretation of the discovery rules de novo, United States v. Cedano-\\nArellano, 332 F.3d 568, 570–71 (9th Cir. 2003).\\n 16\\n The “possession” element of Brady is treated as coextensive with\\nthat of Rule 16. See, e.g., United States v. Bryan, 868 F.2d 1032, 1037\\n(9th Cir. 1989) (using the same “knowledge and access” test to determine\\n“possession” for both Rule 16 and Brady); United States v. Grace, 401 F.\\nSupp. 2d 1069, 1076 (D. Mont. 2005) (“Whether exculpatory information\\nis in the government’s possession for Brady purposes is measured by the\\nsame . . . test used under Rule 16(a)(1)(E) for discovery.”).\\n\\f UNITED STATES V. CANO 37\\n\\nthe “individual prosecutor has a duty to learn of any favorable\\nevidence known to [those] acting on the government’s\\nbehalf”); Youngblood v. West Virginia, 547 U.S. 867, 869–70\\n(2006) (per curiam). This includes information held by\\nsubordinates such as investigating police officers, see Kyles,\\n514 U.S. at 438; United States v. Price, 566 F.3d 900, 908–09\\n(9th Cir. 2009), and sometimes extends to information held\\nby other executive branch agencies, see United States v.\\nSantiago, 46 F.3d 885, 893 (9th Cir. 1995); United States v.\\nJennings, 960 F.2d 1488, 1490–91 (9th Cir. 1992).\\n\\n Documents held by another executive branch agency are\\ndeemed to be “in the possession of the government” if the\\nprosecutor has “knowledge of and access to” the documents.\\nBryan, 868 F.2d at 1036. Knowledge and access are\\npresumed if the agency participates in the investigation of the\\ndefendant. Id. (“The prosecutor will be deemed to have\\nknowledge of and access to anything in the possession,\\ncustody or control of any federal agency participating in the\\nsame investigation of the defendant.”). However, “a federal\\nprosecutor need not comb the files of every federal agency\\nwhich might have documents regarding the defendant in order\\nto fulfill his or her obligations under [Rule 16].” Id.; see also\\nKyles, 514 U.S. at 437 (“We have never held that the\\nConstitution demands an open file policy . . . .”).\\n\\n Here, Cano asserted a third-party defense theory: he was\\nstaying in Tijuana with his cousin, Jose Medina; Medina was\\na member of the Latin Kings gang which was involved in the\\ndrug trade; and Medina had access to Cano’s car before Cano\\nwas stopped at the border. Cano requested that the U.S.\\nAttorney’s Office turn over any material held by HSI, the\\nFBI, and the DEA relating to: (1) records linking his cousin\\nJose Medina to drug sales, distribution, and trafficking; and\\n\\f38 UNITED STATES V. CANO\\n\\n(2) documentation showing a link between the Latin Kings\\nand drug trafficking through the United States-Mexico\\nborder. The district court found that both requests might\\nproduce evidence that was exculpatory under Brady and\\nmaterial under Rule 16, but limited Cano’s discovery to only\\nmaterial held by HSI. The court concluded that the\\nprosecutor did not have access to evidence held by the FBI\\nand DEA, and thus had no obligation to provide such\\nevidence, because both agencies had “rebuffed” the\\nprosecutor’s attempts to obtain information. Thus, the only\\nissue raised on appeal is whether any material held by the\\nDEA and FBI should be deemed “within the government’s\\npossession.”\\n\\n We find no evidence that the prosecution had knowledge\\nor possession of evidence showing that Medina or the Latin\\nKings were involved in drug trafficking at the Mexico-\\nCalifornia border. Medina had one drug-related conviction,\\nand it was for simple possession of cocaine, not trafficking.\\nBefore trial, however, the prosecution team reached out to\\nMedina and promised him immunity and immigration\\ndocuments in exchange for cooperation and information\\nconcerning drug importation. Although Medina originally\\nrebuffed the government, he eventually offered to work with\\nthe government and “stated that he would be able to assist the\\nGovernment with the . . . biggest RICO . . . case and drug\\nseizures of 20 to 25 kilograms at a time.” The district court\\nfound that Medina’s statements “spawn[ed] an inference that\\n[he] is closely connected to the drug-traffickers in Tijuana.”\\nBased on this inference, Cano argues that the government had\\nsufficient knowledge of a possible connection between\\nMedina and drug trafficking to trigger the government’s\\ndiscovery obligations.\\n\\f UNITED STATES V. CANO 39\\n\\n Cano’s argument, however, misstates the test we first set\\nout in Bryan. Cano has argued only that the prosecutor had\\nknowledge that certain facts might exist. However, we have\\nsaid that the prosecutor’s disclosure obligations turn on “the\\nextent to which the prosecutor has knowledge of and access\\nto the documents sought by the defendant.” Bryan, 868 F.2d\\nat 1036 (emphasis added); see also Santiago, 46 F.3d at 894\\n(analyzing whether the prosecutor had knowledge of and\\naccess to certain inmate files). We have required disclosure\\nonly of documents that the prosecutor knew existed. Bryan,\\n868 F.2d at 1034–37.\\n\\n Here, although Cano has presented evidence alleging a\\nplausible connection between Medina and drug trafficking,\\nCano has failed to adduce any evidence showing that\\nprosecutors or investigators knew that the FBI or the DEA\\npossessed documents showing that connection. In fact, the\\nrecord established the opposite. One of the HSI agents ran\\nMedina’s name through two different law enforcement\\nclearinghouses—in which the FBI and DEA both\\nparticipate—and neither search returned any hits.\\n\\n Moreover, the prosecutor did not have access to FBI or\\nDEA files and thus was under no obligation to “comb the\\nfiles” of the FBI and DEA for documents relating to\\nMedina.17 We have occasionally presumed that a prosecutor\\n\\n 17\\n Cano sought to introduce a 2015 report from the FBI’s National\\nGang Intelligence Center listing the Latin Kings as one of the top gangs\\ninvolved in cross border crime, and including drug importation in its list\\nof cross-border crimes. (The evidence was not ultimately presented at\\ntrial.) Cano also proffered information concerning two government\\ninformants working within the Latin Kings. Although these reports may\\nsuggest that the FBI may have had further information regarding a\\nconnection between the Latin Kings and drug importation, Cano has not\\n\\f40 UNITED STATES V. CANO\\n\\nhas access to an agency’s files where the prosecutor actually\\nobtained inculpatory information from the agency, even if the\\nagency was not involved in the investigation or prosecution.\\nSee Santiago, 46 F.3d at 894 (concluding that the prosecutor\\nhad access to other inmates’ prison files where the prosecutor\\nwas able to obtain the defendant’s prison file from the Bureau\\nof Prisons). Here, however, the U.S. Attorney’s Office\\nadvised the district court that it did not obtain any evidence—\\ninculpatory or exculpatory—from the FBI or the DEA.\\nFollowing the district court’s initial discovery order, HSI’s\\nagent—Agent Petonak—made a formal request to the legal\\ncounsel for the FBI and the DEA for any “materials related to\\nthe Latin Kings importing cocaine from Mexico to the United\\nStates,” but both agencies “declined to provide [him] with\\nany such information.” Neither agency revealed whether any\\nsuch information existed or provided a reason for its refusal.\\nThe U.S. Attorney’s Office also reached out to the FBI and\\nthe DEA for Latin Kings-related discovery. That request was\\nalso denied.\\n\\n Cano argues that the FBI and DEA’s refusal to turn over\\ninformation in this particular case should not be\\ndeterminative and that the test for access under Bryan and\\nSantiago requires only that the U.S. Attorney’s Office or\\ninvestigating agency generally have access to this type of\\ninformation. Cano points to evidence from both prosecution\\nand defense witnesses that HSI regularly works with the FBI\\nand the DEA; that “interagency cooperation has been\\nemphasized” after September 11, 2001; that agents from the\\ndifferent agencies regularly access information for one\\nanother; that a DEA representative worked in Agent\\nPetonak’s office; and that agents are often cross-listed\\n\\n\\nestablished that the prosecutor had access to the FBI’s or the DEA’s files.\\n\\f UNITED STATES V. CANO 41\\n\\nbetween agencies. From this, Cano argues that HSI generally\\nhas access to FBI and DEA files for inculpatory purposes,\\nand thus asserts that the refusal of the FBI and DEA to\\nprovide information in this particular case should not relieve\\nHSI of its discovery obligations. To rule otherwise, Cano\\ncontends, would allow these withholding agencies “to\\neffectively wall off exculpatory information from the\\ngovernment in a particular defendant’s case, all the while\\nproviding the government free-flowing access to information\\nin its overall investigations.”\\n\\n Although we are sympathetic to Cano’s concerns\\nregarding strategic withholding, the rule Cano urges us to\\nadopt is much too broad. Brady and Rule 16 obligations are\\ncase specific. In Bryan we stated that the test for\\n“possession” turns on the prosecutor’s “knowledge of and\\naccess to the documents sought by the defendant in each\\ncase” and that “[t]he prosecutor will be deemed to have\\nknowledge of and access to anything in the possession,\\ncustody or control of any federal agency participating in the\\nsame investigation of the defendant.” 868 F.2d at 1036\\n(emphases added). Such a case-by-case approach makes\\nsense, as the FBI and DEA may have valid concerns over\\nrevealing sensitive information in cases wholly unrelated to\\nthe agencies’ own workload; the agencies may be reluctant to\\ncooperate in a particular investigation if it means opening\\ntheir files in other investigations. If Cano thinks that the FBI\\nor the DEA have other information, not known to the U.S.\\nAttorney’s Office or the investigating officers, he may file a\\nrequest under the Freedom of Information Act, subject to that\\nAct’s own restrictions on releasing “records or information\\ncompiled for law enforcement purposes.” 5 U.S.C.\\n§ 552(b)(7). Brady and Rule 16 are not a means for a\\ndefendant to require the prosecutor to do this work for him.\\n\\f42 UNITED STATES V. CANO\\n\\nSee generally Roth v. U.S. Dep’t of Justice, 642 F.3d 1161,\\n1175–76 (D.C. Cir. 2011); Boyd v. Crim. Div. of U.S. Dep’t\\nof Justice, 475 F.3d 381, 386–89 (D.C. Cir. 2007).\\n\\n Cano is unable to identify any case in which the\\nprosecutor was required to obtain discovery from an agency\\nwholly unrelated to the investigation of the defendant in spite\\nof that agency’s refusal to comply; all of the cases cited by\\nCano imposing a “duty to learn” on the prosecutor involve\\nindependent federal agencies that had participated in the\\ninvestigation of the defendant. See Price, 566 F.3d at\\n908–09; Carriger, 132 F.3d at 479–80; United States v.\\nPerdomo, 929 F.2d 967, 971 (3d Cir. 1991); United States v.\\nOsorio, 929 F.2d 753, 762 (1st Cir. 1991). Indeed, the Third\\nCircuit has held that a Brady obligation is not triggered where\\nthe agency did not participate in the investigation in any way,\\ndid not share any information with the prosecuting team, and\\nwhere the prosecutor had no authority or control over the\\nagency’s members. United States v. Pelullo, 399 F.3d 197,\\n218 (3d Cir. 2005); see also United States v. Salyer, 271\\nF.R.D. 148, 156 (E.D. Cal. 2010) (concluding that “[t]he need\\nfor formal process in the acquisition of documents [from\\nanother agency] is the antithesis of ‘access’”). We similarly\\nnow hold that the prosecutor should not be held to have\\n“access” to any information that an agency not involved in\\nthe investigation or prosecution of the case refuses to turn\\nover.\\n\\n Because the HSI agents and prosecutors in Cano’s case\\nneither knew of nor had access to any additional files relating\\nto Medina and the Latin Kings, we conclude that the\\ngovernment has satisfied its discovery obligations under\\nBrady and Rule 16.\\n\\f UNITED STATES V. CANO 43\\n\\n IV. CONCLUSION\\n\\n We REVERSE the district court’s order denying Cano’s\\nmotion to suppress and VACATE Cano’s conviction.\"}]}"},"judge1":{"kind":"string","value":"SUSAN P GRABER"},"judge2":{"kind":"string","value":"JAY S BYBEE"},"judge3":{"kind":"string","value":"M DOUGLAS HARPOOL"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":1,"string":"1"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649091/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Miguel Cano\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-16\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: Yes\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITED STATES OF AMERICA, No. 17-50151\n Plaintiff-Appellee,\n D.C. No.\n v. 3:16-cr-01770-BTM-1\n\n MIGUEL ANGEL CANO,\n Defendant-Appellant. OPINION\n\n\n Appeal from the United States District Court\n for the Southern District of California\n Barry Ted Moskowitz, District Judge, Presiding\n\n Argued and Submitted April 10, 2019\n Pasadena, California\n\n Filed August 16, 2019\n\nBefore: Susan P. Graber and Jay S. Bybee, Circuit Judges,\n and M. Douglas Harpool,* District Judge.\n\n Opinion by Judge Bybee\n\n\n\n\n *\n The Honorable M. Douglas Harpool, United States District Judge\nfor the Western District of Missouri, sitting by designation\n\f2 UNITED STATES V. CANO\n\n SUMMARY**\n\n\n Criminal Law\n\n The panel reversed the district court’s order denying the\ndefendant’s motion to suppress evidence obtained from\nwarrantless searches of his cell phone by Customs and Border\nProtection officials, and vacated his conviction for importing\ncocaine.\n\n Applying United States v. Cotterman, 709 F.3d 952 (9th\nCir. 2013) (en banc), the panel held that manual cell phone\nsearches may be conducted by border officials without\nreasonable suspicion but that forensic cell phone searches\nrequire reasonable suspicion. The panel clarified Cotterman\nby holding that “reasonable suspicion” in this context means\nthat officials must reasonably suspect that the cell phone\ncontains digital contraband. The panel further concluded that\ncell phone searches at the border, whether manual or forensic,\nmust be limited in scope to whether the phone contains digital\ncontraband; and that a broader search for evidence of a crime\ncannot be justified by the purposes of the border search\nexception to the Fourth Amendment warrant requirement.\n\n The panel held that to the extent that a Border Patrol\nagent’s search of the defendant’s phone – which included the\nrecording of phone numbers and text messages for further\nprocessing – went beyond a verification that the phone lacked\ndigital contraband, the search exceeded the proper scope of a\nborder search and was unreasonable as a border search under\n\n **\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f UNITED STATES V. CANO 3\n\nthe Fourth Amendment. The panel held that although the\nagents had reason to suspect the defendant’s phone would\ncontain evidence leading to additional drugs, the record does\nnot give rise to an objectively reasonable suspicion that the\ndigital data in the phone contained contraband, and the border\nsearch exception therefore did not authorize the agents to\nconduct a warrantless forensic search of the defendant’s\nphone. The panel held that the good faith exception to the\nexclusionary rule does not apply because the border officials\ndid not rely on binding appellate precedent specifically\nauthorizing the cell phone searches at issue here.\n\n Rejecting the defendant’s contention that the government\nviolated his rights under Brady v, Maryland, 373 U.S. 83\n(1963), and Fed. R. Crim. P. 16, by failing to turn over certain\ninformation he requested from the FBI and DEA in pursuit of\nthis third-party defense, the panel found no evidence that the\nprosecution had knowledge or possession of evidence\nshowing that the defendant’s cousin or his cousin’s gang were\ninvolved in drug trafficking at the Mexico-California border,\nand held that the prosecutor should not be held to have\n“access” to any information that an agency not involved in\nthe investigation or prosecution of the case refuses to turn\nover.\n\f4 UNITED STATES V. CANO\n\n COUNSEL\n\nHarini P. Raghupathi (argued), Federal Defenders of San\nDiego, Inc., San Diego, California, for Defendant-Appellant.\n\nMark R. Rehe (argued), Assistant United States Attorney;\nHelen H. Hong, Assistant United States Attorney, Chief,\nAppellate Section, Criminal Division; Adam L. Braverman,\nUnited States Attorney; United States Attorney’s Office, San\nDiego, California; for Plaintiff-Appellee.\n\nSophia Cope and Adam Schwartz, Electronic Frontier\nFoundation, San Francisco, California, for Amicus Curiae\nElectronic Frontier Foundation.\n\n\n OPINION\n\nBYBEE, Circuit Judge:\n\n Defendant-Appellant Miguel Cano was arrested for\ncarrying cocaine as he attempted to cross into the United\nStates from Mexico at the San Ysidro Port of Entry.\nFollowing his arrest, a Customs and Border Protection\nofficial seized Cano’s cell phone and searched it, first\nmanually and then using software that accesses all text\nmessages, contacts, call logs, media, and application data.\nWhen Cano moved to suppress the evidence obtained from\nthe warrantless searches of his cell phone, the district court\nheld that the searches were valid under the border search\nexception to the Fourth Amendment’s warrant requirement.\n\n Applying United States v. Cotterman, 709 F.3d 952 (9th\nCir. 2013) (en banc), we conclude that manual cell phone\n\f UNITED STATES V. CANO 5\n\nsearches may be conducted by border officials without\nreasonable suspicion but that forensic cell phone searches\nrequire reasonable suspicion. We clarify Cotterman by\nholding that “reasonable suspicion” in this context means that\nofficials must reasonably suspect that the cell phone contains\ndigital contraband. We further conclude that cell phone\nsearches at the border, whether manual or forensic, must be\nlimited in scope to a search for digital contraband. In this\ncase, the officials violated the Fourth Amendment when their\nwarrantless searches exceeded the permissible scope of a\nborder search. Accordingly, we hold that most of the\nevidence from the searches of Cano’s cell phone should have\nbeen suppressed. We also conclude that Cano’s Brady claims\nare unpersuasive. Because we vacate Cano’s conviction, we\ndo not reach his claim of prosecutorial misconduct.\n\n We reverse the district court’s order denying Cano’s\nmotion to suppress and vacate Cano’s conviction.\n\n I. THE BACKGROUND\n\nA. The Facts\n\n Defendant-Appellant Miguel Cano worked in the flooring\nand carpet installation trade and lived with his wife and\nchildren in the Mission Hills community north of Los\nAngeles. In the summer of 2016, however, Cano moved from\nLos Angeles to Tijuana, Mexico, where he stayed with his\ncousin Jose Medina. While staying with Medina, Cano\ncrossed the border into the United States six times, sometimes\nremaining in the United States for less than thirty minutes.\nOn two of those trips, Cano was referred to secondary\ninspection, but no contraband was found.\n\f6 UNITED STATES V. CANO\n\n On July 25, 2016, Cano arrived at the San Ysidro Port of\nEntry from Tijuana. In primary inspection, Cano stated that\n“he was living in Mexico, working in San Diego, but going to\nLA on that day.” Pursuant to a random Customs and Border\nProtection (CBP) computer referral, Cano was referred to\nsecondary inspection, where a narcotic-detecting dog alerted\nto the vehicle’s spare tire. A CBP official removed the spare\ntire from the undercarriage of the truck and discovered 14\nvacuum-sealed packages inside, containing 14.03 kilograms\n(30.93 pounds) of cocaine.\n\n Cano was arrested, and a CBP official administratively\nseized his cell phone. The CBP officials called Homeland\nSecurity Investigations (HSI), which dispatched Agents\nPetonak and Medrano to investigate. After arriving, Agent\nPetonak “briefly” and manually reviewed Cano’s cell phone,\nnoticing a “lengthy call log” but no text messages. Agent\nPetonak later stated that the purpose of this manual search\nwas “two-pronged”: “to find some brief investigative leads in\nthe current case,” and “to see if there’s evidence of other\nthings coming across the border.”\n\n Agent Petonak proceeded to question Cano, who waived\nhis Miranda rights and agreed to talk. During that interview,\nCano denied any knowledge of the cocaine. Cano stated that\nhe had moved to Tijuana to look for work in nearby San\nDiego, because work was slow in Los Angeles. He also said\nhe had crossed the border every day for the previous three\nweeks looking for work. He told Agent Petonak that he was\nheaded to a carpet store in Chula Vista that day to seek work.\nWhen pressed, Cano was not able to provide the name or\naddress of the store, claiming that he intended to look it up on\nGoogle after crossing the border. Cano also explained that he\ndid not have his flooring tools with him in his pickup truck so\n\f UNITED STATES V. CANO 7\n\nas to avoid problems with border crossings; Cano intended to\ndrive to Los Angeles to retrieve his tools if he located work\nin San Diego.\n\n During the interrogation, Agent Petonak specifically\nasked Cano about the lack of text messages on his cell phone.\nCano responded that his cousin had advised him to delete his\ntext messages “just in case” he got pulled over in Mexico and\npolice were to check his cell phone. Cano stated that he\nerased his messages to avoid “any problems” with the\nMexican police.\n\n While Agent Petonak questioned Cano, Agent Medrano\nconducted a second manual search of the cell phone. Agent\nMedrano browsed the call log and wrote down some of the\nphone numbers on a piece of paper. He also noticed two\nmessages that arrived after Cano had reached the border, and\nhe took a photograph of the messages. The first message\nstated, “Good morning,” and the second message stated,\n“Primo, are you coming to the house?” Agent Medrano gave\nall of this information—the recorded list of calls and the\nphotograph—to Agent Petonak.\n\n Finally, Agent Medrano conducted a “logical download”\nof the phone using Cellebrite software. A Cellebrite search\nenables the user to access text messages, contacts, call logs,\nmedia, and application data on a cell phone and to select\nwhich types of data to download. It does not, however, allow\nthe user to access data stored within third-party applications.\nAgent Medrano typically does not select the option to\ndownload photographs.\n\n After Agent Petonak interviewed Cano, he reviewed the\nresults of the Cellebrite download of Cano’s phone by Agent\n\f8 UNITED STATES V. CANO\n\nMedrano. The Cellebrite results revealed that Cano had sent\nno text messages, and it listed all the calls made by Cano.\nAgent Petonak later concluded that none of the phone\nnumbers in the call log corresponded to carpeting stores in\nSan Diego.\n\nB. The Proceedings\n\n Cano was indicted for importing cocaine. Before trial,\nCano moved to suppress any evidence obtained from Agents\nPetonak and Medrano’s warrantless searches of his cell phone\nat the border. The district court denied Cano’s motion, ruling\nthat the manual searches and the Cellebrite search of Cano’s\nphone were valid border searches. During trial, the\ngovernment introduced evidence that resulted from the\nmanual searches of the phone and from Agent Medrano’s\nCellebrite download of the phone.1\n\n In preparation for trial, Cano indicated his intent to\npresent a third-party culpability defense claiming that his\ncousin, Jose Medina, was responsible for placing the drugs in\nCano’s spare tire without Cano’s knowledge. Cano proffered\nevidence that Medina had a key to Cano’s car and had driven\nit shortly before Cano’s attempted border crossing, that\n\n\n 1\n Some—but not all—of the evidence was available through\nalternative channels. For example, the government introduced a call log,\nunchallenged by Cano, that the government received from Cano’s phone\ncompany. Similarly, the government later obtained a warrant to search the\nphone, and an agent conducted further searches. Because the government\nintroduced at trial much evidence pre-dating those events, and because the\ngovernment has not argued that any Fourth Amendment error was\nharmless, those later events do not affect our Fourth Amendment analysis\nof the warrantless searches. United States v. Rodriguez, 880 F.3d 1151,\n1163 (9th Cir. 2018)\n\f UNITED STATES V. CANO 9\n\nMedina had a criminal record including a conviction for\ncocaine possession, that Medina was a member of a Chicago-\nbased gang called the Latin Kings, and that the Latin Kings\nsold cocaine within the United States and were involved with\na cartel that trafficked drugs across the border.\n\n Following Cano’s implication of Medina, the government\ncontacted Medina and promised him immunity and\nimmigration papers in exchange for his cooperation. Medina\ninitially denied being involved with drugs, but later contacted\nthe government on his own and offered to help them with the\n“biggest RICO case” and “drug seizures of 20 to 25\nkilograms at a time.” All of this information was made\navailable to Cano.\n\n As part of his defense, Cano sought additional discovery\nfrom HSI, the Federal Bureau of Investigation (FBI), and the\nDrug Enforcement Agency (DEA) regarding: (1) records\nlinking Medina to drug sales, distribution, or trafficking; and\n(2) records linking the Latin Kings to drug trafficking from\nMexico to Southern California. The government opposed\nCano’s discovery motion, arguing that the evidence was not\nmaterial under Federal Rule of Criminal Procedure\n16(a)(1)(E)(i) and that discovery should be limited to HSI, as\nneither the DEA nor the FBI had participated in the\ninvestigation of Cano. The district court originally overruled\nboth objections, finding the evidence material under Rule 16\nand exculpatory under Brady v. Maryland, 373 U.S. 83\n(1963). The court also reasoned that, because HSI could\ninquire of the DEA and FBI if it sought inculpatory evidence,\nHSI had access to the files and was required to provide any\nexculpatory evidence held by the DEA or FBI.\n\f10 UNITED STATES V. CANO\n\n In response to the court’s discovery order, HSI produced\nMedina’s immigration file and his Bureau of Prisons record.\nAgent Petonak also searched for Medina’s name in two\ndifferent police clearinghouses, but neither returned any hits.2\nBoth Agent Petonak and the United States Attorney’s Office\n(USAO) subsequently requested information showing a link\nbetween the Latin Kings and drug trafficking from Mexico\nfrom the legal counsel of both the FBI and DEA. Both\nagencies denied the requests without providing any\nexplanation or any indication as to whether the requested\ninformation existed.\n\n Following these attempts, the government moved for the\ndistrict court to reconsider its discovery order and excuse it\nfrom discovery relating to files held by the FBI and DEA.\nThe district court granted the motion to reconsider, finding\nthat the prosecutor did not have access to the evidence when\nhe was “rebuffed” by agencies over which he had no control.\n\n The case proceeded to trial and Cano presented his third-\nparty culpability defense. The first trial resulted in a hung\njury and a mistrial. On retrial, Cano again relied on his third-\nparty culpability defense. The second trial resulted in Cano’s\nconviction. This appeal followed, in which Cano raises three\nissues: (1) whether the warrantless searches of his cell phone\nviolated the Fourth Amendment and whether the resulting\nevidence should be suppressed; (2) whether the government’s\nnon-disclosure of materials that may have been held by the\nDEA and FBI violated his right to due process under Brady\n\n\n 2\n A police clearinghouse works for the purpose of “deconfliction” by\nnotifying an agency if another agency has an investigation pending against\nthe same person or item. The DEA and FBI participate in the two\nclearinghouses searched by Agent Petonak.\n\f UNITED STATES V. CANO 11\n\nand Federal Rule of Criminal Procedure 16; and (3) whether\nthe government raised an improper propensity inference in its\nclosing argument. We address Cano’s first two arguments in\nturn. Because we conclude that the district court erred in\ndenying Cano’s motion to suppress, we vacate Cano’s\nconviction and do not reach his claim of prosecutorial\nmisconduct.\n\n II. THE WARRANTLESS SEARCH OF CANO’S\n CELL PHONE\n\n The Fourth Amendment protects “[t]he right of the people\nto be secure in their persons, houses, papers, and effects,\nagainst unreasonable searches and seizures.” U.S. Const.\namend. IV.3 Ordinarily, before conducting a search, police\nmust obtain a warrant issued by a judicial officer based “upon\nprobable cause, supported by Oath or affirmation, and\nparticularly describing the place to be searched, and the\npersons or things to be seized.” Id. Warrants are generally\nrequired “unless ‘the exigencies of the situation’ make the\nneeds of law enforcement so compelling that the warrantless\nsearch is objectively reasonable under the Fourth\nAmendment.” Mincey v. Arizona, 437 U.S. 385, 393–94\n(1978) (citation omitted). Consequently, “searches conducted\noutside the judicial process, without prior approval by judge\nor magistrate, are per se unreasonable under the Fourth\nAmendment—subject only to a few specifically established\nand well-delineated exceptions.” Katz v. United States, 389\nU.S. 347, 357 (1967) (footnote omitted). Such “specifically\nestablished and well-delineated exceptions” include exigent\n\n\n 3\n We review de novo “the district court’s determination that [a]\nwarrantless search . . . was a valid border search.” United States v.\nCardona, 769 F.2d 625, 628 (9th Cir. 1985).\n\f12 UNITED STATES V. CANO\n\ncircumstances, searches incident to arrest, vehicle searches,\nand border searches. See Arizona v. Gant, 556 U.S. 332, 343\n(2009) (vehicle searches); Brigham City v. Stuart, 547 U.S.\n398, 403 (2006) (exigent circumstances; listing other\nexceptions, including warrantless entry to fight a fire, to\nprevent the imminent destruction of evidence, or in “hot\npursuit” of a fleeing suspect); United States v. Ramsey, 431\nU.S. 606, 616 (1977) (border searches); Weeks v. United\nStates, 232 U.S. 383, 392 (1914) (searches incident to arrest),\noverruled in part on other grounds by Mapp v. Ohio, 367\nU.S. 643 (1961).\n\n Exceptions to the warrant requirement are subject to two\nimportant constraints. First, any search conducted under an\nexception must be within the scope of the exception. Second,\nsome searches, even when conducted within the scope of the\nexception, are so intrusive that they require additional\njustification, up to and including probable cause and a\nwarrant.\n\n The first constraint is illustrated by the Supreme Court’s\ndecision in Riley v. California, 573 U.S. 373 (2014), a case\ninvolving the search incident to arrest exception. In Riley, the\nCourt addressed “whether the police may, without a warrant,\nsearch digital information on a cell phone seized from an\nindividual who has been arrested”; in other words, whether\ncell phones fell within the scope of the search incident to\narrest exception. Id. at 378. The Court began by recognizing\nthe increasing role in our lives of “minicomputers that also\nhappen to have the capacity to be used as a telephone”;\n“[m]odern cell phones, as a category, implicate privacy\nconcerns far beyond those implicated by the search of a\ncigarette pack, a wallet, or a purse.” Id. at 393.\nAcknowledging that “it has been well accepted that [a search\n\f UNITED STATES V. CANO 13\n\nincident to lawful arrest] constitutes an exception to the\nwarrant requirement,” id. at 382, the Court pointed out that\nsuch searches serve two purposes: (1) to secure “the officer’s\nsafety” and (2) to “prevent . . . concealment or destruction [of\nevidence],” id. at 383 (citation omitted). The Court then\nconsidered whether a cell phone search qualified as a search\nincident to arrest by considering “whether application of the\nsearch incident to arrest doctrine to [cell phones] would\n‘untether the rule from the justifications underlying the . . .\nexception.’” Id. at 386 (quoting Gant, 556 U.S. at 343).\n\n The Court concluded that neither purpose for the search\nincident to arrest exception justified the search of a cell\nphone. The Court rejected the government’s argument that\nsearching a cell phone incident to arrest would “help ensure\nofficer safety in . . . indirect ways, for example by alerting\nofficers that confederates of the arrestee are headed to the\nscene.” Id. at 387. The Court reasoned that the government’s\nposition “would . . . represent a broadening” of the\nexception’s foundational concern that “an arrestee himself\nmight grab a weapon and use it against an officer.” Id. at\n387–88. The Court observed that “once law enforcement\nofficers have secured a cell phone, there is no longer any risk\nthat the arrestee himself will be able to delete incriminating\ndata from the phone,” id. at 388, and police have means to\nensure that data cannot be wiped from the phone remotely,\nid. at 390. The Court concluded “not that the information on\na cell phone is immune from search; [but rather] that a\nwarrant is generally required before such a search, even when\na cell phone is seized incident to arrest.” Id. at 401.\n\n The second constraint on warrantless searches is\nillustrated by the Court’s decision in United States v.\nMontoya de Hernandez, 473 U.S. 531 (1985). Montoya was\n\f14 UNITED STATES V. CANO\n\nstopped at Los Angeles International Airport and referred to\nsecondary inspection. Id. at 533. She had arrived from\nBogota and was carrying $5,000 in cash. Id. She had no\ncredit cards and no hotel reservations. Id. at 533–34.\nBecause border officials suspected that Montoya may have\nswallowed cocaine-filled balloons, Montoya was held in the\ncustoms office and, after a magistrate judge issued an order,\ntaken to a hospital for a rectal examination. Id. at 534–35.\nOver the next four days, she passed 88 balloons containing\ncocaine. Id. at 536. Montoya argued that the search she was\nsubjected to, though a border search, was so intrusive that it\ncould not be conducted without a high level of particularized\nsuspicion. Id. at 536–37, 540. The Court balanced her\nprivacy interests against the interests of the government at the\nborder and concluded that, while routine searches may be\nconducted at the border without any showing of suspicion, a\nmore intrusive, nonroutine search must be supported by\n“reasonable suspicion.” Id. at 537–41; see also United States\nv. Flores-Montano, 541 U.S. 149, 152 (2004) (suggesting that\nnonroutine searches are limited to “highly intrusive searches\nof the person” involving “dignity and privacy interests”).\n\n Cano recognizes that he was subject to search at the\nborder, but Cano and amicus Electronic Frontier Foundation\n(“EFF”) raise two categorical challenges and one as-applied\nchallenge to the searches conducted here. First, EFF argues\nthat any warrantless search of a cell phone falls outside the\nscope of the border search exception. Second, EFF argues\nthat even if the search is within the scope of the border search\nexception, a warrantless cell phone search is so intrusive that\nit requires probable cause. We address these categorical\nchallenges in Part II.A. Third, Cano asserts that, even if cell\nphones are generally subject to search at the border, the\nmanual and forensic searches of his cell phone exceeded the\n\f UNITED STATES V. CANO 15\n\n“well delineated” scope of the border search. We address this\nas-applied question in Part II.B. Finally, the government\nargues that even if the border search exceeded the limits of\nthe Fourth Amendment, the search was conducted in good\nfaith, and the evidence is admissible. We consider the good\nfaith exception in Part II.C.\n\nA. Border Searches and Cell Phones\n\n “[B]order searches constitute a ‘historically recognized\nexception to the Fourth Amendment’s general principle that\na warrant be obtained.’” Cotterman, 709 F.3d at 957 (quoting\nRamsey, 431 U.S. at 621). Indeed, border searches typically\ndo not require any particularized suspicion, so long as they\nare “routine inspections and searches of individuals or\nconveyances seeking to cross our borders.” Almeida-Sanchez\nv. United States, 413 U.S. 266, 272 (1973); see United States\nv. Seljan, 547 F.3d 993, 999 (9th Cir. 2008) (en banc). Such\nsearches are “reasonable simply by virtue of the fact they\noccur at the border.” Ramsey, 431 U.S. at 616. The\nexception is “rooted in ‘the long-standing right of the\nsovereign to protect itself by stopping and examining persons\nand property crossing into this country,’” Cotterman, 709\nF.3d at 960 (quoting Ramsey, 431 U.S. at 616), to “prevent[]\nthe entry of unwanted persons and effects,” id. (quoting\nFlores-Montano, 541 U.S. at 152).\n\n The sovereign’s right to conduct suspicionless searches at\nthe border “does not mean, however, that at the border\n‘anything goes.’” Id. (quoting Seljan, 547 F.3d at 1000).\nRather, the border search exception is a “narrow exception”\nthat is limited in two important ways. Id. (citation omitted).\nFirst, “[t]he authorizing statute limits the persons who may\nlegally conduct a ‘border search’ to ‘persons authorized to\n\f16 UNITED STATES V. CANO\n\nboard or search vessels.’” United States v. Soto-Soto, 598\nF.2d 545, 549 (9th Cir. 1979) (citing 19 U.S.C. § 482).4 This\nincludes customs and immigration officials, but not general\nlaw enforcement officers such as FBI agents. Id.; see United\nStates v. Diamond, 471 F.2d 771, 773 (9th Cir. 1973) (stating\nthat “customs agents are not general guardians of the public\npeace”). Second, a border search must be conducted “in\nenforcement of customs laws.” Soto-Soto, 598 F.2d at 549.\nA border search must be conducted to “enforce importation\nlaws,” and not for “general law enforcement purposes.” Id.\n\n\n 4\n Section 482 now reads in relevant part:\n\n Any of the officers or persons authorized to board\n or search vessels may stop, search, and examine . . . any\n vehicle, beast, or person, on which or whom he or they\n shall suspect there is merchandise which is subject to\n duty, or shall have been introduced into the United\n States in any manner contrary to law . . . . [and may]\n seize and secure the same for trial.\n\n19 U.S.C. § 482(a); see id. § 1467 (“[T]he appropriate customs officer for\n[a] port or place of arrival may . . . enforce, cause inspection, examination,\nand search to be made of the persons, baggage, and merchandise\ndischarged or unladen from [an arriving] vessel . . . .”); id. § 1496 (“The\nappropriate customs officer may cause an examination to be made of the\nbaggage of any persons arriving in the United States in order to ascertain\nwhat articles are contained therein and whether subject to duty, free of\nduty, or prohibited . . . .”); id. § 1582 (“[A]ll persons coming into the\nUnited States from foreign countries shall be liable to detention and search\nby authorized officers or agents . . . .”).\n\n The Court has described § 482 as granting the executive “plenary\nauthority to conduct routine searches and seizures at the border, without\nprobable cause or a warrant.” Montoya de Hernandez, 473 U.S. at 537.\nWe have held that the “outer limits of authority delegated by [§ 482 are]\navailable only in border searches.” Corngold v. United States, 367 F.2d\n1, 3 (9th Cir. 1966) (en banc).\n\f UNITED STATES V. CANO 17\n\nA general search cannot be “justif[ied] . . . on the mere basis\nthat it occurred at the border.” Id. (affirming the suppression\nof evidence where an FBI agent stopped and searched the\nvehicle of an alien to determine whether the car had been\nstolen).\n\n 1. Cell Phone Data as Contraband\n\n As we discussed briefly above, the Supreme Court has\nidentified two principal purposes behind warrantless border\nsearches: First, to identify “[t]ravellers . . . entitled to come\nin” and, second, to verify their “belongings as effects which\nmay be lawfully brought in.” Carroll v. United States, 267\nU.S. 132, 154 (1925); see Ramsey, 431 U.S. at 620 (“The\nborder-search exception is grounded in the recognized right\nof the sovereign to control . . . who and what may enter the\ncountry.”).\n\n EFF argues that applying the border search exception to\na cell phone’s data would “untether” the exception from the\npurposes underlying it. EFF contends that a border search\nencompasses only a search for illegal persons and physical\ncontraband located on the body of the applicant for\nadmission or among his effects. Because digital data on a cell\nphone cannot conceal objects such as drugs, guns, or\nsmuggled persons, EFF asserts that digital cell phone\nsearches are always beyond the scope of the border search\nexception.\n\n We agree with EFF that the purpose of the border search\nis to interdict contraband, but we disagree with its premise\nthat cell phones cannot contain contraband. Although cell\n\f18 UNITED STATES V. CANO\n\nphone data cannot hide physical objects,5 the data can contain\ndigital contraband. The best example is child pornography.\nSee United States v. Molina-Isidoro, 884 F.3d 287, 295 n.3\n(5th Cir. 2018) (Costa, J., specially concurring) (“One type of\ncontraband that can be stored within the data of a cell\nphone . . . is child pornography.”). And because cell phones\nmay ultimately be released into the interior, even if the owner\nhas been detained, the United States has a strong interest in\npreventing the entry of such material. See, e.g., United States\nv. Vergara, 884 F.3d 1309, 1311 (11th Cir.) (describing how\nagents returned one of the defendant’s phones to a family\nmember after defendant had been arrested for possessing\nchild pornography on his other two phones), cert. denied, 139\nS. Ct. 70 (2018). We find no basis for the proposition that the\nborder search exception is limited to searching for physical\ncontraband. At the very least, a cell phone that has photos\nstored on it is the equivalent of photographs, magazines, and\nbooks.6 See Riley, 573 U.S. at 394; Cotterman, 709 F.3d at\n964. The contents may be digital when they are on the phone,\nbut the physicality of the phone itself and the possibility that\n\n\n 5\n No one contests that a border official could, consistent with the\nFourth Amendment, examine the physical body of a cell phone to see if\nthe phone itself is contraband—because, for example, it is a pirated copy\nof a patented U.S. phone—or if the phone itself presents a physical threat\nto officers. See Riley, 573 U.S. at 387 (“Law enforcement officers remain\nfree to examine the physical aspects of a phone to ensure that it will not\nbe used as a weapon—say, to determine whether there is a razor blade\nhidden between the phone and its case.”). The dispute here concerns only\nwhether border officials may search the digital data contained within the\nphone.\n 6\n We need not address here questions surrounding the use of “cloud\ncomputing,” where the phone gives access to, but does not contain in its\nown memory, digital data stored in the cloud. See Riley, 573 U.S. at\n397–98; Cotterman, 709 F.3d at 965 & n.12.\n\f UNITED STATES V. CANO 19\n\nthe phone’s contents can be printed or shared electronically\ngives border officials sufficient reason to inspect it at the\nborder. We conclude that cell phones—including the phones’\ndata—are subject to search at the border.\n\n 2. Forensic Cell Phone Searches as an Intrusive Search\n\n The second question we must address in response to\namicus EFF is whether forensic searches of a cell phone are\nso intrusive that they require reasonable suspicion or even\nprobable cause. We answered this question in our en banc\ndecision in Cotterman, but with respect to laptop computers.7\nCotterman, 709 F.3d at 962–68. Cotterman was a United\nStates citizen returning to the United States from Mexico. Id.\nat 957. When he reached the port of entry, border officials\nnoted that Cotterman had various convictions for sexual\nconduct with children. Id. Concerned that Cotterman might\nbe involved in child sex tourism, officials conducted a brief\nsearch of his laptop computers and digital cameras and noted\nthat the laptops had password-protected files. Id. at 958. The\nofficials detained the computers for several days in order to\nrun a comprehensive forensic search of the hard drive, which\nrevealed hundreds of images of child pornography. Id. at\n958–59. For us, “the legitimacy of the initial search of\nCotterman’s electronic devices at the border [was] not in\ndoubt,” id. at 960, “[t]he difficult question . . . [was] the\nreasonableness, without a warrant, of the forensic\n\n\n\n 7\n Although Cotterman referred to “electronic devices” generally, see\n709 F.3d at 962–68, our holding was limited to the “examination of\nCotterman’s computer,” id. at 968, and did not address cell phones. We\nmentioned cell phones only once—in the first paragraph of the\nintroduction describing the modern “digital world.” Id. at 956.\n\f20 UNITED STATES V. CANO\n\nexamination that comprehensively analyzed the hard drive of\nthe computer,” id. at 961.\n\n We acknowledged the “substantial personal privacy\ninterests” in “[e]lectronic devices . . . capable of storing\nwarehouses full of information.” Id. at 964. At the same\ntime, we recognized “the important security concerns that\nprevail at the border” and the legitimacy of “[t]he effort to\ninterdict child pornography.” Id. at 966. We held that a\nroutine, manual search of files on a laptop computer—“a\nquick look and unintrusive search”—is reasonable “even\nwithout particularized suspicion,” but that officials must\n“possess a particularized and objective basis for suspecting\nthe person stopped of criminal activity” to engage in a\nforensic examination, which is “essentially a computer strip\nsearch.” Id. at 960–61, 966, 967 (citation omitted). We\nconcluded that reasonable suspicion was “a modest, workable\nstandard that is already applied in the extended border search,\nTerry stop, and other contexts.” Id. at 966; see id. at 968\n(defining reasonable suspicion as “a particularized and\nobjective basis for suspecting the particular person stopped of\ncriminal activity” (quoting United States v. Cortez, 449 U.S.\n411, 417–18 (1981))).\n\n We think that Cotterman’s reasoning applies equally to\ncell phones. In large measure, we anticipated the Supreme\nCourt’s reasoning in Riley, 573 U.S. at 393–97, when we\nrecognized in Cotterman that digital devices “contain the\nmost intimate details of our lives” and “the uniquely sensitive\nnature of data on electronic devices carries with it a\nsignificant expectation of privacy,” Cotterman, 709 F.3d at\n965–66; see Riley, 573 U.S. at 385, 393 (describing cell\nphones as “a pervasive and insistent part of daily life” that,\n“as a category, implicate privacy concerns far beyond those\n\f UNITED STATES V. CANO 21\n\nimplicated by the search of a cigarette pack, a wallet, or a\npurse”). The Court’s view of cell phones in Riley so closely\nresembles our own analysis of laptop computers in Cotterman\nthat we find no basis to distinguish a forensic cell phone\nsearch from a forensic laptop search.8\n\n Nor do we believe that Riley renders the Cotterman\nstandard insufficiently protective. Riley, of course, held that\n“a warrant is generally required” before searching a cell\nphone, “even when a cell phone is seized incident to arrest.”\n573 U.S. at 401. But here we deal with the border search\nexception—not the search incident to arrest exception—and\nthe difference in context is critical. In light of the\ngovernment’s enhanced interest in protecting the “integrity of\nthe border” and the individual’s decreased expectation of\nprivacy, the Court has emphasized that “the Fourth\nAmendment’s balance of reasonableness is qualitatively\ndifferent at the international border than in the interior” and\nis “struck much more favorably to the Government.”\nMontoya de Hernandez, 473 U.S. at 538–40. As a result,\npost-Riley, no court has required more than reasonable\nsuspicion to justify even an intrusive border search. See\nUnited States v. Wanjiku, 919 F.3d 472, 485 (7th Cir. 2019)\n(“[N]o circuit court, before or after Riley, has required more\nthan reasonable suspicion for a border search of cell phones\nor electronically-stored data.”); Touset, 890 F.3d at 1234\n\n 8\n We note that the Eleventh Circuit disagreed with Cotterman in\nUnited States v. Touset, 890 F.3d 1227, 1234 (11th Cir. 2018). The court\nheld that no level of suspicion was required to conduct a forensic search\nof a cell phone. Id. at 1234–35. Nevertheless, the Touset court held, in\nthe alternative, that the forensic search of various electronic devices seized\nat the border were supported by reasonable suspicion. Id. at 1237. As\nwith most cell phone search cases, in Touset border agents were looking\nfor child pornography.\n\f22 UNITED STATES V. CANO\n\n(“Riley, which involved the search-incident-to-arrest\nexception, does not apply to searches at the border.”);\nMolina-Isidoro, 884 F.3d at 291 (“For border searches both\nroutine and not, no case has required a warrant.”); id. at 293\n(“The bottom line is that only two of the many federal cases\naddressing border searches of electronic devices have ever\nrequired any level of suspicion. They both required only\nreasonable suspicion and that was for the more intrusive\nforensic search.”); see also Kolsuz, 890 F.3d 133, 137 (4th\nCir. 2018) (concluding that a “forensic examination of\nKolsuz’s phone must be considered a nonroutine border\nsearch, requiring some measure of individualized suspicion”\nbut declining to decide whether the standard should be\nreasonable suspicion or probable cause).\n\n Accordingly, we hold that manual searches of cell phones\nat the border are reasonable without individualized suspicion,\nwhereas the forensic examination of a cell phone requires a\nshowing of reasonable suspicion. See Cotterman, 709 F.3d\nat 968.\n\nB. The Searches of Cano’s Cell Phone and the Scope of the\n Border Search Exception\n\n Having concluded that border officials may conduct\nsuspicionless manual searches of cell phones, but must have\nreasonable suspicion before they conduct a forensic search,\nwe still must address the core of Cano’s argument: whether\nthe manual and forensic searches of his cell phone were not\nsearches for digital contraband, but searches for evidence of\na crime, and thus exceeded the proper scope of a border\nsearch.\n\f UNITED STATES V. CANO 23\n\n 1. The Border Exception and the Search for Contraband\n\n As a threshold matter, Cano argues that border searches\nare limited in both purpose and scope to searches for\ncontraband.9 In response, the government argues that\nsearches for evidence that would aid in prosecuting past and\npreventing future border-related crimes are tethered to the\npurpose of the border search exception—namely, interdicting\nforeign contraband—and thus fall within its scope.\n\n\n\n\n 9\n Cano emphasizes that the officials who arrested him were looking\nfor evidence of a crime, not contraband that could be seized at the border,\nand this renders the search unconstitutional. He points to Officers Petonak\nand Medrano, who searched Cano’s cell phone, and who testified that their\nsearches had a dual purpose: “to find some brief investigative leads in the\ncurrent case” and “to see if there[] [was] evidence of other things coming\nacross the border.” Because the agents acknowledged that they sought\nevidence to use against Cano in building a criminal case, Cano argues that\nthe court should treat the search as one conducted for “general law\nenforcement purposes” rather than a border search.\n\n Cano’s focus on the officials’ subjective motivations is misplaced,\nhowever. As the district court recognized, “courts have repeatedly held\nthat the Fourth Amendment’s reasonableness analysis is ‘predominantly\nan objective inquiry.’” See Whren v. United States, 517 U.S. 806, 813\n(1996) (upholding a “pretextual” stop because “[s]ubjective intentions\nplay no role in ordinary . . . Fourth Amendment analysis”). We have\nupheld border searches of persons seeking entry even when those searches\nwere conducted “at the behest” of DEA agents seeking criminal evidence.\nSee United States v. Schoor, 597 F.2d 1303, 1305–06 (9th Cir. 1979)\n(holding a border search reasonable where it was conducted “at the\nbehest” of DEA agents and included a search for certain items of evidence\nin addition to a search for contraband). Thus, the mere fact that Officers\nPetonak and Medrano subjectively hoped to find “investigative leads”\npertaining to the seized shipment of cocaine does not render their searches\nof Cano’s phone beyond the border search exception.\n\f24 UNITED STATES V. CANO\n\n This is a close question, but we think Cano has the better\nof the argument. There is a difference between a search for\ncontraband and a search for evidence of border-related\ncrimes, although the distinction may not be apparent.\nCotterman helps us focus on the difference. There, border\nofficials had been alerted that Cotterman had a criminal\nrecord of sex abuse of minors and might be involved in “child\nsex tourism.” Cotterman, 709 F.3d at 957. The officials\nseized his laptop and subjected it to searches for child\npornography, which they found. In Cotterman, the child\npornography was contraband subject to seizure at the border.\nAs contraband, the child pornography is also evidence of\nvarious crimes, including possession of child pornography,\n18 U.S.C. § 2252A(a)(5)(B), and importation of obscene\nmaterial, 18 U.S.C. § 1462(a). But nothing in Cotterman\nauthorized border officials to conduct a search for evidence\nthat Cotterman was involved in sex-related crimes generally.\n\n Border officials are authorized to seize “merchandise\nwhich . . . shall have been introduced into the United States\nin any manner contrary to law.” 19 U.S.C. § 482(a)\n(emphasis added). The photos on Cotterman’s laptop\ncomputer were such merchandise. 18 U.S.C. § 2252(a). But\nborder officials have no general authority to search for crime.\nThis is true even if there is a possibility that such crimes may\nbe perpetrated at the border in the future. So, for example, if\nU.S. officials reasonably suspect that a person who has\npresented himself at the border may be engaged in price\nfixing, see 15 U.S.C. § 1, they may not conduct a forensic\nsearch of his phone or laptop. Evidence of price fixing—\ntexts or emails, for example—is not itself contraband whose\nimportation is prohibited by law. Such emails may be\nevidence of a crime, but they are not contraband, and there is\nno law prohibiting the importation of mere evidence of crime.\n\f UNITED STATES V. CANO 25\n\n We recognize that our analysis is in tension with the\nFourth Circuit’s decision in Kolsuz. Kolsuz was detained at\nWashington Dulles International Airport when customs\nagents discovered firearm parts in his luggage. Kolsuz, 890\nF.3d at 138–39. Kolsuz was arrested and his cell phone\nseized. Id. at 139. The agents subjected the phone to a\nmonth-long forensic search, producing a 896-page report. Id.\nKolsuz challenged the search, which the district court upheld\nand the Fourth Circuit affirmed. Id. at 139–42. The court\napproved the forensic search because the agents had “reason\nto believe . . . that Kolsuz was attempting to export firearms\nillegally” and that “their search would reveal not only\nevidence of the export violation they already had detected,\nbut also ‘information related to other ongoing attempts to\nexport illegally various firearm parts.’” Id. at 143 (quoting\nthe district court; citation omitted). According to the Fourth\nCircuit, “[t]he justification behind the border search\nexception is broad enough to accommodate not only the\ndirect interception of contraband as it crosses the border, but\nalso the prevention and disruption of ongoing efforts to\nexport contraband illegally.” Id. (emphasis added).10\n\n We agree with much of the Fourth Circuit’s discussion of\nfoundational principles, but we respectfully disagree with the\nfinal step approving the search for further evidence that\n\n\n 10\n As support for this proposition, the Fourth Circuit cited two district\ncourt cases originating within our circuit. Both of those cases addressed\nfact-patterns almost identical to Cano’s, and in each case the district court\nheld that the border-search exception was not limited to searching for\ncontraband directly. See United States v. Mendez, 240 F. Supp. 3d 1005,\n1007–08 (D. Ariz. 2017); United States v. Ramos, 190 F. Supp. 3d 992,\n999 (S.D. Cal. 2016). In neither case was the issue appealed to our circuit.\nThus, Cano’s case presents the first opportunity for us to consider the\nmatter.\n\f26 UNITED STATES V. CANO\n\nKolsuz was smuggling weapons. Our disagreement focuses\nprecisely on the critical question that we previously\nidentified: Does the proper scope of a border search include\nthe power to search for evidence of contraband that is not\npresent at the border? Or, put differently, can border agents\nconduct a warrantless search for evidence of past or future\nborder-related crimes? We think that the answer must be\n“no.” The “[d]etection of . . . contraband is the strongest\nhistoric rationale for the border-search exception.” Molina-\nIsidoro, 884 F.3d at 295 (Costa, J., specially concurring).\nIndeed, “every border-search case the Supreme Court has\ndecided involved searches to locate items being smuggled”\nrather than evidence. Id. (emphasis added); see Montoya de\nHernandez, 473 U.S. at 537 (the border search is “to prevent\nthe introduction of contraband into this country”); United\nStates v. 12 200-Foot Reels of Super 8mm. Film, 413 U.S.\n123, 125 (1973) (border searches are “necessary to prevent\nsmuggling and to prevent prohibited articles from entry”);\nUnited States v. Thirty-Seven Photographs, 402 U.S. 363, 376\n(1971) (“Customs officers characteristically inspect luggage\nand their power to do so is not questioned in this case; it is an\nold practice and is intimately associated with excluding\nillegal articles from the country”). In fact, the Court has long\n“draw[n] a sharp distinction between searches for contraband\nand those for evidence that may reveal the importation of\ncontraband.” Molina-Isidoro, 884 F.3d at 296 (Costa, J.,\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Miguel Cano\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-16\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: Yes\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n UNITED STATES OF AMERICA, No. 17-50151\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:16-cr-01770-BTM-1\\n\\n MIGUEL ANGEL CANO,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the Southern District of California\\n Barry Ted Moskowitz, District Judge, Presiding\\n\\n Argued and Submitted April 10, 2019\\n Pasadena, California\\n\\n Filed August 16, 2019\\n\\nBefore: Susan P. Graber and Jay S. Bybee, Circuit Judges,\\n and M. Douglas Harpool,* District Judge.\\n\\n Opinion by Judge Bybee\\n\\n\\n\\n\\n *\\n The Honorable M. Douglas Harpool, United States District Judge\\nfor the Western District of Missouri, sitting by designation\\n\\f2 UNITED STATES V. CANO\\n\\n SUMMARY**\\n\\n\\n Criminal Law\\n\\n The panel reversed the district court’s order denying the\\ndefendant’s motion to suppress evidence obtained from\\nwarrantless searches of his cell phone by Customs and Border\\nProtection officials, and vacated his conviction for importing\\ncocaine.\\n\\n Applying United States v. Cotterman, 709 F.3d 952 (9th\\nCir. 2013) (en banc), the panel held that manual cell phone\\nsearches may be conducted by border officials without\\nreasonable suspicion but that forensic cell phone searches\\nrequire reasonable suspicion. The panel clarified Cotterman\\nby holding that “reasonable suspicion” in this context means\\nthat officials must reasonably suspect that the cell phone\\ncontains digital contraband. The panel further concluded that\\ncell phone searches at the border, whether manual or forensic,\\nmust be limited in scope to whether the phone contains digital\\ncontraband; and that a broader search for evidence of a crime\\ncannot be justified by the purposes of the border search\\nexception to the Fourth Amendment warrant requirement.\\n\\n The panel held that to the extent that a Border Patrol\\nagent’s search of the defendant’s phone – which included the\\nrecording of phone numbers and text messages for further\\nprocessing – went beyond a verification that the phone lacked\\ndigital contraband, the search exceeded the proper scope of a\\nborder search and was unreasonable as a border search under\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. CANO 3\\n\\nthe Fourth Amendment. The panel held that although the\\nagents had reason to suspect the defendant’s phone would\\ncontain evidence leading to additional drugs, the record does\\nnot give rise to an objectively reasonable suspicion that the\\ndigital data in the phone contained contraband, and the border\\nsearch exception therefore did not authorize the agents to\\nconduct a warrantless forensic search of the defendant’s\\nphone. The panel held that the good faith exception to the\\nexclusionary rule does not apply because the border officials\\ndid not rely on binding appellate precedent specifically\\nauthorizing the cell phone searches at issue here.\\n\\n Rejecting the defendant’s contention that the government\\nviolated his rights under Brady v, Maryland, 373 U.S. 83\\n(1963), and Fed. R. Crim. P. 16, by failing to turn over certain\\ninformation he requested from the FBI and DEA in pursuit of\\nthis third-party defense, the panel found no evidence that the\\nprosecution had knowledge or possession of evidence\\nshowing that the defendant’s cousin or his cousin’s gang were\\ninvolved in drug trafficking at the Mexico-California border,\\nand held that the prosecutor should not be held to have\\n“access” to any information that an agency not involved in\\nthe investigation or prosecution of the case refuses to turn\\nover.\\n\\f4 UNITED STATES V. CANO\\n\\n COUNSEL\\n\\nHarini P. Raghupathi (argued), Federal Defenders of San\\nDiego, Inc., San Diego, California, for Defendant-Appellant.\\n\\nMark R. Rehe (argued), Assistant United States Attorney;\\nHelen H. Hong, Assistant United States Attorney, Chief,\\nAppellate Section, Criminal Division; Adam L. Braverman,\\nUnited States Attorney; United States Attorney’s Office, San\\nDiego, California; for Plaintiff-Appellee.\\n\\nSophia Cope and Adam Schwartz, Electronic Frontier\\nFoundation, San Francisco, California, for Amicus Curiae\\nElectronic Frontier Foundation.\\n\\n\\n OPINION\\n\\nBYBEE, Circuit Judge:\\n\\n Defendant-Appellant Miguel Cano was arrested for\\ncarrying cocaine as he attempted to cross into the United\\nStates from Mexico at the San Ysidro Port of Entry.\\nFollowing his arrest, a Customs and Border Protection\\nofficial seized Cano’s cell phone and searched it, first\\nmanually and then using software that accesses all text\\nmessages, contacts, call logs, media, and application data.\\nWhen Cano moved to suppress the evidence obtained from\\nthe warrantless searches of his cell phone, the district court\\nheld that the searches were valid under the border search\\nexception to the Fourth Amendment’s warrant requirement.\\n\\n Applying United States v. Cotterman, 709 F.3d 952 (9th\\nCir. 2013) (en banc), we conclude that manual cell phone\\n\\f UNITED STATES V. CANO 5\\n\\nsearches may be conducted by border officials without\\nreasonable suspicion but that forensic cell phone searches\\nrequire reasonable suspicion. We clarify Cotterman by\\nholding that “reasonable suspicion” in this context means that\\nofficials must reasonably suspect that the cell phone contains\\ndigital contraband. We further conclude that cell phone\\nsearches at the border, whether manual or forensic, must be\\nlimited in scope to a search for digital contraband. In this\\ncase, the officials violated the Fourth Amendment when their\\nwarrantless searches exceeded the permissible scope of a\\nborder search. Accordingly, we hold that most of the\\nevidence from the searches of Cano’s cell phone should have\\nbeen suppressed. We also conclude that Cano’s Brady claims\\nare unpersuasive. Because we vacate Cano’s conviction, we\\ndo not reach his claim of prosecutorial misconduct.\\n\\n We reverse the district court’s order denying Cano’s\\nmotion to suppress and vacate Cano’s conviction.\\n\\n I. THE BACKGROUND\\n\\nA. The Facts\\n\\n Defendant-Appellant Miguel Cano worked in the flooring\\nand carpet installation trade and lived with his wife and\\nchildren in the Mission Hills community north of Los\\nAngeles. In the summer of 2016, however, Cano moved from\\nLos Angeles to Tijuana, Mexico, where he stayed with his\\ncousin Jose Medina. While staying with Medina, Cano\\ncrossed the border into the United States six times, sometimes\\nremaining in the United States for less than thirty minutes.\\nOn two of those trips, Cano was referred to secondary\\ninspection, but no contraband was found.\\n\\f6 UNITED STATES V. CANO\\n\\n On July 25, 2016, Cano arrived at the San Ysidro Port of\\nEntry from Tijuana. In primary inspection, Cano stated that\\n“he was living in Mexico, working in San Diego, but going to\\nLA on that day.” Pursuant to a random Customs and Border\\nProtection (CBP) computer referral, Cano was referred to\\nsecondary inspection, where a narcotic-detecting dog alerted\\nto the vehicle’s spare tire. A CBP official removed the spare\\ntire from the undercarriage of the truck and discovered 14\\nvacuum-sealed packages inside, containing 14.03 kilograms\\n(30.93 pounds) of cocaine.\\n\\n Cano was arrested, and a CBP official administratively\\nseized his cell phone. The CBP officials called Homeland\\nSecurity Investigations (HSI), which dispatched Agents\\nPetonak and Medrano to investigate. After arriving, Agent\\nPetonak “briefly” and manually reviewed Cano’s cell phone,\\nnoticing a “lengthy call log” but no text messages. Agent\\nPetonak later stated that the purpose of this manual search\\nwas “two-pronged”: “to find some brief investigative leads in\\nthe current case,” and “to see if there’s evidence of other\\nthings coming across the border.”\\n\\n Agent Petonak proceeded to question Cano, who waived\\nhis Miranda rights and agreed to talk. During that interview,\\nCano denied any knowledge of the cocaine. Cano stated that\\nhe had moved to Tijuana to look for work in nearby San\\nDiego, because work was slow in Los Angeles. He also said\\nhe had crossed the border every day for the previous three\\nweeks looking for work. He told Agent Petonak that he was\\nheaded to a carpet store in Chula Vista that day to seek work.\\nWhen pressed, Cano was not able to provide the name or\\naddress of the store, claiming that he intended to look it up on\\nGoogle after crossing the border. Cano also explained that he\\ndid not have his flooring tools with him in his pickup truck so\\n\\f UNITED STATES V. CANO 7\\n\\nas to avoid problems with border crossings; Cano intended to\\ndrive to Los Angeles to retrieve his tools if he located work\\nin San Diego.\\n\\n During the interrogation, Agent Petonak specifically\\nasked Cano about the lack of text messages on his cell phone.\\nCano responded that his cousin had advised him to delete his\\ntext messages “just in case” he got pulled over in Mexico and\\npolice were to check his cell phone. Cano stated that he\\nerased his messages to avoid “any problems” with the\\nMexican police.\\n\\n While Agent Petonak questioned Cano, Agent Medrano\\nconducted a second manual search of the cell phone. Agent\\nMedrano browsed the call log and wrote down some of the\\nphone numbers on a piece of paper. He also noticed two\\nmessages that arrived after Cano had reached the border, and\\nhe took a photograph of the messages. The first message\\nstated, “Good morning,” and the second message stated,\\n“Primo, are you coming to the house?” Agent Medrano gave\\nall of this information—the recorded list of calls and the\\nphotograph—to Agent Petonak.\\n\\n Finally, Agent Medrano conducted a “logical download”\\nof the phone using Cellebrite software. A Cellebrite search\\nenables the user to access text messages, contacts, call logs,\\nmedia, and application data on a cell phone and to select\\nwhich types of data to download. It does not, however, allow\\nthe user to access data stored within third-party applications.\\nAgent Medrano typically does not select the option to\\ndownload photographs.\\n\\n After Agent Petonak interviewed Cano, he reviewed the\\nresults of the Cellebrite download of Cano’s phone by Agent\\n\\f8 UNITED STATES V. CANO\\n\\nMedrano. The Cellebrite results revealed that Cano had sent\\nno text messages, and it listed all the calls made by Cano.\\nAgent Petonak later concluded that none of the phone\\nnumbers in the call log corresponded to carpeting stores in\\nSan Diego.\\n\\nB. The Proceedings\\n\\n Cano was indicted for importing cocaine. Before trial,\\nCano moved to suppress any evidence obtained from Agents\\nPetonak and Medrano’s warrantless searches of his cell phone\\nat the border. The district court denied Cano’s motion, ruling\\nthat the manual searches and the Cellebrite search of Cano’s\\nphone were valid border searches. During trial, the\\ngovernment introduced evidence that resulted from the\\nmanual searches of the phone and from Agent Medrano’s\\nCellebrite download of the phone.1\\n\\n In preparation for trial, Cano indicated his intent to\\npresent a third-party culpability defense claiming that his\\ncousin, Jose Medina, was responsible for placing the drugs in\\nCano’s spare tire without Cano’s knowledge. Cano proffered\\nevidence that Medina had a key to Cano’s car and had driven\\nit shortly before Cano’s attempted border crossing, that\\n\\n\\n 1\\n Some—but not all—of the evidence was available through\\nalternative channels. For example, the government introduced a call log,\\nunchallenged by Cano, that the government received from Cano’s phone\\ncompany. Similarly, the government later obtained a warrant to search the\\nphone, and an agent conducted further searches. Because the government\\nintroduced at trial much evidence pre-dating those events, and because the\\ngovernment has not argued that any Fourth Amendment error was\\nharmless, those later events do not affect our Fourth Amendment analysis\\nof the warrantless searches. United States v. Rodriguez, 880 F.3d 1151,\\n1163 (9th Cir. 2018)\\n\\f UNITED STATES V. CANO 9\\n\\nMedina had a criminal record including a conviction for\\ncocaine possession, that Medina was a member of a Chicago-\\nbased gang called the Latin Kings, and that the Latin Kings\\nsold cocaine within the United States and were involved with\\na cartel that trafficked drugs across the border.\\n\\n Following Cano’s implication of Medina, the government\\ncontacted Medina and promised him immunity and\\nimmigration papers in exchange for his cooperation. Medina\\ninitially denied being involved with drugs, but later contacted\\nthe government on his own and offered to help them with the\\n“biggest RICO case” and “drug seizures of 20 to 25\\nkilograms at a time.” All of this information was made\\navailable to Cano.\\n\\n As part of his defense, Cano sought additional discovery\\nfrom HSI, the Federal Bureau of Investigation (FBI), and the\\nDrug Enforcement Agency (DEA) regarding: (1) records\\nlinking Medina to drug sales, distribution, or trafficking; and\\n(2) records linking the Latin Kings to drug trafficking from\\nMexico to Southern California. The government opposed\\nCano’s discovery motion, arguing that the evidence was not\\nmaterial under Federal Rule of Criminal Procedure\\n16(a)(1)(E)(i) and that discovery should be limited to HSI, as\\nneither the DEA nor the FBI had participated in the\\ninvestigation of Cano. The district court originally overruled\\nboth objections, finding the evidence material under Rule 16\\nand exculpatory under Brady v. Maryland, 373 U.S. 83\\n(1963). The court also reasoned that, because HSI could\\ninquire of the DEA and FBI if it sought inculpatory evidence,\\nHSI had access to the files and was required to provide any\\nexculpatory evidence held by the DEA or FBI.\\n\\f10 UNITED STATES V. CANO\\n\\n In response to the court’s discovery order, HSI produced\\nMedina’s immigration file and his Bureau of Prisons record.\\nAgent Petonak also searched for Medina’s name in two\\ndifferent police clearinghouses, but neither returned any hits.2\\nBoth Agent Petonak and the United States Attorney’s Office\\n(USAO) subsequently requested information showing a link\\nbetween the Latin Kings and drug trafficking from Mexico\\nfrom the legal counsel of both the FBI and DEA. Both\\nagencies denied the requests without providing any\\nexplanation or any indication as to whether the requested\\ninformation existed.\\n\\n Following these attempts, the government moved for the\\ndistrict court to reconsider its discovery order and excuse it\\nfrom discovery relating to files held by the FBI and DEA.\\nThe district court granted the motion to reconsider, finding\\nthat the prosecutor did not have access to the evidence when\\nhe was “rebuffed” by agencies over which he had no control.\\n\\n The case proceeded to trial and Cano presented his third-\\nparty culpability defense. The first trial resulted in a hung\\njury and a mistrial. On retrial, Cano again relied on his third-\\nparty culpability defense. The second trial resulted in Cano’s\\nconviction. This appeal followed, in which Cano raises three\\nissues: (1) whether the warrantless searches of his cell phone\\nviolated the Fourth Amendment and whether the resulting\\nevidence should be suppressed; (2) whether the government’s\\nnon-disclosure of materials that may have been held by the\\nDEA and FBI violated his right to due process under Brady\\n\\n\\n 2\\n A police clearinghouse works for the purpose of “deconfliction” by\\nnotifying an agency if another agency has an investigation pending against\\nthe same person or item. The DEA and FBI participate in the two\\nclearinghouses searched by Agent Petonak.\\n\\f UNITED STATES V. CANO 11\\n\\nand Federal Rule of Criminal Procedure 16; and (3) whether\\nthe government raised an improper propensity inference in its\\nclosing argument. We address Cano’s first two arguments in\\nturn. Because we conclude that the district court erred in\\ndenying Cano’s motion to suppress, we vacate Cano’s\\nconviction and do not reach his claim of prosecutorial\\nmisconduct.\\n\\n II. THE WARRANTLESS SEARCH OF CANO’S\\n CELL PHONE\\n\\n The Fourth Amendment protects “[t]he right of the people\\nto be secure in their persons, houses, papers, and effects,\\nagainst unreasonable searches and seizures.” U.S. Const.\\namend. IV.3 Ordinarily, before conducting a search, police\\nmust obtain a warrant issued by a judicial officer based “upon\\nprobable cause, supported by Oath or affirmation, and\\nparticularly describing the place to be searched, and the\\npersons or things to be seized.” Id. Warrants are generally\\nrequired “unless ‘the exigencies of the situation’ make the\\nneeds of law enforcement so compelling that the warrantless\\nsearch is objectively reasonable under the Fourth\\nAmendment.” Mincey v. Arizona, 437 U.S. 385, 393–94\\n(1978) (citation omitted). Consequently, “searches conducted\\noutside the judicial process, without prior approval by judge\\nor magistrate, are per se unreasonable under the Fourth\\nAmendment—subject only to a few specifically established\\nand well-delineated exceptions.” Katz v. United States, 389\\nU.S. 347, 357 (1967) (footnote omitted). Such “specifically\\nestablished and well-delineated exceptions” include exigent\\n\\n\\n 3\\n We review de novo “the district court’s determination that [a]\\nwarrantless search . . . was a valid border search.” United States v.\\nCardona, 769 F.2d 625, 628 (9th Cir. 1985).\\n\\f12 UNITED STATES V. CANO\\n\\ncircumstances, searches incident to arrest, vehicle searches,\\nand border searches. See Arizona v. Gant, 556 U.S. 332, 343\\n(2009) (vehicle searches); Brigham City v. Stuart, 547 U.S.\\n398, 403 (2006) (exigent circumstances; listing other\\nexceptions, including warrantless entry to fight a fire, to\\nprevent the imminent destruction of evidence, or in “hot\\npursuit” of a fleeing suspect); United States v. Ramsey, 431\\nU.S. 606, 616 (1977) (border searches); Weeks v. United\\nStates, 232 U.S. 383, 392 (1914) (searches incident to arrest),\\noverruled in part on other grounds by Mapp v. Ohio, 367\\nU.S. 643 (1961).\\n\\n Exceptions to the warrant requirement are subject to two\\nimportant constraints. First, any search conducted under an\\nexception must be within the scope of the exception. Second,\\nsome searches, even when conducted within the scope of the\\nexception, are so intrusive that they require additional\\njustification, up to and including probable cause and a\\nwarrant.\\n\\n The first constraint is illustrated by the Supreme Court’s\\ndecision in Riley v. California, 573 U.S. 373 (2014), a case\\ninvolving the search incident to arrest exception. In Riley, the\\nCourt addressed “whether the police may, without a warrant,\\nsearch digital information on a cell phone seized from an\\nindividual who has been arrested”; in other words, whether\\ncell phones fell within the scope of the search incident to\\narrest exception. Id. at 378. The Court began by recognizing\\nthe increasing role in our lives of “minicomputers that also\\nhappen to have the capacity to be used as a telephone”;\\n“[m]odern cell phones, as a category, implicate privacy\\nconcerns far beyond those implicated by the search of a\\ncigarette pack, a wallet, or a purse.” Id. at 393.\\nAcknowledging that “it has been well accepted that [a search\\n\\f UNITED STATES V. CANO 13\\n\\nincident to lawful arrest] constitutes an exception to the\\nwarrant requirement,” id. at 382, the Court pointed out that\\nsuch searches serve two purposes: (1) to secure “the officer’s\\nsafety” and (2) to “prevent . . . concealment or destruction [of\\nevidence],” id. at 383 (citation omitted). The Court then\\nconsidered whether a cell phone search qualified as a search\\nincident to arrest by considering “whether application of the\\nsearch incident to arrest doctrine to [cell phones] would\\n‘untether the rule from the justifications underlying the . . .\\nexception.’” Id. at 386 (quoting Gant, 556 U.S. at 343).\\n\\n The Court concluded that neither purpose for the search\\nincident to arrest exception justified the search of a cell\\nphone. The Court rejected the government’s argument that\\nsearching a cell phone incident to arrest would “help ensure\\nofficer safety in . . . indirect ways, for example by alerting\\nofficers that confederates of the arrestee are headed to the\\nscene.” Id. at 387. The Court reasoned that the government’s\\nposition “would . . . represent a broadening” of the\\nexception’s foundational concern that “an arrestee himself\\nmight grab a weapon and use it against an officer.” Id. at\\n387–88. The Court observed that “once law enforcement\\nofficers have secured a cell phone, there is no longer any risk\\nthat the arrestee himself will be able to delete incriminating\\ndata from the phone,” id. at 388, and police have means to\\nensure that data cannot be wiped from the phone remotely,\\nid. at 390. The Court concluded “not that the information on\\na cell phone is immune from search; [but rather] that a\\nwarrant is generally required before such a search, even when\\na cell phone is seized incident to arrest.” Id. at 401.\\n\\n The second constraint on warrantless searches is\\nillustrated by the Court’s decision in United States v.\\nMontoya de Hernandez, 473 U.S. 531 (1985). Montoya was\\n\\f14 UNITED STATES V. CANO\\n\\nstopped at Los Angeles International Airport and referred to\\nsecondary inspection. Id. at 533. She had arrived from\\nBogota and was carrying $5,000 in cash. Id. She had no\\ncredit cards and no hotel reservations. Id. at 533–34.\\nBecause border officials suspected that Montoya may have\\nswallowed cocaine-filled balloons, Montoya was held in the\\ncustoms office and, after a magistrate judge issued an order,\\ntaken to a hospital for a rectal examination. Id. at 534–35.\\nOver the next four days, she passed 88 balloons containing\\ncocaine. Id. at 536. Montoya argued that the search she was\\nsubjected to, though a border search, was so intrusive that it\\ncould not be conducted without a high level of particularized\\nsuspicion. Id. at 536–37, 540. The Court balanced her\\nprivacy interests against the interests of the government at the\\nborder and concluded that, while routine searches may be\\nconducted at the border without any showing of suspicion, a\\nmore intrusive, nonroutine search must be supported by\\n“reasonable suspicion.” Id. at 537–41; see also United States\\nv. Flores-Montano, 541 U.S. 149, 152 (2004) (suggesting that\\nnonroutine searches are limited to “highly intrusive searches\\nof the person” involving “dignity and privacy interests”).\\n\\n Cano recognizes that he was subject to search at the\\nborder, but Cano and amicus Electronic Frontier Foundation\\n(“EFF”) raise two categorical challenges and one as-applied\\nchallenge to the searches conducted here. First, EFF argues\\nthat any warrantless search of a cell phone falls outside the\\nscope of the border search exception. Second, EFF argues\\nthat even if the search is within the scope of the border search\\nexception, a warrantless cell phone search is so intrusive that\\nit requires probable cause. We address these categorical\\nchallenges in Part II.A. Third, Cano asserts that, even if cell\\nphones are generally subject to search at the border, the\\nmanual and forensic searches of his cell phone exceeded the\\n\\f UNITED STATES V. CANO 15\\n\\n“well delineated” scope of the border search. We address this\\nas-applied question in Part II.B. Finally, the government\\nargues that even if the border search exceeded the limits of\\nthe Fourth Amendment, the search was conducted in good\\nfaith, and the evidence is admissible. We consider the good\\nfaith exception in Part II.C.\\n\\nA. Border Searches and Cell Phones\\n\\n “[B]order searches constitute a ‘historically recognized\\nexception to the Fourth Amendment’s general principle that\\na warrant be obtained.’” Cotterman, 709 F.3d at 957 (quoting\\nRamsey, 431 U.S. at 621). Indeed, border searches typically\\ndo not require any particularized suspicion, so long as they\\nare “routine inspections and searches of individuals or\\nconveyances seeking to cross our borders.” Almeida-Sanchez\\nv. United States, 413 U.S. 266, 272 (1973); see United States\\nv. Seljan, 547 F.3d 993, 999 (9th Cir. 2008) (en banc). Such\\nsearches are “reasonable simply by virtue of the fact they\\noccur at the border.” Ramsey, 431 U.S. at 616. The\\nexception is “rooted in ‘the long-standing right of the\\nsovereign to protect itself by stopping and examining persons\\nand property crossing into this country,’” Cotterman, 709\\nF.3d at 960 (quoting Ramsey, 431 U.S. at 616), to “prevent[]\\nthe entry of unwanted persons and effects,” id. (quoting\\nFlores-Montano, 541 U.S. at 152).\\n\\n The sovereign’s right to conduct suspicionless searches at\\nthe border “does not mean, however, that at the border\\n‘anything goes.’” Id. (quoting Seljan, 547 F.3d at 1000).\\nRather, the border search exception is a “narrow exception”\\nthat is limited in two important ways. Id. (citation omitted).\\nFirst, “[t]he authorizing statute limits the persons who may\\nlegally conduct a ‘border search’ to ‘persons authorized to\\n\\f16 UNITED STATES V. CANO\\n\\nboard or search vessels.’” United States v. Soto-Soto, 598\\nF.2d 545, 549 (9th Cir. 1979) (citing 19 U.S.C. § 482).4 This\\nincludes customs and immigration officials, but not general\\nlaw enforcement officers such as FBI agents. Id.; see United\\nStates v. Diamond, 471 F.2d 771, 773 (9th Cir. 1973) (stating\\nthat “customs agents are not general guardians of the public\\npeace”). Second, a border search must be conducted “in\\nenforcement of customs laws.” Soto-Soto, 598 F.2d at 549.\\nA border search must be conducted to “enforce importation\\nlaws,” and not for “general law enforcement purposes.” Id.\\n\\n\\n 4\\n Section 482 now reads in relevant part:\\n\\n Any of the officers or persons authorized to board\\n or search vessels may stop, search, and examine . . . any\\n vehicle, beast, or person, on which or whom he or they\\n shall suspect there is merchandise which is subject to\\n duty, or shall have been introduced into the United\\n States in any manner contrary to law . . . . [and may]\\n seize and secure the same for trial.\\n\\n19 U.S.C. § 482(a); see id. § 1467 (“[T]he appropriate customs officer for\\n[a] port or place of arrival may . . . enforce, cause inspection, examination,\\nand search to be made of the persons, baggage, and merchandise\\ndischarged or unladen from [an arriving] vessel . . . .”); id. § 1496 (“The\\nappropriate customs officer may cause an examination to be made of the\\nbaggage of any persons arriving in the United States in order to ascertain\\nwhat articles are contained therein and whether subject to duty, free of\\nduty, or prohibited . . . .”); id. § 1582 (“[A]ll persons coming into the\\nUnited States from foreign countries shall be liable to detention and search\\nby authorized officers or agents . . . .”).\\n\\n The Court has described § 482 as granting the executive “plenary\\nauthority to conduct routine searches and seizures at the border, without\\nprobable cause or a warrant.” Montoya de Hernandez, 473 U.S. at 537.\\nWe have held that the “outer limits of authority delegated by [§ 482 are]\\navailable only in border searches.” Corngold v. United States, 367 F.2d\\n1, 3 (9th Cir. 1966) (en banc).\\n\\f UNITED STATES V. CANO 17\\n\\nA general search cannot be “justif[ied] . . . on the mere basis\\nthat it occurred at the border.” Id. (affirming the suppression\\nof evidence where an FBI agent stopped and searched the\\nvehicle of an alien to determine whether the car had been\\nstolen).\\n\\n 1. Cell Phone Data as Contraband\\n\\n As we discussed briefly above, the Supreme Court has\\nidentified two principal purposes behind warrantless border\\nsearches: First, to identify “[t]ravellers . . . entitled to come\\nin” and, second, to verify their “belongings as effects which\\nmay be lawfully brought in.” Carroll v. United States, 267\\nU.S. 132, 154 (1925); see Ramsey, 431 U.S. at 620 (“The\\nborder-search exception is grounded in the recognized right\\nof the sovereign to control . . . who and what may enter the\\ncountry.”).\\n\\n EFF argues that applying the border search exception to\\na cell phone’s data would “untether” the exception from the\\npurposes underlying it. EFF contends that a border search\\nencompasses only a search for illegal persons and physical\\ncontraband located on the body of the applicant for\\nadmission or among his effects. Because digital data on a cell\\nphone cannot conceal objects such as drugs, guns, or\\nsmuggled persons, EFF asserts that digital cell phone\\nsearches are always beyond the scope of the border search\\nexception.\\n\\n We agree with EFF that the purpose of the border search\\nis to interdict contraband, but we disagree with its premise\\nthat cell phones cannot contain contraband. Although cell\\n\\f18 UNITED STATES V. CANO\\n\\nphone data cannot hide physical objects,5 the data can contain\\ndigital contraband. The best example is child pornography.\\nSee United States v. Molina-Isidoro, 884 F.3d 287, 295 n.3\\n(5th Cir. 2018) (Costa, J., specially concurring) (“One type of\\ncontraband that can be stored within the data of a cell\\nphone . . . is child pornography.”). And because cell phones\\nmay ultimately be released into the interior, even if the owner\\nhas been detained, the United States has a strong interest in\\npreventing the entry of such material. See, e.g., United States\\nv. Vergara, 884 F.3d 1309, 1311 (11th Cir.) (describing how\\nagents returned one of the defendant’s phones to a family\\nmember after defendant had been arrested for possessing\\nchild pornography on his other two phones), cert. denied, 139\\nS. Ct. 70 (2018). We find no basis for the proposition that the\\nborder search exception is limited to searching for physical\\ncontraband. At the very least, a cell phone that has photos\\nstored on it is the equivalent of photographs, magazines, and\\nbooks.6 See Riley, 573 U.S. at 394; Cotterman, 709 F.3d at\\n964. The contents may be digital when they are on the phone,\\nbut the physicality of the phone itself and the possibility that\\n\\n\\n 5\\n No one contests that a border official could, consistent with the\\nFourth Amendment, examine the physical body of a cell phone to see if\\nthe phone itself is contraband—because, for example, it is a pirated copy\\nof a patented U.S. phone—or if the phone itself presents a physical threat\\nto officers. See Riley, 573 U.S. at 387 (“Law enforcement officers remain\\nfree to examine the physical aspects of a phone to ensure that it will not\\nbe used as a weapon—say, to determine whether there is a razor blade\\nhidden between the phone and its case.”). The dispute here concerns only\\nwhether border officials may search the digital data contained within the\\nphone.\\n 6\\n We need not address here questions surrounding the use of “cloud\\ncomputing,” where the phone gives access to, but does not contain in its\\nown memory, digital data stored in the cloud. See Riley, 573 U.S. at\\n397–98; Cotterman, 709 F.3d at 965 & n.12.\\n\\f UNITED STATES V. CANO 19\\n\\nthe phone’s contents can be printed or shared electronically\\ngives border officials sufficient reason to inspect it at the\\nborder. We conclude that cell phones—including the phones’\\ndata—are subject to search at the border.\\n\\n 2. Forensic Cell Phone Searches as an Intrusive Search\\n\\n The second question we must address in response to\\namicus EFF is whether forensic searches of a cell phone are\\nso intrusive that they require reasonable suspicion or even\\nprobable cause. We answered this question in our en banc\\ndecision in Cotterman, but with respect to laptop computers.7\\nCotterman, 709 F.3d at 962–68. Cotterman was a United\\nStates citizen returning to the United States from Mexico. Id.\\nat 957. When he reached the port of entry, border officials\\nnoted that Cotterman had various convictions for sexual\\nconduct with children. Id. Concerned that Cotterman might\\nbe involved in child sex tourism, officials conducted a brief\\nsearch of his laptop computers and digital cameras and noted\\nthat the laptops had password-protected files. Id. at 958. The\\nofficials detained the computers for several days in order to\\nrun a comprehensive forensic search of the hard drive, which\\nrevealed hundreds of images of child pornography. Id. at\\n958–59. For us, “the legitimacy of the initial search of\\nCotterman’s electronic devices at the border [was] not in\\ndoubt,” id. at 960, “[t]he difficult question . . . [was] the\\nreasonableness, without a warrant, of the forensic\\n\\n\\n\\n 7\\n Although Cotterman referred to “electronic devices” generally, see\\n709 F.3d at 962–68, our holding was limited to the “examination of\\nCotterman’s computer,” id. at 968, and did not address cell phones. We\\nmentioned cell phones only once—in the first paragraph of the\\nintroduction describing the modern “digital world.” Id. at 956.\\n\\f20 UNITED STATES V. CANO\\n\\nexamination that comprehensively analyzed the hard drive of\\nthe computer,” id. at 961.\\n\\n We acknowledged the “substantial personal privacy\\ninterests” in “[e]lectronic devices . . . capable of storing\\nwarehouses full of information.” Id. at 964. At the same\\ntime, we recognized “the important security concerns that\\nprevail at the border” and the legitimacy of “[t]he effort to\\ninterdict child pornography.” Id. at 966. We held that a\\nroutine, manual search of files on a laptop computer—“a\\nquick look and unintrusive search”—is reasonable “even\\nwithout particularized suspicion,” but that officials must\\n“possess a particularized and objective basis for suspecting\\nthe person stopped of criminal activity” to engage in a\\nforensic examination, which is “essentially a computer strip\\nsearch.” Id. at 960–61, 966, 967 (citation omitted). We\\nconcluded that reasonable suspicion was “a modest, workable\\nstandard that is already applied in the extended border search,\\nTerry stop, and other contexts.” Id. at 966; see id. at 968\\n(defining reasonable suspicion as “a particularized and\\nobjective basis for suspecting the particular person stopped of\\ncriminal activity” (quoting United States v. Cortez, 449 U.S.\\n411, 417–18 (1981))).\\n\\n We think that Cotterman’s reasoning applies equally to\\ncell phones. In large measure, we anticipated the Supreme\\nCourt’s reasoning in Riley, 573 U.S. at 393–97, when we\\nrecognized in Cotterman that digital devices “contain the\\nmost intimate details of our lives” and “the uniquely sensitive\\nnature of data on electronic devices carries with it a\\nsignificant expectation of privacy,” Cotterman, 709 F.3d at\\n965–66; see Riley, 573 U.S. at 385, 393 (describing cell\\nphones as “a pervasive and insistent part of daily life” that,\\n“as a category, implicate privacy concerns far beyond those\\n\\f UNITED STATES V. CANO 21\\n\\nimplicated by the search of a cigarette pack, a wallet, or a\\npurse”). The Court’s view of cell phones in Riley so closely\\nresembles our own analysis of laptop computers in Cotterman\\nthat we find no basis to distinguish a forensic cell phone\\nsearch from a forensic laptop search.8\\n\\n Nor do we believe that Riley renders the Cotterman\\nstandard insufficiently protective. Riley, of course, held that\\n“a warrant is generally required” before searching a cell\\nphone, “even when a cell phone is seized incident to arrest.”\\n573 U.S. at 401. But here we deal with the border search\\nexception—not the search incident to arrest exception—and\\nthe difference in context is critical. In light of the\\ngovernment’s enhanced interest in protecting the “integrity of\\nthe border” and the individual’s decreased expectation of\\nprivacy, the Court has emphasized that “the Fourth\\nAmendment’s balance of reasonableness is qualitatively\\ndifferent at the international border than in the interior” and\\nis “struck much more favorably to the Government.”\\nMontoya de Hernandez, 473 U.S. at 538–40. As a result,\\npost-Riley, no court has required more than reasonable\\nsuspicion to justify even an intrusive border search. See\\nUnited States v. Wanjiku, 919 F.3d 472, 485 (7th Cir. 2019)\\n(“[N]o circuit court, before or after Riley, has required more\\nthan reasonable suspicion for a border search of cell phones\\nor electronically-stored data.”); Touset, 890 F.3d at 1234\\n\\n 8\\n We note that the Eleventh Circuit disagreed with Cotterman in\\nUnited States v. Touset, 890 F.3d 1227, 1234 (11th Cir. 2018). The court\\nheld that no level of suspicion was required to conduct a forensic search\\nof a cell phone. Id. at 1234–35. Nevertheless, the Touset court held, in\\nthe alternative, that the forensic search of various electronic devices seized\\nat the border were supported by reasonable suspicion. Id. at 1237. As\\nwith most cell phone search cases, in Touset border agents were looking\\nfor child pornography.\\n\\f22 UNITED STATES V. CANO\\n\\n(“Riley, which involved the search-incident-to-arrest\\nexception, does not apply to searches at the border.”);\\nMolina-Isidoro, 884 F.3d at 291 (“For border searches both\\nroutine and not, no case has required a warrant.”); id. at 293\\n(“The bottom line is that only two of the many federal cases\\naddressing border searches of electronic devices have ever\\nrequired any level of suspicion. They both required only\\nreasonable suspicion and that was for the more intrusive\\nforensic search.”); see also Kolsuz, 890 F.3d 133, 137 (4th\\nCir. 2018) (concluding that a “forensic examination of\\nKolsuz’s phone must be considered a nonroutine border\\nsearch, requiring some measure of individualized suspicion”\\nbut declining to decide whether the standard should be\\nreasonable suspicion or probable cause).\\n\\n Accordingly, we hold that manual searches of cell phones\\nat the border are reasonable without individualized suspicion,\\nwhereas the forensic examination of a cell phone requires a\\nshowing of reasonable suspicion. See Cotterman, 709 F.3d\\nat 968.\\n\\nB. The Searches of Cano’s Cell Phone and the Scope of the\\n Border Search Exception\\n\\n Having concluded that border officials may conduct\\nsuspicionless manual searches of cell phones, but must have\\nreasonable suspicion before they conduct a forensic search,\\nwe still must address the core of Cano’s argument: whether\\nthe manual and forensic searches of his cell phone were not\\nsearches for digital contraband, but searches for evidence of\\na crime, and thus exceeded the proper scope of a border\\nsearch.\\n\\f UNITED STATES V. CANO 23\\n\\n 1. The Border Exception and the Search for Contraband\\n\\n As a threshold matter, Cano argues that border searches\\nare limited in both purpose and scope to searches for\\ncontraband.9 In response, the government argues that\\nsearches for evidence that would aid in prosecuting past and\\npreventing future border-related crimes are tethered to the\\npurpose of the border search exception—namely, interdicting\\nforeign contraband—and thus fall within its scope.\\n\\n\\n\\n\\n 9\\n Cano emphasizes that the officials who arrested him were looking\\nfor evidence of a crime, not contraband that could be seized at the border,\\nand this renders the search unconstitutional. He points to Officers Petonak\\nand Medrano, who searched Cano’s cell phone, and who testified that their\\nsearches had a dual purpose: “to find some brief investigative leads in the\\ncurrent case” and “to see if there[] [was] evidence of other things coming\\nacross the border.” Because the agents acknowledged that they sought\\nevidence to use against Cano in building a criminal case, Cano argues that\\nthe court should treat the search as one conducted for “general law\\nenforcement purposes” rather than a border search.\\n\\n Cano’s focus on the officials’ subjective motivations is misplaced,\\nhowever. As the district court recognized, “courts have repeatedly held\\nthat the Fourth Amendment’s reasonableness analysis is ‘predominantly\\nan objective inquiry.’” See Whren v. United States, 517 U.S. 806, 813\\n(1996) (upholding a “pretextual” stop because “[s]ubjective intentions\\nplay no role in ordinary . . . Fourth Amendment analysis”). We have\\nupheld border searches of persons seeking entry even when those searches\\nwere conducted “at the behest” of DEA agents seeking criminal evidence.\\nSee United States v. Schoor, 597 F.2d 1303, 1305–06 (9th Cir. 1979)\\n(holding a border search reasonable where it was conducted “at the\\nbehest” of DEA agents and included a search for certain items of evidence\\nin addition to a search for contraband). Thus, the mere fact that Officers\\nPetonak and Medrano subjectively hoped to find “investigative leads”\\npertaining to the seized shipment of cocaine does not render their searches\\nof Cano’s phone beyond the border search exception.\\n\\f24 UNITED STATES V. CANO\\n\\n This is a close question, but we think Cano has the better\\nof the argument. There is a difference between a search for\\ncontraband and a search for evidence of border-related\\ncrimes, although the distinction may not be apparent.\\nCotterman helps us focus on the difference. There, border\\nofficials had been alerted that Cotterman had a criminal\\nrecord of sex abuse of minors and might be involved in “child\\nsex tourism.” Cotterman, 709 F.3d at 957. The officials\\nseized his laptop and subjected it to searches for child\\npornography, which they found. In Cotterman, the child\\npornography was contraband subject to seizure at the border.\\nAs contraband, the child pornography is also evidence of\\nvarious crimes, including possession of child pornography,\\n18 U.S.C. § 2252A(a)(5)(B), and importation of obscene\\nmaterial, 18 U.S.C. § 1462(a). But nothing in Cotterman\\nauthorized border officials to conduct a search for evidence\\nthat Cotterman was involved in sex-related crimes generally.\\n\\n Border officials are authorized to seize “merchandise\\nwhich . . . shall have been introduced into the United States\\nin any manner contrary to law.” 19 U.S.C. § 482(a)\\n(emphasis added). The photos on Cotterman’s laptop\\ncomputer were such merchandise. 18 U.S.C. § 2252(a). But\\nborder officials have no general authority to search for crime.\\nThis is true even if there is a possibility that such crimes may\\nbe perpetrated at the border in the future. So, for example, if\\nU.S. officials reasonably suspect that a person who has\\npresented himself at the border may be engaged in price\\nfixing, see 15 U.S.C. § 1, they may not conduct a forensic\\nsearch of his phone or laptop. Evidence of price fixing—\\ntexts or emails, for example—is not itself contraband whose\\nimportation is prohibited by law. Such emails may be\\nevidence of a crime, but they are not contraband, and there is\\nno law prohibiting the importation of mere evidence of crime.\\n\\f UNITED STATES V. CANO 25\\n\\n We recognize that our analysis is in tension with the\\nFourth Circuit’s decision in Kolsuz. Kolsuz was detained at\\nWashington Dulles International Airport when customs\\nagents discovered firearm parts in his luggage. Kolsuz, 890\\nF.3d at 138–39. Kolsuz was arrested and his cell phone\\nseized. Id. at 139. The agents subjected the phone to a\\nmonth-long forensic search, producing a 896-page report. Id.\\nKolsuz challenged the search, which the district court upheld\\nand the Fourth Circuit affirmed. Id. at 139–42. The court\\napproved the forensic search because the agents had “reason\\nto believe . . . that Kolsuz was attempting to export firearms\\nillegally” and that “their search would reveal not only\\nevidence of the export violation they already had detected,\\nbut also ‘information related to other ongoing attempts to\\nexport illegally various firearm parts.’” Id. at 143 (quoting\\nthe district court; citation omitted). According to the Fourth\\nCircuit, “[t]he justification behind the border search\\nexception is broad enough to accommodate not only the\\ndirect interception of contraband as it crosses the border, but\\nalso the prevention and disruption of ongoing efforts to\\nexport contraband illegally.” Id. (emphasis added).10\\n\\n We agree with much of the Fourth Circuit’s discussion of\\nfoundational principles, but we respectfully disagree with the\\nfinal step approving the search for further evidence that\\n\\n\\n 10\\n As support for this proposition, the Fourth Circuit cited two district\\ncourt cases originating within our circuit. Both of those cases addressed\\nfact-patterns almost identical to Cano’s, and in each case the district court\\nheld that the border-search exception was not limited to searching for\\ncontraband directly. See United States v. Mendez, 240 F. Supp. 3d 1005,\\n1007–08 (D. Ariz. 2017); United States v. Ramos, 190 F. Supp. 3d 992,\\n999 (S.D. Cal. 2016). In neither case was the issue appealed to our circuit.\\nThus, Cano’s case presents the first opportunity for us to consider the\\nmatter.\\n\\f26 UNITED STATES V. CANO\\n\\nKolsuz was smuggling weapons. Our disagreement focuses\\nprecisely on the critical question that we previously\\nidentified: Does the proper scope of a border search include\\nthe power to search for evidence of contraband that is not\\npresent at the border? Or, put differently, can border agents\\nconduct a warrantless search for evidence of past or future\\nborder-related crimes? We think that the answer must be\\n“no.” The “[d]etection of . . . contraband is the strongest\\nhistoric rationale for the border-search exception.” Molina-\\nIsidoro, 884 F.3d at 295 (Costa, J., specially concurring).\\nIndeed, “every border-search case the Supreme Court has\\ndecided involved searches to locate items being smuggled”\\nrather than evidence. Id. (emphasis added); see Montoya de\\nHernandez, 473 U.S. at 537 (the border search is “to prevent\\nthe introduction of contraband into this country”); United\\nStates v. 12 200-Foot Reels of Super 8mm. Film, 413 U.S.\\n123, 125 (1973) (border searches are “necessary to prevent\\nsmuggling and to prevent prohibited articles from entry”);\\nUnited States v. Thirty-Seven Photographs, 402 U.S. 363, 376\\n(1971) (“Customs officers characteristically inspect luggage\\nand their power to do so is not questioned in this case; it is an\\nold practice and is intimately associated with excluding\\nillegal articles from the country”). In fact, the Court has long\\n“draw[n] a sharp distinction between searches for contraband\\nand those for evidence that may reveal the importation of\\ncontraband.” Molina-Isidoro, 884 F.3d at 296 (Costa, J.,\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4073,"cells":{"id":{"kind":"number","value":4649092,"string":"4,649,092"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Marcus Crum"},"decision_date":{"kind":"string","value":"2019-08-16"},"docket_number":{"kind":"string","value":"17-30261"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: William A. Fletcher, Jay S. Bybee, and Paul J. Watford, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PER CURIAM\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-30261\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 1:17-cr-00147-\\n BLW-1\\nMARCUS SCOTT CRUM,\\n Defendant-Appellee. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Idaho\\n B. Lynn Winmill, District Judge, Presiding\\n\\n Argued and Submitted December 5, 2018\\n Seattle, Washington\\n\\n Filed August 16, 2019\\n\\n Before: William A. Fletcher, Jay S. Bybee, and\\n Paul J. Watford, Circuit Judges.\\n\\n Per Curiam Opinion;\\n Dissent by Judge Watford\\n\\n2 UNITED STATES V. CRUM\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel vacated a sentence and remanded for\\nresentencing in a case in which the district court held that\\ndelivery of methamphetamine in violation of Oregon\\nRevised Statutes § 475.890 does not qualify as a “controlled\\nsubstance offense” under U.S.S.G. §§ 2K2.1(a)(4)(A) and\\n4B1.2(b).\\n\\n The district court agreed with the defendant that\\nOregon’s delivery-of-methamphetamine offense is\\noverbroad as compared to the federal definition of\\n“controlled substance offense” because only the former\\nencompasses soliciting the delivery of methamphetamine.\\nThe panel held that United States v. Shumate, 329 F.3d 1026\\n(9th Cir. 2003) (construing the same Oregon definition of\\n“delivery”), compels the holding that § 475.890 is not\\noverbroad on the basis that it encompasses soliciting\\ndelivery. The panel that the district court erred in applying\\nSandoval v. Sessions, 866 F.3d 986 (9th Cir. 2017), which is\\ninapplicable in that it involved the different analysis\\nemployed for determining whether an offense qualifies as a\\n“drug trafficking crime” under the Controlled Substance\\nAct.\\n\\n The defendant asked the panel to reconsider this court’s\\ndecision in Shumate on the ground that the commentary to\\n§ 4B1.2 (Application Note 1), on which Shumate relied to\\nhold that “controlled substance offense” encompasses\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. CRUM 3\\n\\nsolicitation offenses, lacks legal force because it is\\ninconsistent with the text of the guideline. The panel wrote\\nthat if it were free to do so, it would hold that the\\ncommentary improperly expands the definition of\\n“controlled substance offense” to include other offenses not\\nlisted in the text of the guideline, but that it is bound by this\\ncourt’s decision in United States v. Vea-Gonzales, 999 F.2d\\n1326 (9th Cir. 1993), which held that Application Note 1 of\\n§ 4B1.2 is “perfectly consistent” with the text of § 4B1.2.\\n\\n The panel rejected the defendant’s argument that\\nOregon’s delivery-of-methamphetamine offense sweeps\\nmore broadly than the federal definition of “controlled\\nsubstance offense” because the Oregon offense criminalizes\\nthe mere offer to sell methamphetamine. The panel\\nexplained that as noted in Sandoval, offering to sell a\\ncontrolled substance constitutes soliciting delivery of a\\ncontrolled substance, and because solicitation does fall\\nwithin the definition of “controlled substance offense” under\\n§ 4B1.2, an offer to sell a controlled substance under Oregon\\nlaw is a categorical match for solicitation of a “controlled\\nsubstance offense” under § 4B1.2.\\n\\n The panel concluded that the district court should\\ntherefore have applied a base offense level of 20 under\\n§ 2K2.1(a)(4)(A).\\n\\n Dissenting, Judge Watford wrote that the Oregon offense\\ncriminalizes more conduct than the federal offense does,\\nrendering the Oregon offense overbroad, because a mere\\noffer to sell does not constitute solicitation of a “controlled\\nsubstance offense.”\\n\\n4 UNITED STATES V. CRUM\\n\\n COUNSEL\\n\\nFrancis J. Zebari (argued), Special Assistant United States\\nAttorney; Bart M. Davis, United States Attorney; United\\nStates Attorney’s Office, Boise, Idaho; for Plaintiff-\\nAppellant.\\n\\nTheodore Braden Blank (argued) and Robert K. Schwarz,\\nFederal Defender Services of Idaho, Boise, Idaho, for\\nDefendant-Appellee.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n Marcus Crum pleaded guilty to being a felon in\\npossession of a firearm in violation of 18 U.S.C. § 922(g)(1).\\nThe United States Sentencing Guidelines assign a higher\\nbase offense level for that offense if the defendant has\\npreviously been convicted of a “controlled substance\\noffense.” U.S.S.G. § 2K2.1(a)(4)(A). The question before\\nus is whether Crum’s prior conviction for delivery of\\nmethamphetamine in violation of Oregon Revised Statutes\\n§ 475.890 qualifies as a “controlled substance offense.” We\\nconclude that it does, and remand to the district court for\\nresentencing.\\n\\n I\\n\\n We use the categorical approach to determine whether a\\ndefendant’s prior conviction qualifies as a federal\\n“controlled substance offense.” See United States v. Brown,\\n879 F.3d 1043, 1046 (9th Cir. 2018). Under that approach,\\nwe compare the elements of the state offense to the elements\\n\\n UNITED STATES V. CRUM 5\\n\\nof the federal definition of “controlled substance offense” to\\ndetermine whether the state offense “criminalizes a broader\\nrange of conduct than the federal definition captures.”\\nUnited States v. Edling, 895 F.3d 1153, 1155 (9th Cir. 2018).\\n\\n Section 4B1.2(b) of the Sentencing Guidelines defines\\nthe term “controlled substance offense” to mean, as relevant\\nhere, an offense under state law that prohibits the\\n“distribution[] or dispensing of a controlled substance.”\\nU.S.S.G. § 4B1.2(b). 1 The commentary to § 4B1.2,\\nspecifically Application Note 1, further provides: “‘Crime\\nof violence’ and ‘controlled substance offense’ include the\\noffenses of aiding and abetting, conspiring, and attempting\\nto commit such offenses.” § 4B1.2 cmt. n.1. Crum contends\\nthat Oregon’s delivery-of-methamphetamine offense is\\noverbroad as compared to the federal definition of a\\n“controlled substance offense.”\\n\\n The elements of the Oregon offense are fairly simple.\\nOregon Revised Statutes § 475.890 makes it unlawful “for\\nany person to deliver methamphetamine.” Under Oregon\\nlaw, “delivery” of a controlled substance means, as relevant\\nhere, the “actual, constructive or attempted transfer . . . from\\none person to another of a controlled substance.” Or. Rev.\\nStat. § 475.005(8) (emphasis added). Attempted transfer, in\\n\\n 1\\n Section 4B1.2(b) reads in full:\\n\\n The term “controlled substance offense” means an\\n offense under federal or state law, punishable by\\n imprisonment for a term exceeding one year, that\\n prohibits the manufacture, import, export, distribution,\\n or dispensing of a controlled substance (or a\\n counterfeit substance) or the possession of a controlled\\n substance (or a counterfeit substance) with intent to\\n manufacture, import, export, distribute, or dispense.\\n\\n6 UNITED STATES V. CRUM\\n\\nturn, has been construed to include soliciting another person\\nto deliver a controlled substance, see State v. Sargent, 822\\nP.2d 726, 728 (Or. Ct. App. 1991), as well as offering to sell\\na controlled substance, see State v. Pollock, 73 P.3d 297, 300\\n(Or. Ct. App. 2003). Crum argues that neither soliciting\\ndelivery nor offering to sell is encompassed within the\\nfederal offense, thus rendering the Oregon offense\\noverbroad.\\n\\n The district court agreed with Crum, relying primarily on\\nour decision in Sandoval v. Sessions, 866 F.3d 986 (9th Cir.\\n2017). There, we held that delivery of a controlled substance\\nunder Oregon law does not constitute a “drug trafficking\\ncrime” under the Controlled Substances Act because the\\nterm “drug trafficking crime” does not include solicitation,\\nwhereas Oregon’s delivery-of-a-controlled-substance\\noffense does. Id. at 989–93. Having concluded that Crum’s\\nprior conviction did not qualify as a “controlled substance\\noffense” under § 4B1.2(b), the district court sentenced him\\nusing a base offense level of 14 rather than 20.\\n\\n The government challenges the district court’s ruling on\\nappeal.\\n\\n II\\n\\n We first address Crum’s argument that Oregon’s\\ndelivery-of-methamphetamine offense sweeps more broadly\\nthan the federal “controlled substance offense” because it\\ncriminalizes soliciting the delivery of methamphetamine.\\nWe hold that Oregon’s statute is not overbroad on this basis.\\n\\n A\\n\\n Our conclusion is compelled by our court’s prior\\ndecision in United States v. Shumate, 329 F.3d 1026 (9th Cir.\\n\\n UNITED STATES V. CRUM 7\\n\\n2003), which held that delivery of marijuana under Oregon\\nlaw qualifies as a “controlled substance offense” under\\n§ 4B1.2(b). Id. at 1028–31. That case dealt with delivery of\\na controlled substance under Oregon Revised Statutes\\n§ 475.992 (now codified at § 475.752), rather than the\\ndelivery-of-methamphetamine offense under § 475.890 at\\nissue here. But because the definition of “delivery” is the\\nsame under both statutes, see Or. Rev. Stat. § 475.005, the\\nanalysis in Shumate applies here.\\n\\n We concluded in Shumate that the definition of\\n“controlled substance offense” in § 4B1.2 encompasses\\nsolicitation offenses. We acknowledged that the\\ncommentary to § 4B1.2 does not mention solicitation, even\\nthough it expands the definition of “controlled substance\\noffense” to include aiding and abetting, conspiring, and\\nattempting to commit such an offense. Shumate, 329 F.3d at\\n1030–31. However, we concluded that the commentary’s\\n“failure to mention solicitation has no legal significance.”\\nId. at 1031 (internal quotation marks omitted). We\\nexplained that the commentary does not provide an\\nexhaustive list of the offenses that are encompassed by the\\nterm “controlled substance offense” because the\\ncommentary uses the word “include.” Id. at 1030–31. And\\nsince our court had previously relied on the same\\ncommentary to hold that the term “crime of violence” in\\n§ 4B1.2 includes solicitation offenses, we determined that\\nthe term “controlled substance offense” encompasses\\nsolicitation offenses as well. Id. (discussing United States v.\\nCox, 74 F.3d 189 (9th Cir. 1996)). We therefore held that\\ndelivery of a controlled substance under Oregon law is a\\ncategorical match under § 4B1.2, even though the Oregon\\nstatute encompasses soliciting the delivery of a controlled\\nsubstance. Id.\\n\\n8 UNITED STATES V. CRUM\\n\\n Shumate controls here. The district court thus erred in\\napplying Sandoval, which involved the term “drug\\ntrafficking crime” under the Controlled Substances Act.\\nAlthough we held in Sandoval that the term does not\\nencompass solicitation offenses, 866 F.3d at 989–90, the\\nanalysis for determining whether an offense qualifies as a\\n“drug trafficking crime” under the Controlled Substances\\nAct is different from the analysis for determining whether an\\noffense qualifies as a “controlled substance offense” under\\nthe Sentencing Guidelines. See Shumate, 329 F.3d at 1030\\nn.5. The Controlled Substances Act “neither mentions\\nsolicitation nor contains any broad catch-all provision that\\ncould even arguably be read to cover solicitation.” Leyva-\\nLicea v. INS, 187 F.3d 1147, 1150 (9th Cir. 1999). In\\ncontrast, although the commentary to § 4B1.2 does not\\nmention solicitation either, it does contain a catch-all term\\n(“include”) that we have interpreted to encompass\\nsolicitation. See Shumate, 329 F.3d at 1030. In this regard,\\nour decision in Sandoval is inapplicable to this case.\\n\\n B\\n\\n Crum asks us to reconsider our decision in Shumate on\\nthe basis of an argument that was not considered in that case.\\nCrum contends that Application Note 1 of § 4B1.2 lacks\\nlegal force because it is inconsistent with the text of the\\nguideline—an assertion that, if true, would preclude courts\\nfrom relying on the commentary to expand the definition of\\n“controlled substance offense” to include solicitation. See\\nStinson v. United States, 508 U.S. 36, 45–46 (1993). In\\nCrum’s view, because the plain text of § 4B1.2(b) does not\\nencompass solicitation (or any of the inchoate offenses\\ndiscussed in the commentary), the commentary may not\\nexpand the definition of “controlled substance offense” to\\ninclude those offenses.\\n\\n UNITED STATES V. CRUM 9\\n\\n Our sister circuits are split on this issue. The First, Third,\\nand Eleventh Circuits have held that the commentary is\\nconsistent with the text of § 4B1.2(b), as the commentary\\ndoes not include any offense that is explicitly excluded by\\nthe text of the guideline. United States v. Smith, 54 F.3d 690,\\n693 (11th Cir. 1995); United States v. Piper, 35 F.3d 611,\\n617 (1st Cir. 1994); United States v. Hightower, 25 F.3d 182,\\n187 (3d Cir. 1994). On the other side of the split, the Sixth\\nand D.C. Circuits have held that the commentary conflicts\\nwith the text of § 4B1.2(b). United States v. Havis, 927 F.3d\\n382, 385–87 (6th Cir. 2019) (en banc); United States v.\\nWinstead, 890 F.3d 1082, 1090–92 (D.C. Cir. 2018). The\\nD.C. Circuit explained that the text of § 4B1.2(b) provides a\\n“very detailed definition” of “controlled substance offense,”\\nwhich does not include the offenses listed in the\\ncommentary. Winstead, 927 F.3d at 1091 (internal quotation\\nmarks omitted). The court also pointed out that the\\nSentencing Commission included attempt offenses in\\n§ 4B1.2(a) when defining “crime of violence,” but chose not\\nto include such offenses in § 4B1.2(b) when defining\\n“controlled substance offense.” Id. Those drafting choices\\nsupport the conclusion that the definition of “controlled\\nsubstance offense” excludes attempt and the related offenses\\nlisted in the commentary. Id.\\n\\n If we were free to do so, we would follow the Sixth and\\nD.C. Circuits’ lead. In our view, the commentary\\nimproperly expands the definition of “controlled substance\\noffense” to include other offenses not listed in the text of the\\nguideline. Like the Sixth and D.C. Circuits, we are troubled\\nthat the Sentencing Commission has exercised its\\ninterpretive authority to expand the definition of “controlled\\nsubstance offense” in this way, without any grounding in the\\ntext of § 4B1.2(b) and without affording any opportunity for\\ncongressional review. See Havis, 927 F.3d at 386–87;\\n\\n10 UNITED STATES V. CRUM\\n\\nWinstead, 890 F.3d at 1092. This is especially concerning\\ngiven that the Commission’s interpretation will likely\\nincrease the sentencing ranges for numerous defendants\\nwhose prior convictions qualify as controlled substance\\noffenses due solely to Application Note 1.\\n\\n We are nonetheless compelled by our court’s prior\\ndecision in United States v. Vea-Gonzales, 999 F.2d 1326\\n(9th Cir. 1993), overruled on other grounds by Custis v.\\nUnited States, 511 U.S. 485 (1994), to reject the Sixth and\\nD.C. Circuits’ view. In Vea-Gonzales, we held that\\nApplication Note 1 of § 4B1.2 is “perfectly consistent” with\\nthe text of § 4B1.2(b). 999 F.2d at 1330. We explained that\\nthe text of § 4B1.2(b) defines the term “controlled substance\\noffense” as encompassing violations of laws prohibiting the\\nmanufacture, import, export, distribution, or dispensing of\\ndrugs, and that aiding and abetting, conspiring, and\\nattempting to commit such offenses constitute violations of\\nthose laws. Id. We thus concluded that Application Note 1\\nproperly interprets the definition of the term “controlled\\nsubstance offense” to encompass aiding and abetting,\\nconspiracy, attempt, and other forms of the underlying\\noffense. Id. No intervening higher authority is “clearly\\nirreconcilable” with the reasoning of Vea-Gonzales, so we\\ncannot overrule that precedent as a three-judge panel. See\\nUnited States v. Pepe, 895 F.3d 679, 685–86 (9th Cir. 2018).\\nAs a result, we are not free to depart from the holding in our\\nprior cases that the term “controlled substance offense” as\\ndefined in § 4B1.2(b) encompasses both solicitation and\\nattempt offenses. See Shumate, 329 F.3d at 1029–31; Vea-\\nGonzales, 999 F.2d at 1330.\\n\\n III\\n\\n We turn next to Crum’s argument that Oregon’s\\ndelivery-of-methamphetamine offense sweeps more broadly\\n\\n UNITED STATES V. CRUM 11\\n\\nthan the federal definition of “controlled substance offense”\\nbecause the Oregon offense criminalizes the mere offer to\\nsell methamphetamine.\\n\\n Crum’s argument turns on the Oregon Court of Appeals’\\ndecision in Pollock, which was issued after our court decided\\nShumate. In Pollock, the Oregon Court of Appeals held that\\nan individual can be convicted of delivery of a controlled\\nsubstance under Oregon law if he has offered to sell that\\nsubstance to another person. 73 P.3d at 300. In Crum’s\\nview, merely offering to sell a controlled substance does not\\nconstitute either soliciting or attempting to commit a\\n“controlled substance offense.” Thus, even if the definition\\nof “controlled substance offense” under § 4B1.2\\nencompasses solicitation and attempt, Crum argues that\\nOregon’s delivery-of-methamphetamine offense is still\\noverbroad.\\n\\n We reject Crum’s argument. As we noted in Sandoval,\\noffering to sell a controlled substance constitutes soliciting\\ndelivery of a controlled substance. 866 F.3d at 990–91\\n(discussing Pollock, among other Oregon cases); see also\\nUnited States v. Lee, 704 F.3d 785, 790 n.2 (9th Cir. 2012).\\nSolicitation does not fall within the definition of “drug\\ntrafficking crime” under the Controlled Substances Act,\\nwhich is the term we were construing in Sandoval. But\\nsolicitation does fall within the definition of “controlled\\nsubstance offense” under § 4B1.2 of the Sentencing\\nGuidelines. See Shumate, 329 F.3d at 1030. Thus, an offer\\nto sell a controlled substance under Oregon law is a\\ncategorical match for solicitation of a “controlled substance\\noffense” under § 4B1.2.\\n\\n * * *\\n\\n12 UNITED STATES V. CRUM\\n\\n In sum, Crum’s prior conviction for delivery of\\nmethamphetamine qualifies as a “controlled substance\\noffense,” as that term is defined in § 4B1.2 of the Guidelines.\\nThe district court should therefore have applied a base\\noffense level of 20 rather than 14. See U.S.S.G.\\n§ 2K2.1(a)(4)(A). We vacate Crum’s sentence and remand\\nfor resentencing.\\n\\n VACATED and REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426345/\", \"author_raw\": \"PER CURIAM\"}, {\"author\": \"WATFORD, Circuit Judge, dissenting\", \"type\": \"dissent\", \"text\": \"WATFORD, Circuit Judge, dissenting:\\n\\n I would affirm. In my view, Oregon’s delivery-of-\\nmethamphetamine offense is overbroad, even if the term\\n“controlled substance offense” under U.S.S.G. § 4B1.2(b)\\nencompasses solicitation, as the majority concludes. Oregon\\nlaw permits conviction for delivery of a controlled substance\\nbased on a mere offer to sell the drug to someone else. See\\nState v. Pollock, 73 P.3d 297, 300 (Or. Ct. App. 2003).\\nBecause a mere offer to sell does not constitute solicitation\\nof a “controlled substance offense,” the Oregon offense\\ncriminalizes more conduct than the federal offense does,\\nrendering the Oregon offense overbroad.\\n\\n The problem with the majority’s solicitation analysis, as\\nI see it, is this. Solicitation is enticing or encouraging\\nsomeone else to commit a crime. See Model Penal Code\\n§ 5.02(1) (American Law Institute 1985). Here, for our\\npurposes, the crime that’s covered by the federal definition\\nof “controlled substance offense” is distributing or\\ndispensing a controlled substance. To solicit that offense,\\nthe defendant must entice or encourage someone else to\\ndistribute or dispense drugs to a third party. If the defendant\\nmerely offers to sell drugs to someone else, he has not\\n\\n UNITED STATES V. CRUM 13\\n\\nsolicited a “controlled substance offense” under the\\nGuidelines. At most, a mere offer to sell amounts to\\nsoliciting the other person to commit the crime of simple\\npossession. Simple possession, however, is not covered by\\nthe Guidelines’ definition of “controlled substance offense”;\\nonly possession with the intent to distribute is. U.S.S.G.\\n§ 4B1.2(b).\\n\\n Our decision in Sandoval v. Sessions, 866 F.3d 986 (9th\\nCir. 2017), on which the majority relies, reflects an incorrect\\nview of what solicitation means. In Sandoval, we equated\\noffering to sell a controlled substance with soliciting\\ndelivery of a controlled substance, id. at 990–91, but for the\\nreason just stated they are not the same thing. That analytical\\nerror was not necessary to the conclusion we ultimately\\nreached. So I do not view that aspect of Sandoval’s\\nreasoning as binding here, and I would not perpetuate the\\nerror we made there.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426345/\", \"author_raw\": \"WATFORD, Circuit Judge, dissenting\"}]}"},"judge1":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge2":{"kind":"string","value":"JAY S BYBEE"},"judge3":{"kind":"string","value":"PAUL J WATFORD"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649092/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Marcus Crum\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-16\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-30261\n Plaintiff-Appellant,\n D.C. No.\n v. 1:17-cr-00147-\n BLW-1\nMARCUS SCOTT CRUM,\n Defendant-Appellee. OPINION\n\n Appeal from the United States District Court\n for the District of Idaho\n B. Lynn Winmill, District Judge, Presiding\n\n Argued and Submitted December 5, 2018\n Seattle, Washington\n\n Filed August 16, 2019\n\n Before: William A. Fletcher, Jay S. Bybee, and\n Paul J. Watford, Circuit Judges.\n\n Per Curiam Opinion;\n Dissent by Judge Watford\n\n2 UNITED STATES V. CRUM\n\n SUMMARY *\n\n\n Criminal Law\n\n The panel vacated a sentence and remanded for\nresentencing in a case in which the district court held that\ndelivery of methamphetamine in violation of Oregon\nRevised Statutes § 475.890 does not qualify as a “controlled\nsubstance offense” under U.S.S.G. §§ 2K2.1(a)(4)(A) and\n4B1.2(b).\n\n The district court agreed with the defendant that\nOregon’s delivery-of-methamphetamine offense is\noverbroad as compared to the federal definition of\n“controlled substance offense” because only the former\nencompasses soliciting the delivery of methamphetamine.\nThe panel held that United States v. Shumate, 329 F.3d 1026\n(9th Cir. 2003) (construing the same Oregon definition of\n“delivery”), compels the holding that § 475.890 is not\noverbroad on the basis that it encompasses soliciting\ndelivery. The panel that the district court erred in applying\nSandoval v. Sessions, 866 F.3d 986 (9th Cir. 2017), which is\ninapplicable in that it involved the different analysis\nemployed for determining whether an offense qualifies as a\n“drug trafficking crime” under the Controlled Substance\nAct.\n\n The defendant asked the panel to reconsider this court’s\ndecision in Shumate on the ground that the commentary to\n§ 4B1.2 (Application Note 1), on which Shumate relied to\nhold that “controlled substance offense” encompasses\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n UNITED STATES V. CRUM 3\n\nsolicitation offenses, lacks legal force because it is\ninconsistent with the text of the guideline. The panel wrote\nthat if it were free to do so, it would hold that the\ncommentary improperly expands the definition of\n“controlled substance offense” to include other offenses not\nlisted in the text of the guideline, but that it is bound by this\ncourt’s decision in United States v. Vea-Gonzales, 999 F.2d\n1326 (9th Cir. 1993), which held that Application Note 1 of\n§ 4B1.2 is “perfectly consistent” with the text of § 4B1.2.\n\n The panel rejected the defendant’s argument that\nOregon’s delivery-of-methamphetamine offense sweeps\nmore broadly than the federal definition of “controlled\nsubstance offense” because the Oregon offense criminalizes\nthe mere offer to sell methamphetamine. The panel\nexplained that as noted in Sandoval, offering to sell a\ncontrolled substance constitutes soliciting delivery of a\ncontrolled substance, and because solicitation does fall\nwithin the definition of “controlled substance offense” under\n§ 4B1.2, an offer to sell a controlled substance under Oregon\nlaw is a categorical match for solicitation of a “controlled\nsubstance offense” under § 4B1.2.\n\n The panel concluded that the district court should\ntherefore have applied a base offense level of 20 under\n§ 2K2.1(a)(4)(A).\n\n Dissenting, Judge Watford wrote that the Oregon offense\ncriminalizes more conduct than the federal offense does,\nrendering the Oregon offense overbroad, because a mere\noffer to sell does not constitute solicitation of a “controlled\nsubstance offense.”\n\n4 UNITED STATES V. CRUM\n\n COUNSEL\n\nFrancis J. Zebari (argued), Special Assistant United States\nAttorney; Bart M. Davis, United States Attorney; United\nStates Attorney’s Office, Boise, Idaho; for Plaintiff-\nAppellant.\n\nTheodore Braden Blank (argued) and Robert K. Schwarz,\nFederal Defender Services of Idaho, Boise, Idaho, for\nDefendant-Appellee.\n\n\n OPINION\n\nPER CURIAM:\n\n Marcus Crum pleaded guilty to being a felon in\npossession of a firearm in violation of 18 U.S.C. § 922(g)(1).\nThe United States Sentencing Guidelines assign a higher\nbase offense level for that offense if the defendant has\npreviously been convicted of a “controlled substance\noffense.” U.S.S.G. § 2K2.1(a)(4)(A). The question before\nus is whether Crum’s prior conviction for delivery of\nmethamphetamine in violation of Oregon Revised Statutes\n§ 475.890 qualifies as a “controlled substance offense.” We\nconclude that it does, and remand to the district court for\nresentencing.\n\n I\n\n We use the categorical approach to determine whether a\ndefendant’s prior conviction qualifies as a federal\n“controlled substance offense.” See United States v. Brown,\n879 F.3d 1043, 1046 (9th Cir. 2018). Under that approach,\nwe compare the elements of the state offense to the elements\n\n UNITED STATES V. CRUM 5\n\nof the federal definition of “controlled substance offense” to\ndetermine whether the state offense “criminalizes a broader\nrange of conduct than the federal definition captures.”\nUnited States v. Edling, 895 F.3d 1153, 1155 (9th Cir. 2018).\n\n Section 4B1.2(b) of the Sentencing Guidelines defines\nthe term “controlled substance offense” to mean, as relevant\nhere, an offense under state law that prohibits the\n“distribution[] or dispensing of a controlled substance.”\nU.S.S.G. § 4B1.2(b). 1 The commentary to § 4B1.2,\nspecifically Application Note 1, further provides: “‘Crime\nof violence’ and ‘controlled substance offense’ include the\noffenses of aiding and abetting, conspiring, and attempting\nto commit such offenses.” § 4B1.2 cmt. n.1. Crum contends\nthat Oregon’s delivery-of-methamphetamine offense is\noverbroad as compared to the federal definition of a\n“controlled substance offense.”\n\n The elements of the Oregon offense are fairly simple.\nOregon Revised Statutes § 475.890 makes it unlawful “for\nany person to deliver methamphetamine.” Under Oregon\nlaw, “delivery” of a controlled substance means, as relevant\nhere, the “actual, constructive or attempted transfer . . . from\none person to another of a controlled substance.” Or. Rev.\nStat. § 475.005(8) (emphasis added). Attempted transfer, in\n\n 1\n Section 4B1.2(b) reads in full:\n\n The term “controlled substance offense” means an\n offense under federal or state law, punishable by\n imprisonment for a term exceeding one year, that\n prohibits the manufacture, import, export, distribution,\n or dispensing of a controlled substance (or a\n counterfeit substance) or the possession of a controlled\n substance (or a counterfeit substance) with intent to\n manufacture, import, export, distribute, or dispense.\n\n6 UNITED STATES V. CRUM\n\nturn, has been construed to include soliciting another person\nto deliver a controlled substance, see State v. Sargent, 822\nP.2d 726, 728 (Or. Ct. App. 1991), as well as offering to sell\na controlled substance, see State v. Pollock, 73 P.3d 297, 300\n(Or. Ct. App. 2003). Crum argues that neither soliciting\ndelivery nor offering to sell is encompassed within the\nfederal offense, thus rendering the Oregon offense\noverbroad.\n\n The district court agreed with Crum, relying primarily on\nour decision in Sandoval v. Sessions, 866 F.3d 986 (9th Cir.\n2017). There, we held that delivery of a controlled substance\nunder Oregon law does not constitute a “drug trafficking\ncrime” under the Controlled Substances Act because the\nterm “drug trafficking crime” does not include solicitation,\nwhereas Oregon’s delivery-of-a-controlled-substance\noffense does. Id. at 989–93. Having concluded that Crum’s\nprior conviction did not qualify as a “controlled substance\noffense” under § 4B1.2(b), the district court sentenced him\nusing a base offense level of 14 rather than 20.\n\n The government challenges the district court’s ruling on\nappeal.\n\n II\n\n We first address Crum’s argument that Oregon’s\ndelivery-of-methamphetamine offense sweeps more broadly\nthan the federal “controlled substance offense” because it\ncriminalizes soliciting the delivery of methamphetamine.\nWe hold that Oregon’s statute is not overbroad on this basis.\n\n A\n\n Our conclusion is compelled by our court’s prior\ndecision in United States v. Shumate, 329 F.3d 1026 (9th Cir.\n\n UNITED STATES V. CRUM 7\n\n2003), which held that delivery of marijuana under Oregon\nlaw qualifies as a “controlled substance offense” under\n§ 4B1.2(b). Id. at 1028–31. That case dealt with delivery of\na controlled substance under Oregon Revised Statutes\n§ 475.992 (now codified at § 475.752), rather than the\ndelivery-of-methamphetamine offense under § 475.890 at\nissue here. But because the definition of “delivery” is the\nsame under both statutes, see Or. Rev. Stat. § 475.005, the\nanalysis in Shumate applies here.\n\n We concluded in Shumate that the definition of\n“controlled substance offense” in § 4B1.2 encompasses\nsolicitation offenses. We acknowledged that the\ncommentary to § 4B1.2 does not mention solicitation, even\nthough it expands the definition of “controlled substance\noffense” to include aiding and abetting, conspiring, and\nattempting to commit such an offense. Shumate, 329 F.3d at\n1030–31. However, we concluded that the commentary’s\n“failure to mention solicitation has no legal significance.”\nId. at 1031 (internal quotation marks omitted). We\nexplained that the commentary does not provide an\nexhaustive list of the offenses that are encompassed by the\nterm “controlled substance offense” because the\ncommentary uses the word “include.” Id. at 1030–31. And\nsince our court had previously relied on the same\ncommentary to hold that the term “crime of violence” in\n§ 4B1.2 includes solicitation offenses, we determined that\nthe term “controlled substance offense” encompasses\nsolicitation offenses as well. Id. (discussing United States v.\nCox, 74 F.3d 189 (9th Cir. 1996)). We therefore held that\ndelivery of a controlled substance under Oregon law is a\ncategorical match under § 4B1.2, even though the Oregon\nstatute encompasses soliciting the delivery of a controlled\nsubstance. Id.\n\n8 UNITED STATES V. CRUM\n\n Shumate controls here. The district court thus erred in\napplying Sandoval, which involved the term “drug\ntrafficking crime” under the Controlled Substances Act.\nAlthough we held in Sandoval that the term does not\nencompass solicitation offenses, 866 F.3d at 989–90, the\nanalysis for determining whether an offense qualifies as a\n“drug trafficking crime” under the Controlled Substances\nAct is different from the analysis for determining whether an\noffense qualifies as a “controlled substance offense” under\nthe Sentencing Guidelines. See Shumate, 329 F.3d at 1030\nn.5. The Controlled Substances Act “neither mentions\nsolicitation nor contains any broad catch-all provision that\ncould even arguably be read to cover solicitation.” Leyva-\nLicea v. INS, 187 F.3d 1147, 1150 (9th Cir. 1999). In\ncontrast, although the commentary to § 4B1.2 does not\nmention solicitation either, it does contain a catch-all term\n(“include”) that we have interpreted to encompass\nsolicitation. See Shumate, 329 F.3d at 1030. In this regard,\nour decision in Sandoval is inapplicable to this case.\n\n B\n\n Crum asks us to reconsider our decision in Shumate on\nthe basis of an argument that was not considered in that case.\nCrum contends that Application Note 1 of § 4B1.2 lacks\nlegal force because it is inconsistent with the text of the\nguideline—an assertion that, if true, would preclude courts\nfrom relying on the commentary to expand the definition of\n“controlled substance offense” to include solicitation. See\nStinson v. United States, 508 U.S. 36, 45–46 (1993). In\nCrum’s view, because the plain text of § 4B1.2(b) does not\nencompass solicitation (or any of the inchoate offenses\ndiscussed in the commentary), the commentary may not\nexpand the definition of “controlled substance offense” to\ninclude those offenses.\n\n UNITED STATES V. CRUM 9\n\n Our sister circuits are split on this issue. The First, Third,\nand Eleventh Circuits have held that the commentary is\nconsistent with the text of § 4B1.2(b), as the commentary\ndoes not include any offense that is explicitly excluded by\nthe text of the guideline. United States v. Smith, 54 F.3d 690,\n693 (11th Cir. 1995); United States v. Piper, 35 F.3d 611,\n617 (1st Cir. 1994); United States v. Hightower, 25 F.3d 182,\n187 (3d Cir. 1994). On the other side of the split, the Sixth\nand D.C. Circuits have held that the commentary conflicts\nwith the text of § 4B1.2(b). United States v. Havis, 927 F.3d\n382, 385–87 (6th Cir. 2019) (en banc); United States v.\nWinstead, 890 F.3d 1082, 1090–92 (D.C. Cir. 2018). The\nD.C. Circuit explained that the text of § 4B1.2(b) provides a\n“very detailed definition” of “controlled substance offense,”\nwhich does not include the offenses listed in the\ncommentary. Winstead, 927 F.3d at 1091 (internal quotation\nmarks omitted). The court also pointed out that the\nSentencing Commission included attempt offenses in\n§ 4B1.2(a) when defining “crime of violence,” but chose not\nto include such offenses in § 4B1.2(b) when defining\n“controlled substance offense.” Id. Those drafting choices\nsupport the conclusion that the definition of “controlled\nsubstance offense” excludes attempt and the related offenses\nlisted in the commentary. Id.\n\n If we were free to do so, we would follow the Sixth and\nD.C. Circuits’ lead. In our view, the commentary\nimproperly expands the definition of “controlled substance\noffense” to include other offenses not listed in the text of the\nguideline. Like the Sixth and D.C. Circuits, we are troubled\nthat the Sentencing Commission has exercised its\ninterpretive authority to expand the definition of “controlled\nsubstance offense” in this way, without any grounding in the\ntext of § 4B1.2(b) and without affording any opportunity for\ncongressional review. See Havis, 927 F.3d at 386–87;\n\n10 UNITED STATES V. CRUM\n\nWinstead, 890 F.3d at 1092. This is especially concerning\ngiven that the Commission’s interpretation will likely\nincrease the sentencing ranges for numerous defendants\nwhose prior convictions qualify as controlled substance\noffenses due solely to Application Note 1.\n\n We are nonetheless compelled by our court’s prior\ndecision in United States v. Vea-Gonzales, 999 F.2d 1326\n(9th Cir. 1993), overruled on other grounds by Custis v.\nUnited States, 511 U.S. 485 (1994), to reject the Sixth and\nD.C. Circuits’ view. In Vea-Gonzales, we held that\nApplication Note 1 of § 4B1.2 is “perfectly consistent” with\nthe text of § 4B1.2(b). 999 F.2d at 1330. We explained that\nthe text of § 4B1.2(b) defines the term “controlled substance\noffense” as encompassing violations of laws prohibiting the\nmanufacture, import, export, distribution, or dispensing of\ndrugs, and that aiding and abetting, conspiring, and\nattempting to commit such offenses constitute violations of\nthose laws. Id. We thus concluded that Application Note 1\nproperly interprets the definition of the term “controlled\nsubstance offense” to encompass aiding and abetting,\nconspiracy, attempt, and other forms of the underlying\noffense. Id. No intervening higher authority is “clearly\nirreconcilable” with the reasoning of Vea-Gonzales, so we\ncannot overrule that precedent as a three-judge panel. See\nUnited States v. Pepe, 895 F.3d 679, 685–86 (9th Cir. 2018).\nAs a result, we are not free to depart from the holding in our\nprior cases that the term “controlled substance offense” as\ndefined in § 4B1.2(b) encompasses both solicitation and\nattempt offenses. See Shumate, 329 F.3d at 1029–31; Vea-\nGonzales, 999 F.2d at 1330.\n\n III\n\n We turn next to Crum’s argument that Oregon’s\ndelivery-of-methamphetamine offense sweeps more broadly\n\n UNITED STATES V. CRUM 11\n\nthan the federal definition of “controlled substance offense”\nbecause the Oregon offense criminalizes the mere offer to\nsell methamphetamine.\n\n Crum’s argument turns on the Oregon Court of Appeals’\ndecision in Pollock, which was issued after our court decided\nShumate. In Pollock, the Oregon Court of Appeals held that\nan individual can be convicted of delivery of a controlled\nsubstance under Oregon law if he has offered to sell that\nsubstance to another person. 73 P.3d at 300. In Crum’s\nview, merely offering to sell a controlled substance does not\nconstitute either soliciting or attempting to commit a\n“controlled substance offense.” Thus, even if the definition\nof “controlled substance offense” under § 4B1.2\nencompasses solicitation and attempt, Crum argues that\nOregon’s delivery-of-methamphetamine offense is still\noverbroad.\n\n We reject Crum’s argument. As we noted in Sandoval,\noffering to sell a controlled substance constitutes soliciting\ndelivery of a controlled substance. 866 F.3d at 990–91\n(discussing Pollock, among other Oregon cases); see also\nUnited States v. Lee, 704 F.3d 785, 790 n.2 (9th Cir. 2012).\nSolicitation does not fall within the definition of “drug\ntrafficking crime” under the Controlled Substances Act,\nwhich is the term we were construing in Sandoval. But\nsolicitation does fall within the definition of “controlled\nsubstance offense” under § 4B1.2 of the Sentencing\nGuidelines. See Shumate, 329 F.3d at 1030. Thus, an offer\nto sell a controlled substance under Oregon law is a\ncategorical match for solicitation of a “controlled substance\noffense” under § 4B1.2.\n\n * * *\n\n12 UNITED STATES V. CRUM\n\n In sum, Crum’s prior conviction for delivery of\nmethamphetamine qualifies as a “controlled substance\noffense,” as that term is defined in § 4B1.2 of the Guidelines.\nThe district court should therefore have applied a base\noffense level of 20 rather than 14. See U.S.S.G.\n§ 2K2.1(a)(4)(A). We vacate Crum’s sentence and remand\nfor resentencing.\n\n VACATED and REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Marcus Crum\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-16\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-30261\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 1:17-cr-00147-\\n BLW-1\\nMARCUS SCOTT CRUM,\\n Defendant-Appellee. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Idaho\\n B. Lynn Winmill, District Judge, Presiding\\n\\n Argued and Submitted December 5, 2018\\n Seattle, Washington\\n\\n Filed August 16, 2019\\n\\n Before: William A. Fletcher, Jay S. Bybee, and\\n Paul J. Watford, Circuit Judges.\\n\\n Per Curiam Opinion;\\n Dissent by Judge Watford\\n\\n2 UNITED STATES V. CRUM\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel vacated a sentence and remanded for\\nresentencing in a case in which the district court held that\\ndelivery of methamphetamine in violation of Oregon\\nRevised Statutes § 475.890 does not qualify as a “controlled\\nsubstance offense” under U.S.S.G. §§ 2K2.1(a)(4)(A) and\\n4B1.2(b).\\n\\n The district court agreed with the defendant that\\nOregon’s delivery-of-methamphetamine offense is\\noverbroad as compared to the federal definition of\\n“controlled substance offense” because only the former\\nencompasses soliciting the delivery of methamphetamine.\\nThe panel held that United States v. Shumate, 329 F.3d 1026\\n(9th Cir. 2003) (construing the same Oregon definition of\\n“delivery”), compels the holding that § 475.890 is not\\noverbroad on the basis that it encompasses soliciting\\ndelivery. The panel that the district court erred in applying\\nSandoval v. Sessions, 866 F.3d 986 (9th Cir. 2017), which is\\ninapplicable in that it involved the different analysis\\nemployed for determining whether an offense qualifies as a\\n“drug trafficking crime” under the Controlled Substance\\nAct.\\n\\n The defendant asked the panel to reconsider this court’s\\ndecision in Shumate on the ground that the commentary to\\n§ 4B1.2 (Application Note 1), on which Shumate relied to\\nhold that “controlled substance offense” encompasses\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. CRUM 3\\n\\nsolicitation offenses, lacks legal force because it is\\ninconsistent with the text of the guideline. The panel wrote\\nthat if it were free to do so, it would hold that the\\ncommentary improperly expands the definition of\\n“controlled substance offense” to include other offenses not\\nlisted in the text of the guideline, but that it is bound by this\\ncourt’s decision in United States v. Vea-Gonzales, 999 F.2d\\n1326 (9th Cir. 1993), which held that Application Note 1 of\\n§ 4B1.2 is “perfectly consistent” with the text of § 4B1.2.\\n\\n The panel rejected the defendant’s argument that\\nOregon’s delivery-of-methamphetamine offense sweeps\\nmore broadly than the federal definition of “controlled\\nsubstance offense” because the Oregon offense criminalizes\\nthe mere offer to sell methamphetamine. The panel\\nexplained that as noted in Sandoval, offering to sell a\\ncontrolled substance constitutes soliciting delivery of a\\ncontrolled substance, and because solicitation does fall\\nwithin the definition of “controlled substance offense” under\\n§ 4B1.2, an offer to sell a controlled substance under Oregon\\nlaw is a categorical match for solicitation of a “controlled\\nsubstance offense” under § 4B1.2.\\n\\n The panel concluded that the district court should\\ntherefore have applied a base offense level of 20 under\\n§ 2K2.1(a)(4)(A).\\n\\n Dissenting, Judge Watford wrote that the Oregon offense\\ncriminalizes more conduct than the federal offense does,\\nrendering the Oregon offense overbroad, because a mere\\noffer to sell does not constitute solicitation of a “controlled\\nsubstance offense.”\\n\\n4 UNITED STATES V. CRUM\\n\\n COUNSEL\\n\\nFrancis J. Zebari (argued), Special Assistant United States\\nAttorney; Bart M. Davis, United States Attorney; United\\nStates Attorney’s Office, Boise, Idaho; for Plaintiff-\\nAppellant.\\n\\nTheodore Braden Blank (argued) and Robert K. Schwarz,\\nFederal Defender Services of Idaho, Boise, Idaho, for\\nDefendant-Appellee.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n Marcus Crum pleaded guilty to being a felon in\\npossession of a firearm in violation of 18 U.S.C. § 922(g)(1).\\nThe United States Sentencing Guidelines assign a higher\\nbase offense level for that offense if the defendant has\\npreviously been convicted of a “controlled substance\\noffense.” U.S.S.G. § 2K2.1(a)(4)(A). The question before\\nus is whether Crum’s prior conviction for delivery of\\nmethamphetamine in violation of Oregon Revised Statutes\\n§ 475.890 qualifies as a “controlled substance offense.” We\\nconclude that it does, and remand to the district court for\\nresentencing.\\n\\n I\\n\\n We use the categorical approach to determine whether a\\ndefendant’s prior conviction qualifies as a federal\\n“controlled substance offense.” See United States v. Brown,\\n879 F.3d 1043, 1046 (9th Cir. 2018). Under that approach,\\nwe compare the elements of the state offense to the elements\\n\\n UNITED STATES V. CRUM 5\\n\\nof the federal definition of “controlled substance offense” to\\ndetermine whether the state offense “criminalizes a broader\\nrange of conduct than the federal definition captures.”\\nUnited States v. Edling, 895 F.3d 1153, 1155 (9th Cir. 2018).\\n\\n Section 4B1.2(b) of the Sentencing Guidelines defines\\nthe term “controlled substance offense” to mean, as relevant\\nhere, an offense under state law that prohibits the\\n“distribution[] or dispensing of a controlled substance.”\\nU.S.S.G. § 4B1.2(b). 1 The commentary to § 4B1.2,\\nspecifically Application Note 1, further provides: “‘Crime\\nof violence’ and ‘controlled substance offense’ include the\\noffenses of aiding and abetting, conspiring, and attempting\\nto commit such offenses.” § 4B1.2 cmt. n.1. Crum contends\\nthat Oregon’s delivery-of-methamphetamine offense is\\noverbroad as compared to the federal definition of a\\n“controlled substance offense.”\\n\\n The elements of the Oregon offense are fairly simple.\\nOregon Revised Statutes § 475.890 makes it unlawful “for\\nany person to deliver methamphetamine.” Under Oregon\\nlaw, “delivery” of a controlled substance means, as relevant\\nhere, the “actual, constructive or attempted transfer . . . from\\none person to another of a controlled substance.” Or. Rev.\\nStat. § 475.005(8) (emphasis added). Attempted transfer, in\\n\\n 1\\n Section 4B1.2(b) reads in full:\\n\\n The term “controlled substance offense” means an\\n offense under federal or state law, punishable by\\n imprisonment for a term exceeding one year, that\\n prohibits the manufacture, import, export, distribution,\\n or dispensing of a controlled substance (or a\\n counterfeit substance) or the possession of a controlled\\n substance (or a counterfeit substance) with intent to\\n manufacture, import, export, distribute, or dispense.\\n\\n6 UNITED STATES V. CRUM\\n\\nturn, has been construed to include soliciting another person\\nto deliver a controlled substance, see State v. Sargent, 822\\nP.2d 726, 728 (Or. Ct. App. 1991), as well as offering to sell\\na controlled substance, see State v. Pollock, 73 P.3d 297, 300\\n(Or. Ct. App. 2003). Crum argues that neither soliciting\\ndelivery nor offering to sell is encompassed within the\\nfederal offense, thus rendering the Oregon offense\\noverbroad.\\n\\n The district court agreed with Crum, relying primarily on\\nour decision in Sandoval v. Sessions, 866 F.3d 986 (9th Cir.\\n2017). There, we held that delivery of a controlled substance\\nunder Oregon law does not constitute a “drug trafficking\\ncrime” under the Controlled Substances Act because the\\nterm “drug trafficking crime” does not include solicitation,\\nwhereas Oregon’s delivery-of-a-controlled-substance\\noffense does. Id. at 989–93. Having concluded that Crum’s\\nprior conviction did not qualify as a “controlled substance\\noffense” under § 4B1.2(b), the district court sentenced him\\nusing a base offense level of 14 rather than 20.\\n\\n The government challenges the district court’s ruling on\\nappeal.\\n\\n II\\n\\n We first address Crum’s argument that Oregon’s\\ndelivery-of-methamphetamine offense sweeps more broadly\\nthan the federal “controlled substance offense” because it\\ncriminalizes soliciting the delivery of methamphetamine.\\nWe hold that Oregon’s statute is not overbroad on this basis.\\n\\n A\\n\\n Our conclusion is compelled by our court’s prior\\ndecision in United States v. Shumate, 329 F.3d 1026 (9th Cir.\\n\\n UNITED STATES V. CRUM 7\\n\\n2003), which held that delivery of marijuana under Oregon\\nlaw qualifies as a “controlled substance offense” under\\n§ 4B1.2(b). Id. at 1028–31. That case dealt with delivery of\\na controlled substance under Oregon Revised Statutes\\n§ 475.992 (now codified at § 475.752), rather than the\\ndelivery-of-methamphetamine offense under § 475.890 at\\nissue here. But because the definition of “delivery” is the\\nsame under both statutes, see Or. Rev. Stat. § 475.005, the\\nanalysis in Shumate applies here.\\n\\n We concluded in Shumate that the definition of\\n“controlled substance offense” in § 4B1.2 encompasses\\nsolicitation offenses. We acknowledged that the\\ncommentary to § 4B1.2 does not mention solicitation, even\\nthough it expands the definition of “controlled substance\\noffense” to include aiding and abetting, conspiring, and\\nattempting to commit such an offense. Shumate, 329 F.3d at\\n1030–31. However, we concluded that the commentary’s\\n“failure to mention solicitation has no legal significance.”\\nId. at 1031 (internal quotation marks omitted). We\\nexplained that the commentary does not provide an\\nexhaustive list of the offenses that are encompassed by the\\nterm “controlled substance offense” because the\\ncommentary uses the word “include.” Id. at 1030–31. And\\nsince our court had previously relied on the same\\ncommentary to hold that the term “crime of violence” in\\n§ 4B1.2 includes solicitation offenses, we determined that\\nthe term “controlled substance offense” encompasses\\nsolicitation offenses as well. Id. (discussing United States v.\\nCox, 74 F.3d 189 (9th Cir. 1996)). We therefore held that\\ndelivery of a controlled substance under Oregon law is a\\ncategorical match under § 4B1.2, even though the Oregon\\nstatute encompasses soliciting the delivery of a controlled\\nsubstance. Id.\\n\\n8 UNITED STATES V. CRUM\\n\\n Shumate controls here. The district court thus erred in\\napplying Sandoval, which involved the term “drug\\ntrafficking crime” under the Controlled Substances Act.\\nAlthough we held in Sandoval that the term does not\\nencompass solicitation offenses, 866 F.3d at 989–90, the\\nanalysis for determining whether an offense qualifies as a\\n“drug trafficking crime” under the Controlled Substances\\nAct is different from the analysis for determining whether an\\noffense qualifies as a “controlled substance offense” under\\nthe Sentencing Guidelines. See Shumate, 329 F.3d at 1030\\nn.5. The Controlled Substances Act “neither mentions\\nsolicitation nor contains any broad catch-all provision that\\ncould even arguably be read to cover solicitation.” Leyva-\\nLicea v. INS, 187 F.3d 1147, 1150 (9th Cir. 1999). In\\ncontrast, although the commentary to § 4B1.2 does not\\nmention solicitation either, it does contain a catch-all term\\n(“include”) that we have interpreted to encompass\\nsolicitation. See Shumate, 329 F.3d at 1030. In this regard,\\nour decision in Sandoval is inapplicable to this case.\\n\\n B\\n\\n Crum asks us to reconsider our decision in Shumate on\\nthe basis of an argument that was not considered in that case.\\nCrum contends that Application Note 1 of § 4B1.2 lacks\\nlegal force because it is inconsistent with the text of the\\nguideline—an assertion that, if true, would preclude courts\\nfrom relying on the commentary to expand the definition of\\n“controlled substance offense” to include solicitation. See\\nStinson v. United States, 508 U.S. 36, 45–46 (1993). In\\nCrum’s view, because the plain text of § 4B1.2(b) does not\\nencompass solicitation (or any of the inchoate offenses\\ndiscussed in the commentary), the commentary may not\\nexpand the definition of “controlled substance offense” to\\ninclude those offenses.\\n\\n UNITED STATES V. CRUM 9\\n\\n Our sister circuits are split on this issue. The First, Third,\\nand Eleventh Circuits have held that the commentary is\\nconsistent with the text of § 4B1.2(b), as the commentary\\ndoes not include any offense that is explicitly excluded by\\nthe text of the guideline. United States v. Smith, 54 F.3d 690,\\n693 (11th Cir. 1995); United States v. Piper, 35 F.3d 611,\\n617 (1st Cir. 1994); United States v. Hightower, 25 F.3d 182,\\n187 (3d Cir. 1994). On the other side of the split, the Sixth\\nand D.C. Circuits have held that the commentary conflicts\\nwith the text of § 4B1.2(b). United States v. Havis, 927 F.3d\\n382, 385–87 (6th Cir. 2019) (en banc); United States v.\\nWinstead, 890 F.3d 1082, 1090–92 (D.C. Cir. 2018). The\\nD.C. Circuit explained that the text of § 4B1.2(b) provides a\\n“very detailed definition” of “controlled substance offense,”\\nwhich does not include the offenses listed in the\\ncommentary. Winstead, 927 F.3d at 1091 (internal quotation\\nmarks omitted). The court also pointed out that the\\nSentencing Commission included attempt offenses in\\n§ 4B1.2(a) when defining “crime of violence,” but chose not\\nto include such offenses in § 4B1.2(b) when defining\\n“controlled substance offense.” Id. Those drafting choices\\nsupport the conclusion that the definition of “controlled\\nsubstance offense” excludes attempt and the related offenses\\nlisted in the commentary. Id.\\n\\n If we were free to do so, we would follow the Sixth and\\nD.C. Circuits’ lead. In our view, the commentary\\nimproperly expands the definition of “controlled substance\\noffense” to include other offenses not listed in the text of the\\nguideline. Like the Sixth and D.C. Circuits, we are troubled\\nthat the Sentencing Commission has exercised its\\ninterpretive authority to expand the definition of “controlled\\nsubstance offense” in this way, without any grounding in the\\ntext of § 4B1.2(b) and without affording any opportunity for\\ncongressional review. See Havis, 927 F.3d at 386–87;\\n\\n10 UNITED STATES V. CRUM\\n\\nWinstead, 890 F.3d at 1092. This is especially concerning\\ngiven that the Commission’s interpretation will likely\\nincrease the sentencing ranges for numerous defendants\\nwhose prior convictions qualify as controlled substance\\noffenses due solely to Application Note 1.\\n\\n We are nonetheless compelled by our court’s prior\\ndecision in United States v. Vea-Gonzales, 999 F.2d 1326\\n(9th Cir. 1993), overruled on other grounds by Custis v.\\nUnited States, 511 U.S. 485 (1994), to reject the Sixth and\\nD.C. Circuits’ view. In Vea-Gonzales, we held that\\nApplication Note 1 of § 4B1.2 is “perfectly consistent” with\\nthe text of § 4B1.2(b). 999 F.2d at 1330. We explained that\\nthe text of § 4B1.2(b) defines the term “controlled substance\\noffense” as encompassing violations of laws prohibiting the\\nmanufacture, import, export, distribution, or dispensing of\\ndrugs, and that aiding and abetting, conspiring, and\\nattempting to commit such offenses constitute violations of\\nthose laws. Id. We thus concluded that Application Note 1\\nproperly interprets the definition of the term “controlled\\nsubstance offense” to encompass aiding and abetting,\\nconspiracy, attempt, and other forms of the underlying\\noffense. Id. No intervening higher authority is “clearly\\nirreconcilable” with the reasoning of Vea-Gonzales, so we\\ncannot overrule that precedent as a three-judge panel. See\\nUnited States v. Pepe, 895 F.3d 679, 685–86 (9th Cir. 2018).\\nAs a result, we are not free to depart from the holding in our\\nprior cases that the term “controlled substance offense” as\\ndefined in § 4B1.2(b) encompasses both solicitation and\\nattempt offenses. See Shumate, 329 F.3d at 1029–31; Vea-\\nGonzales, 999 F.2d at 1330.\\n\\n III\\n\\n We turn next to Crum’s argument that Oregon’s\\ndelivery-of-methamphetamine offense sweeps more broadly\\n\\n UNITED STATES V. CRUM 11\\n\\nthan the federal definition of “controlled substance offense”\\nbecause the Oregon offense criminalizes the mere offer to\\nsell methamphetamine.\\n\\n Crum’s argument turns on the Oregon Court of Appeals’\\ndecision in Pollock, which was issued after our court decided\\nShumate. In Pollock, the Oregon Court of Appeals held that\\nan individual can be convicted of delivery of a controlled\\nsubstance under Oregon law if he has offered to sell that\\nsubstance to another person. 73 P.3d at 300. In Crum’s\\nview, merely offering to sell a controlled substance does not\\nconstitute either soliciting or attempting to commit a\\n“controlled substance offense.” Thus, even if the definition\\nof “controlled substance offense” under § 4B1.2\\nencompasses solicitation and attempt, Crum argues that\\nOregon’s delivery-of-methamphetamine offense is still\\noverbroad.\\n\\n We reject Crum’s argument. As we noted in Sandoval,\\noffering to sell a controlled substance constitutes soliciting\\ndelivery of a controlled substance. 866 F.3d at 990–91\\n(discussing Pollock, among other Oregon cases); see also\\nUnited States v. Lee, 704 F.3d 785, 790 n.2 (9th Cir. 2012).\\nSolicitation does not fall within the definition of “drug\\ntrafficking crime” under the Controlled Substances Act,\\nwhich is the term we were construing in Sandoval. But\\nsolicitation does fall within the definition of “controlled\\nsubstance offense” under § 4B1.2 of the Sentencing\\nGuidelines. See Shumate, 329 F.3d at 1030. Thus, an offer\\nto sell a controlled substance under Oregon law is a\\ncategorical match for solicitation of a “controlled substance\\noffense” under § 4B1.2.\\n\\n * * *\\n\\n12 UNITED STATES V. CRUM\\n\\n In sum, Crum’s prior conviction for delivery of\\nmethamphetamine qualifies as a “controlled substance\\noffense,” as that term is defined in § 4B1.2 of the Guidelines.\\nThe district court should therefore have applied a base\\noffense level of 20 rather than 14. See U.S.S.G.\\n§ 2K2.1(a)(4)(A). We vacate Crum’s sentence and remand\\nfor resentencing.\\n\\n VACATED and REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4074,"cells":{"id":{"kind":"number","value":4649093,"string":"4,649,093"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Josephine Tijerino v. Stetson Desert Project, LLC"},"decision_date":{"kind":"string","value":"2019-08-16"},"docket_number":{"kind":"string","value":"18-16013"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Ferdinand F. Fernandez, A. Wallace Tashima, and John B. Owens, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"TASHIMA, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nJOSEPHINE TIJERINO; TAMICKA No. 18-16013\\nTOLIVER; JANE ROE DANCER,\\n Plaintiffs-Appellants, D.C. Nos.\\n CV 15-2563-SMM\\n v. 15-2564-SMM\\n 16-0408-SMM\\nSTETSON DESERT PROJECT, LLC,\\nDBA Lé Girls Cabaret; CORY J.\\nANDERSON; CARY ANDERSON, OPINION\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Stephen M. McNamee, District Judge, Presiding\\n\\n Argued and Submitted July 9, 2019\\n Portland, Oregon\\n\\n Filed August 16, 2019\\n\\n Before: Ferdinand F. Fernandez, A. Wallace Tashima,\\n and John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Tashima\\n\\f2 TIJERINO V. STETSON DESERT PROJECT\\n\\n SUMMARY*\\n\\n\\n Labor Law\\n\\n The panel reversed the district court’s dismissal for lack\\nof subject-matter jurisdiction of an action brought under the\\nFair Labor Standards Act and Arizona state law by a group of\\nexotic dancers against the club at which they worked.\\n\\n The district court held that it lacked jurisdiction because\\nthe dancers did not prove at the outset of the case that they\\nwere employees rather than independent contractors.\\nReversing, the panel held that the statutory requirement that\\nplaintiffs must be employees as defined in the FLSA is a\\nmerits-based determination, not a jurisdictional limitation.\\nFurther, the dancers’ complaints sufficiently alleged\\nsubstantial issues of federal law. The panel remanded the\\ncase for further proceedings.\\n\\n\\n COUNSEL\\n\\nAndrew Sterling (argued) and Michael J. Rusing, Rusing\\nLopez & Lizardi PLLC, Tucson, Arizona, for Plaintiff-\\nAppellant Jane Roe Dancer.\\n\\nClifford P. Bendau II, The Bendau Law Firm PLLC, Phoenix,\\nArizona, for Plaintiffs-Appellants Josephine Tijerino and\\nTamika Toliver.\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f TIJERINO V. STETSON DESERT PROJECT 3\\n\\nLaurent R.G. Badoux (argued), Buchalter APC, Scottsdale,\\nArizona, for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nTASHIMA, Circuit Judge:\\n\\n These are actions under the Fair Labor Standards Act\\n(“FLSA”) brought by a group of exotic dancers against the\\nclub at which they worked, for asserted violations of the\\nFLSA and Arizona state law. The district court dismissed the\\nsuit with prejudice for lack of subject-matter jurisdiction\\nbecause the Dancers did not prove at the outset of the case\\nthat they were employees rather than independent contractors.\\nAs we explain below, the district court erred in reaching that\\ndecision. We reverse and remand for further proceedings.\\n\\n BACKGROUND\\n\\n Plaintiffs-Appellants Josephine Tijerino, Tamicka\\nToliver, and Jane Roe Dancer (collectively, the “Dancers”) all\\nworked as exotic dancers at Lé Girls Gentlemen’s Club, a\\nstrip club in Phoenix, Arizona. The club is owned and\\noperated by Defendants-Appellees Stetson Desert Project,\\nLLC, dba Lé Girls Cabaret, Cory Anderson, and Cary\\nAnderson (collectively, the “Club”). Initially, each dancer\\nfiled a separate complaint against the Club alleging violations\\nof employment law. The Dancers all allege that the Club\\nfailed to pay them any wages in violation of the FLSA and\\nwillfully failed to pay them at the rate of Arizona’s minimum\\nwage in violation of the Arizona Minimum Wage Act. See\\n29 U.S.C. § 206(a); Ariz. Rev. Stat. § 23-363. Separately,\\nTijerino and Toliver allege that the Club failed to pay them\\n\\f4 TIJERINO V. STETSON DESERT PROJECT\\n\\novertime wages in violation of the FLSA and willfully failed\\nto pay their wages for labor in violation of the Arizona Wage\\nLaw. See 29 U.S.C § 207; Ariz. Rev. Stat. § 23-351. Jane\\nRoe Dancer separately alleges that the Club was unjustly\\nenriched in violation of Ariz. Rev. Stat. § 23-202.\\n\\n Jane Roe Dancer filed her complaint as a collective action\\non behalf of herself and all other similarly situated\\nindividuals who worked at the Club as dancers. See\\n29 U.S.C. § 216(b). She filed a motion to consolidate the\\nDancers’ cases, which the district court granted. She also\\nfiled a motion for conditional certification and to authorize\\nnotice to putative class members, and she included a\\ndeclaration in which she provided information suggesting that\\nthe exotic dancers who worked at the Club were similarly\\nsituated for purposes of the FLSA. Shortly after the cases\\nwere consolidated, the parties submitted a proposed joint case\\nmanagement order in which they agreed that “[t]he basis for\\nthe court’s subject-matter jurisdiction over plaintiff’s claims\\nis the Fair Labor Standards Act, 29 U.S.C. § 201 et seq. . . .”\\n\\n The district court denied the Dancers’ motion to\\nconditionally certify an opt-in class and dismissed the case\\nwith prejudice for lack of subject-matter jurisdiction. First,\\nin denying the motion to conditionally certify a FLSA opt-in\\nclass, the district court called such a motion “premature.” It\\nheld that Jane Roe “must satisfy her initial burden of alleging\\nspecific facts that permit an inference that she is an\\n‘employee’ (and the Defendants her ‘employer’) within the\\nmeaning of the FLSA.” It also noted that the Dancers’ status\\nas employees or independent contractors of the Club was\\n“unsettled,” and ordered the Dancers to file a brief “on\\nwhether Plaintiff and her putative class are employees within\\nthe meaning of FLSA under the economic realities test.”\\n\\f TIJERINO V. STETSON DESERT PROJECT 5\\n\\n The district court next denied the Dancers’ renewed\\nmotion for conditional class certification and held that the\\nDancers were independent contractors, not employees. It\\nheld that § 216(b)’s requirement that employees bring actions\\nunder the statute constituted an “antecedent issue” of\\njurisdiction and thus that, if the Dancers were not employees\\nunder the FLSA, there could be no violation of the FLSA and\\ntherefore no federal question jurisdiction. The district court\\nproceeded to evaluate the Dancers’ employment status under\\nthe FLSA’s economic realities test and declared them to be\\nindependent contractors. It then denied the Dancers’ motion\\nto conditionally certify a FLSA opt-in class based on a lack\\nof subject-matter jurisdiction.1\\n\\n Finally, the district court dismissed all three plaintiffs’\\ncases with prejudice for lack of subject-matter jurisdiction.2\\nThe district court “reiterate[d] its position that employee\\nstatus under the FLSA is an antecedent jurisdictional issue,”\\nand that plaintiffs had not “come forward with any evidence\\nfrom which the Court could find a reasonable indicia of\\nemployment.”\\n\\n The Dancers appeal from the final judgment, the order\\ndismissing the cases with prejudice for lack of subject-matter\\njurisdiction, and the order denying their renewed motion for\\nconditional certification of an FLSA opt-in class.\\n\\n 1\\n The district court later denied the Dancers’ motion to reconsider this\\norder.\\n 2\\n Although we need not examine the issue in detail because we\\nreverse the district court’s dismissal for other reasons, we note that, in\\ngeneral, dismissal for lack of subject matter jurisdiction should be without\\nprejudice. See Mo. ex rel. Koster v. Harris, 847 F.3d 646, 656 (9th Cir.\\n2017).\\n\\f6 TIJERINO V. STETSON DESERT PROJECT\\n\\nAPPELLATE JURISDICTION AND STANDARD OF\\n REVIEW\\n\\n We have appellate jurisdiction under 28 U.S.C. § 1291.\\nWe review de novo a dismissal for lack of subject-matter\\njurisdiction. Leeson v. Transamerica Disability Income Plan,\\n671 F.3d 969, 974 (9th Cir. 2012). A district court’s\\ninterpretation of federal law, such as the FLSA, is also\\nreviewed de novo. Id.\\n\\n DISCUSSION\\n\\nI. The district court erred in concluding that it lacked\\n subject-matter jurisdiction.\\n\\n Congress granted federal district courts “original\\njurisdiction of all civil actions arising under the Constitution,\\nlaws, or treaties of the United States.” 28 U.S.C. § 1331.\\nThe Supreme Court has held that “Congress has given the\\nlower federal courts jurisdiction to hear . . . only those cases\\nin which a well-pleaded complaint establishes either that\\nfederal law creates the cause of action or that the plaintiff’s\\nright to relief necessarily depends on resolution of a\\nsubstantial question of federal law.” Franchise Tax Bd. v.\\nConstr. Laborers Vacation Trust for S. Cal., 463 U.S. 1,\\n27–28 (1983). “[A] federally created claim for relief is\\ngenerally a ‘sufficient condition for federal-question\\njurisdiction.’” Mims v. Arrow Fin. Servs., LLC, 565 U.S. 368,\\n377 (2012) (quoting Grable & Sons Metal Prods., Inc. v.\\nDarue Eng’g & Mfg., 545 U.S. 308, 317 (2005)). The FLSA\\nprovides such a claim for aggrieved employees to sue to\\nrecover unpaid minimum wages or unpaid overtime\\ncompensation, either individually or collectively. 29 U.S.C.\\n§ 216(b).\\n\\f TIJERINO V. STETSON DESERT PROJECT 7\\n\\n In order to determine whether the district court had\\nsubject-matter jurisdiction over the Dancers’ claims, we ask\\ntwo questions. First, we ask whether the employment status\\nprovision—the statutory requirement that plaintiffs must be\\nemployees as defined in the FLSA, 20 U.S.C. § 203(e), in\\norder to prevail in a § 216(b) action—is a jurisdictional\\nlimitation on the statute’s coverage such that it prevents the\\nDancers from moving forward with their claims. Second, we\\nask whether the Dancers’ complaints survive the well-pleaded\\ncomplaint rule to invoke federal question subject-matter\\njurisdiction.\\n\\n A. The Dancers’ employment status is a merits-based\\n determination, not an antecedent jurisdictional\\n issue.\\n\\n If the employment status provision of the statute is\\njurisdictional, then district courts would be required to\\nanalyze whether plaintiffs are employees under the FLSA at\\nthe outset of the litigation under Federal Rule of Civil\\nProcedure 12(b)(1), as the district court did in this case. But\\nif the employment status provision is an essential ingredient\\nrelated to the merits of a claim, then whether plaintiffs are\\nemployees should be decided at summary judgment or trial,\\nassuming the complaint sufficiently alleges that plaintiffs are\\nemployees.\\n\\n In determining whether a statutory provision constitutes\\na jurisdictional limitation or an “essential ingredient” of a\\nclaim, the Supreme Court has set forth a bright-line test:\\n\\n If the Legislature clearly states that a\\n threshold limitation on a statute’s scope shall\\n count as jurisdictional, then courts and\\n\\f8 TIJERINO V. STETSON DESERT PROJECT\\n\\n litigants will be duly instructed and will not\\n left to wrestle with the issue. But when\\n Congress does not rank a statutory limitation\\n on coverage as jurisdictional, courts should\\n treat the restriction as nonjurisdictional in\\n character.\\n\\nArbaugh v. Y & H Corp., 546 U.S. 500, 515–16 (2006)\\n(internal citation omitted). A limitation on a statute’s scope\\ncounts as a jurisdictional limitation only if Congress clearly\\n“intended” it to be jurisdictional such that courts should raise\\nthe issue “on their own motion.” Id. at 514. Otherwise courts\\nshould treat the statutory threshold as nonjurisdictional. Id.\\n\\n In Arbaugh, a restaurant challenged its former waitress’\\nTitle VII sexual harassment claim after the close of trial,\\narguing that the district court lacked subject-matter\\njurisdiction over the case because Title VII’s statutory\\ndefinition of “employer” did not include restaurants of its\\nsize. Id. at 504 (“Y & H asserted that it had fewer than\\n15 employees on its payroll and therefore was not amenable\\nto suit under Title VII.”). Noting the tendency of courts to\\nconflate subject-matter jurisdiction with the “plaintiff’s need\\nand ability to prove the defendant bound by the federal law\\nasserted as the predicate for relief—a merits-related\\ndetermination,” the Supreme Court set out to clarify this\\n“dichotomy.” Id. at 511 (emphasis added). The Court held\\nthat, “if subject-matter jurisdiction turns on contested facts,\\nthe trial judge may be authorized to review the evidence and\\nresolve the dispute on her own. If satisfaction of an essential\\nelement of a claim for relief is at issue, however, the jury is\\nthe proper trier of contested facts.” Id. at 514 (internal\\ncitations omitted).\\n\\f TIJERINO V. STETSON DESERT PROJECT 9\\n\\n Subsequently, in Reed Elsevier, Inc. v. Muchnick,\\n559 U.S. 154 (2010), the Court reasoned that courts should\\ndetermine whether a statutory provision is jurisdictional by\\nlooking at the provision’s “text and structure,” as well as\\n“context, and relevant historical treatment.” Id. at 162,\\n166–68 (applying the Arbaugh test to hold that the Copyright\\nAct’s registration requirement was not a jurisdictional\\nprecondition to filing an infringement claim). The Ninth\\nCircuit has applied Arbaugh and Reed Elsevier to\\nemployment-related statutory provisions. See Day v. AT & T\\nDisability Income Plan, 698 F.3d 1091, 1099–1100 (9th Cir.\\n2012); Leeson, 671 F.3d at 978. A straightforward\\nframework has emerged from this line of cases: when\\ndetermining whether a statutory threshold is a jurisdictional\\nbarrier, we ask whether Congress clearly labeled the statutory\\nrequirement as jurisdictional and whether the requirement is\\nlocated in a jurisdiction-granting provision; if so, Congress\\nlikely intended the provision to be jurisdictional. See Reed\\nElsevier, 559 U.S. at 160–69; Day, 698 F.3d at 1099–1100;\\nLeeson, 671 F.3d at 970–79. Also helpful to our analysis is\\nwhether courts have historically treated the requirement as\\njurisdictional absent an express designation from Congress.\\nReed Elsevier, 559 U.S. aat 169. In this Circuit, we also\\nensure that “no other reasons necessitate[] that the provision\\nbe construed as jurisdictional.” Leeson, 671 F.3d at 977. All\\nof these factors help guide our analysis of the employment\\nstatus provision in the FLSA.\\n\\n Congress did not clearly label the FLSA employment\\nstatus provision as jurisdictional. “As in all cases involving\\nstatutory construction, our starting point must be the language\\nemployed by Congress.” Am. Tobacco Co. v. Patterson,\\n456 U.S. 63, 68 (1982) (internal quotation marks omitted).\\nHere, the relevant provision of FLSA § 216(b) states:\\n\\f10 TIJERINO V. STETSON DESERT PROJECT\\n\\n Any employer who violates the provisions of\\n section 206 or section 207 of this title shall be\\n liable to the employee or employees affected\\n in the amount of their unpaid minimum\\n wages, or their unpaid overtime\\n compensation, as the case may be, and in an\\n additional equal amount as liquidated\\n damages . . . . An action to recover the\\n liability prescribed in the preceding sentences\\n may be maintained against any employer\\n (including a public agency) in any Federal or\\n State court of competent jurisdiction by any\\n one or more employees for and in behalf of\\n himself or themselves and other employees\\n similarly situated.\\n\\n29 U.S.C. § 216(b).3 Section 216(b)’s requirement that\\nemployers be liable to employees for violations of the FLSA\\nis not clearly labeled as jurisdictional. In fact, there is not\\neven a mention of jurisdiction in this section (besides a\\ngeneral reference to a “court of competent jurisdiction”), and\\nthere is no indication whatsoever that employee status is a\\njurisdictional limitation. Compare this to the statute in Reed\\nElsevier, in which even “the presence of the word\\n“jurisdiction” in the last sentence of [the statutory provision]”\\ndid not render that requirement jurisdictional. Reed Elsevier,\\n559 U.S. at 163. In the “Definitions” section of the FLSA,\\nCongress defined the term “employee” to mean “any\\nindividual employed by an employer,” while also setting\\naside exceptions to the term “employee” for certain\\n\\n\\n 3\\n 29 U.S.C. § 206 requires that non-exempt employees be paid the\\nminimum wage and § 207 requires that non-exempt employees be paid\\novertime wages for hours worked in excess of forty hours in one week.\\n\\f TIJERINO V. STETSON DESERT PROJECT 11\\n\\nindividuals employed by governmental entities, individuals\\nwho volunteer to perform services for governmental entities,\\nand immediate family members of employers employed in\\nagriculture. 29 U.S.C. § 203(e)(1)–(4)(A). Here too, there is\\nno indication that the criterion of a plaintiff being an\\n“employee” is in any way related to subject matter\\njurisdiction, let alone clearly labeled as jurisdictional.\\n\\n Section 203(e) and § 216(b) are not located in\\njurisdiction-granting provisions of the statute. In Arbaugh,\\nthe Court concluded that Congress did not intend that Title\\nVII’s definition of the term “employer” be jurisdictional\\nbecause Congress placed the term’s definition in a section\\ntitled “Definitions”—“a separate provision that ‘does not\\nspeak in jurisdictional terms or refer in any way to the\\njurisdiction of the district courts.’” Arbaugh, 546 U.S. at 515\\n(internal citation omitted); see also Leeson, 671 F.3d\\nat 978–79 (Plaintiff’s status as a plan “participant,” for\\npurposes of the Employee Retirement Income Security Act,\\nwas a substantive element of his claim—not a prerequisite for\\nsubject-matter jurisdiction—because the definition of\\n“participant” appears in a provision that “does not speak in\\njurisdictional terms.”). Here, Congress also placed § 203(e),\\nwhich defines the term “employee,” in a section titled\\n“Definitions,” and § 216(b) in a section titled “Penalties,”\\nneither of which speaks to jurisdiction. See 29 U.S.C.\\n§§ 203(e), 216(b); see also Day, 698 F.3d at 1099–1100\\n(statute’s minimum age requirement for plaintiffs alleging\\ndiscrimination claims “appear[ed] in a separate provision that\\nd[id] not speak in jurisdictional terms” and therefore was a\\nsubstantive element of a claim rather than a jurisdictional\\nlimitation). Congress did use jurisdictional language in other\\nsections of the FLSA, which suggests that Congress clearly\\ndid not intend the requirements of § 216(b) and § 203(e) to be\\n\\f12 TIJERINO V. STETSON DESERT PROJECT\\n\\njurisdictional. See 29 U.S.C. § 210(a) (stating that upon the\\nfiling of a petition for judicial review of a § 208 order in a\\nspecific court of appeals, “such court shall have exclusive\\njurisdiction to affirm, modify, . . . or set aside such order”).\\n\\n Furthermore, courts have not historically treated § 203(e)\\nand § 216(b)’s employment status provisions as\\njurisdictional. Although the Court in Reed Elsevier found\\nprevious treatment of the Copyright Act’s registration\\nrequirement as jurisdictional, it still concluded “that\\n§ 411(a)’s registration requirement [was] nonjurisdictional,\\nnotwithstanding its prior jurisdictional treatment.” 559 U.S.\\nat 169. Here, courts have not historically treated the\\nemployment status provision as jurisdictional; in similar\\ncases, both the Fourth and Fifth Circuits have treated this\\nquestion as nonjurisdictional. See McFeeley v. Jackson St.\\nEntm’t LLC, 825 F.3d 235 (4th Cir. 2016); Reich v. Circle C.\\nInvs., Inc., 998 F.2d 324 (5th Cir. 1993). Thus, all factors\\nweigh in favor of construing the statute’s employment status\\nprovision as nonjurisdictional.\\n\\n There is no support for the district court’s theory that the\\nemployment status provision constitutes an antecedent\\njurisdictional issue, and we can find no reasons that\\n“necessitate[] that the provision be construed as\\njurisdictional.” Leeson, 671 F.3d at 977. The district court\\nattempted to distinguish cases like Leeson and Day, writing\\nthat “the controlling cases upon which Plaintiff relie[d] to\\nrefute the Court’s decision that employee status is an\\nantecedent issue [were] specific to Employee Retirement\\nIncome Security Act and Title VII of the Civil Rights Act\\ncases – not FLSA cases.” But, aside from this conclusory\\nstatement, the district court did not provide any further\\nanalysis. It provided no rationale as to why Arbaugh’s\\n\\f TIJERINO V. STETSON DESERT PROJECT 13\\n\\nframework is not controlling, other than to suggest that\\nArbaugh does not control merely because that case involved\\na different statute than the statute in this case. Instead, the\\ndistrict court cited Rutherford Food Corp. v. McComb,\\n331 U.S. 722 (1947), in support of its position that the\\nexercise of jurisdiction over a FLSA complaint turned on\\nwhether the plaintiff was an employee. Id. at 727 (“Whether\\nor not the acts charged in this complaint violate the [FLSA]\\ndepends, so far as the meat boners are concerned, upon a\\ndetermination as to whether either of both respondents are\\nemployers of the boners.”). The district court, however,\\noverlooked that Rutherford was a decision reached on the\\nmerits, not on jurisdictional grounds. See id.4\\n\\n In summary, under the Arbaugh test, it is clear that the\\nFLSA’s employment status provision in § 203(e) and\\n§ 216(b) should be construed as nonjurisdictional. Our\\nconclusion is also supported by applications of this test to\\nother employment-related statutory provisions. See Day,\\n698 F.3d at 1099–1100; Leeson, 671 F.3d at 979. Most\\nimportantly, Congress did not state that the statute’s\\nemployment status provision was jurisdictional, and § 203(e)\\nand § 216(b) are not located in jurisdiction-granting\\nprovisions of the statute. Finally, courts have not historically\\ntreated the employment status provision as jurisdictional.\\nThe term “employee” in the FLSA “serves to identify those\\nplaintiffs who may be entitled to relief, not to limit the\\n\\n\\n\\n 4\\n The district court also cited to Dellinger v. Sci. Applications Int’l\\nCorp., in support of this proposition. Dellinger, however, is\\ndistinguishable in that the plaintiff was never hired by nor performed any\\nwork for the defendant: a prospective employer. See 649 F.3d 226, 228\\n(4th Cir. 2011).\\n\\f14 TIJERINO V. STETSON DESERT PROJECT\\n\\nauthority of federal courts to adjudicate claims” under the\\nFLSA. Id. at 978.\\n\\n B. The allegations were not so patently without merit\\n as to justify dismissal for lack of subject-matter\\n jurisdiction.\\n\\n Having held that the FLSA’s employment status provision\\nin § 203(e) and § 216(b) is nonjurisdictional, we now turn to\\nthe Dancers’ complaints to ensure that they allege a\\nsubstantial issue of federal law that is sufficient to invoke\\nsubject-matter jurisdiction. A complaint for relief properly\\ninvokes federal jurisdiction where its well-pleaded allegations\\nestablish “either that federal law creates the cause of action or\\nthat the plaintiff’s right to relief necessarily depends on\\nresolution of a substantial question of federal law.”\\nFranchise Tax Bd., 463 U.S. at 27–28. There is no federal\\nquestion jurisdiction where the federal claim asserted is “too\\ninsubstantial,” Cook v. Peter Kiewit Sons Co., 775 F.2d 1030,\\n1035 (9th Cir. 1985), defined as so patently without merit that\\nthe claim requires no meaningful consideration. See Yokeno\\nv. Mafnas, 973 F.2d 803, 808 (9th Cir. 1992); see also\\nCement Masons Health & Welfare Trust Fund for N. Cal. v.\\nStone, 197 F.3d 1003, 1008 (9th Cir. 1999) (“Any non-\\nfrivolous assertion of a federal claim suffices to establish\\nfederal question jurisdiction, even if that claim is later\\ndismissed on the merits.”).\\n\\n Here, the Dancers’ complaints facially allege substantial\\nissues of federal law: whether the Dancers could collectively\\nrecover damages under § 216(b) for the Club’s failure to pay\\nthem at the required minimum wage and at the required\\novertime rate, for hours worked in excess of 40 per week, in\\nviolation of § 206 and § 207 of the FLSA. Moreover, the\\n\\f TIJERINO V. STETSON DESERT PROJECT 15\\n\\nDancers allege numerous facts supporting their claims that\\nthe Club owes them wages because they are employees under\\nthe FLSA, and relatedly, under Arizona state law. The claims\\nin the Dancers’ complaints are not “wholly insubstantial and\\nfrivolous,” and are not so patently without merit as to warrant\\ndismissal for lack of subject-matter jurisdiction. Arbaugh,\\n546 U.S. at 513 n.10. Therefore, there is federal question\\nsubject-matter jurisdiction to hear the Dancers’ § 216(b)\\nclaims pursuant to 28 U.S.C. § 1331, and there is\\nsupplemental jurisdiction over the state law claims.\\n\\nII. The Dancers’ other arguments on appeal\\n\\n The Dancers make two additional arguments on appeal.\\nFirst, they ask us to remand with instructions to grant the\\nDancers’ motion for conditional class certification under the\\nFLSA. The district court did not reach a merits-based\\nanalysis of conditional certification because it erroneously\\ndetermined that it lacked subject-matter jurisdiction over the\\ncase. Our holding that employee status under the FLSA is\\nnonjurisdictional necessarily reverses the reasoning of the\\ndistrict court’s denial of conditional certification. On\\nremand, the district court will have the opportunity to\\ndetermine whether to conditionally certify an FLSA\\ncollective action in accordance with our precedent. See\\nCampbell v. City of Los Angeles, 903 F.3d 1090, 1100 (9th\\nCir. 2018) (discussing the “near-universal practice” of the\\ntwo-step approach to conditional certification under\\n§ 216(b)). We deny the Dancers’ request to remand with\\nspecific instructions to grant the Dancers’ motion for\\nconditional certification.\\n\\n Second, the Dancers ask us to remand with instructions\\n“to adjudicate the merits of the Dancers’ claims in accordance\\n\\f16 TIJERINO V. STETSON DESERT PROJECT\\n\\nwith well-established case law applying the FLSA’s\\neconomic realities test.” The district court applied the\\neconomic realities test in reaching its conclusion that it lacked\\nsubject-matter jurisdiction. Because we reverse the district\\ncourt’s conclusion that it lacked subject-matter jurisdiction,\\non remand, it will have the opportunity to address the\\neconomic realities test when it addresses the merits of the\\nDancers’ FLSA claims. See Real v. Driscoll Strawberry\\nAssocs., Inc., 603 F.2d 748, 750 (9th Cir. 1979) (evaluating\\nthe employment status of the plaintiffs on summary\\njudgment). We doubt that the experienced district court is in\\nneed of any further instructions on remand.\\n\\n CONCLUSION\\n\\n For the reasons set forth above, we reverse the district\\ncourt’s dismissal of these actions for lack of subject-matter\\njurisdiction, and remand for further proceedings consistent\\nwith this opinion.\\n\\n REVERSED and REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426346/\", \"author_raw\": \"TASHIMA, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"FERDINAND F FERNANDEZ"},"judge2":{"kind":"string","value":"A WALLACE TASHIMA"},"judge3":{"kind":"string","value":"JOHN B OWENS"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649093/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Josephine Tijerino v. Stetson Desert Project, LLC\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-16\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nJOSEPHINE TIJERINO; TAMICKA No. 18-16013\nTOLIVER; JANE ROE DANCER,\n Plaintiffs-Appellants, D.C. Nos.\n CV 15-2563-SMM\n v. 15-2564-SMM\n 16-0408-SMM\nSTETSON DESERT PROJECT, LLC,\nDBA Lé Girls Cabaret; CORY J.\nANDERSON; CARY ANDERSON, OPINION\n Defendants-Appellees.\n\n\n Appeal from the United States District Court\n for the District of Arizona\n Stephen M. McNamee, District Judge, Presiding\n\n Argued and Submitted July 9, 2019\n Portland, Oregon\n\n Filed August 16, 2019\n\n Before: Ferdinand F. Fernandez, A. Wallace Tashima,\n and John B. Owens, Circuit Judges.\n\n Opinion by Judge Tashima\n\f2 TIJERINO V. STETSON DESERT PROJECT\n\n SUMMARY*\n\n\n Labor Law\n\n The panel reversed the district court’s dismissal for lack\nof subject-matter jurisdiction of an action brought under the\nFair Labor Standards Act and Arizona state law by a group of\nexotic dancers against the club at which they worked.\n\n The district court held that it lacked jurisdiction because\nthe dancers did not prove at the outset of the case that they\nwere employees rather than independent contractors.\nReversing, the panel held that the statutory requirement that\nplaintiffs must be employees as defined in the FLSA is a\nmerits-based determination, not a jurisdictional limitation.\nFurther, the dancers’ complaints sufficiently alleged\nsubstantial issues of federal law. The panel remanded the\ncase for further proceedings.\n\n\n COUNSEL\n\nAndrew Sterling (argued) and Michael J. Rusing, Rusing\nLopez & Lizardi PLLC, Tucson, Arizona, for Plaintiff-\nAppellant Jane Roe Dancer.\n\nClifford P. Bendau II, The Bendau Law Firm PLLC, Phoenix,\nArizona, for Plaintiffs-Appellants Josephine Tijerino and\nTamika Toliver.\n\n\n *\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f TIJERINO V. STETSON DESERT PROJECT 3\n\nLaurent R.G. Badoux (argued), Buchalter APC, Scottsdale,\nArizona, for Defendants-Appellees.\n\n\n OPINION\n\nTASHIMA, Circuit Judge:\n\n These are actions under the Fair Labor Standards Act\n(“FLSA”) brought by a group of exotic dancers against the\nclub at which they worked, for asserted violations of the\nFLSA and Arizona state law. The district court dismissed the\nsuit with prejudice for lack of subject-matter jurisdiction\nbecause the Dancers did not prove at the outset of the case\nthat they were employees rather than independent contractors.\nAs we explain below, the district court erred in reaching that\ndecision. We reverse and remand for further proceedings.\n\n BACKGROUND\n\n Plaintiffs-Appellants Josephine Tijerino, Tamicka\nToliver, and Jane Roe Dancer (collectively, the “Dancers”) all\nworked as exotic dancers at Lé Girls Gentlemen’s Club, a\nstrip club in Phoenix, Arizona. The club is owned and\noperated by Defendants-Appellees Stetson Desert Project,\nLLC, dba Lé Girls Cabaret, Cory Anderson, and Cary\nAnderson (collectively, the “Club”). Initially, each dancer\nfiled a separate complaint against the Club alleging violations\nof employment law. The Dancers all allege that the Club\nfailed to pay them any wages in violation of the FLSA and\nwillfully failed to pay them at the rate of Arizona’s minimum\nwage in violation of the Arizona Minimum Wage Act. See\n29 U.S.C. § 206(a); Ariz. Rev. Stat. § 23-363. Separately,\nTijerino and Toliver allege that the Club failed to pay them\n\f4 TIJERINO V. STETSON DESERT PROJECT\n\novertime wages in violation of the FLSA and willfully failed\nto pay their wages for labor in violation of the Arizona Wage\nLaw. See 29 U.S.C § 207; Ariz. Rev. Stat. § 23-351. Jane\nRoe Dancer separately alleges that the Club was unjustly\nenriched in violation of Ariz. Rev. Stat. § 23-202.\n\n Jane Roe Dancer filed her complaint as a collective action\non behalf of herself and all other similarly situated\nindividuals who worked at the Club as dancers. See\n29 U.S.C. § 216(b). She filed a motion to consolidate the\nDancers’ cases, which the district court granted. She also\nfiled a motion for conditional certification and to authorize\nnotice to putative class members, and she included a\ndeclaration in which she provided information suggesting that\nthe exotic dancers who worked at the Club were similarly\nsituated for purposes of the FLSA. Shortly after the cases\nwere consolidated, the parties submitted a proposed joint case\nmanagement order in which they agreed that “[t]he basis for\nthe court’s subject-matter jurisdiction over plaintiff’s claims\nis the Fair Labor Standards Act, 29 U.S.C. § 201 et seq. . . .”\n\n The district court denied the Dancers’ motion to\nconditionally certify an opt-in class and dismissed the case\nwith prejudice for lack of subject-matter jurisdiction. First,\nin denying the motion to conditionally certify a FLSA opt-in\nclass, the district court called such a motion “premature.” It\nheld that Jane Roe “must satisfy her initial burden of alleging\nspecific facts that permit an inference that she is an\n‘employee’ (and the Defendants her ‘employer’) within the\nmeaning of the FLSA.” It also noted that the Dancers’ status\nas employees or independent contractors of the Club was\n“unsettled,” and ordered the Dancers to file a brief “on\nwhether Plaintiff and her putative class are employees within\nthe meaning of FLSA under the economic realities test.”\n\f TIJERINO V. STETSON DESERT PROJECT 5\n\n The district court next denied the Dancers’ renewed\nmotion for conditional class certification and held that the\nDancers were independent contractors, not employees. It\nheld that § 216(b)’s requirement that employees bring actions\nunder the statute constituted an “antecedent issue” of\njurisdiction and thus that, if the Dancers were not employees\nunder the FLSA, there could be no violation of the FLSA and\ntherefore no federal question jurisdiction. The district court\nproceeded to evaluate the Dancers’ employment status under\nthe FLSA’s economic realities test and declared them to be\nindependent contractors. It then denied the Dancers’ motion\nto conditionally certify a FLSA opt-in class based on a lack\nof subject-matter jurisdiction.1\n\n Finally, the district court dismissed all three plaintiffs’\ncases with prejudice for lack of subject-matter jurisdiction.2\nThe district court “reiterate[d] its position that employee\nstatus under the FLSA is an antecedent jurisdictional issue,”\nand that plaintiffs had not “come forward with any evidence\nfrom which the Court could find a reasonable indicia of\nemployment.”\n\n The Dancers appeal from the final judgment, the order\ndismissing the cases with prejudice for lack of subject-matter\njurisdiction, and the order denying their renewed motion for\nconditional certification of an FLSA opt-in class.\n\n 1\n The district court later denied the Dancers’ motion to reconsider this\norder.\n 2\n Although we need not examine the issue in detail because we\nreverse the district court’s dismissal for other reasons, we note that, in\ngeneral, dismissal for lack of subject matter jurisdiction should be without\nprejudice. See Mo. ex rel. Koster v. Harris, 847 F.3d 646, 656 (9th Cir.\n2017).\n\f6 TIJERINO V. STETSON DESERT PROJECT\n\nAPPELLATE JURISDICTION AND STANDARD OF\n REVIEW\n\n We have appellate jurisdiction under 28 U.S.C. § 1291.\nWe review de novo a dismissal for lack of subject-matter\njurisdiction. Leeson v. Transamerica Disability Income Plan,\n671 F.3d 969, 974 (9th Cir. 2012). A district court’s\ninterpretation of federal law, such as the FLSA, is also\nreviewed de novo. Id.\n\n DISCUSSION\n\nI. The district court erred in concluding that it lacked\n subject-matter jurisdiction.\n\n Congress granted federal district courts “original\njurisdiction of all civil actions arising under the Constitution,\nlaws, or treaties of the United States.” 28 U.S.C. § 1331.\nThe Supreme Court has held that “Congress has given the\nlower federal courts jurisdiction to hear . . . only those cases\nin which a well-pleaded complaint establishes either that\nfederal law creates the cause of action or that the plaintiff’s\nright to relief necessarily depends on resolution of a\nsubstantial question of federal law.” Franchise Tax Bd. v.\nConstr. Laborers Vacation Trust for S. Cal., 463 U.S. 1,\n27–28 (1983). “[A] federally created claim for relief is\ngenerally a ‘sufficient condition for federal-question\njurisdiction.’” Mims v. Arrow Fin. Servs., LLC, 565 U.S. 368,\n377 (2012) (quoting Grable & Sons Metal Prods., Inc. v.\nDarue Eng’g & Mfg., 545 U.S. 308, 317 (2005)). The FLSA\nprovides such a claim for aggrieved employees to sue to\nrecover unpaid minimum wages or unpaid overtime\ncompensation, either individually or collectively. 29 U.S.C.\n§ 216(b).\n\f TIJERINO V. STETSON DESERT PROJECT 7\n\n In order to determine whether the district court had\nsubject-matter jurisdiction over the Dancers’ claims, we ask\ntwo questions. First, we ask whether the employment status\nprovision—the statutory requirement that plaintiffs must be\nemployees as defined in the FLSA, 20 U.S.C. § 203(e), in\norder to prevail in a § 216(b) action—is a jurisdictional\nlimitation on the statute’s coverage such that it prevents the\nDancers from moving forward with their claims. Second, we\nask whether the Dancers’ complaints survive the well-pleaded\ncomplaint rule to invoke federal question subject-matter\njurisdiction.\n\n A. The Dancers’ employment status is a merits-based\n determination, not an antecedent jurisdictional\n issue.\n\n If the employment status provision of the statute is\njurisdictional, then district courts would be required to\nanalyze whether plaintiffs are employees under the FLSA at\nthe outset of the litigation under Federal Rule of Civil\nProcedure 12(b)(1), as the district court did in this case. But\nif the employment status provision is an essential ingredient\nrelated to the merits of a claim, then whether plaintiffs are\nemployees should be decided at summary judgment or trial,\nassuming the complaint sufficiently alleges that plaintiffs are\nemployees.\n\n In determining whether a statutory provision constitutes\na jurisdictional limitation or an “essential ingredient” of a\nclaim, the Supreme Court has set forth a bright-line test:\n\n If the Legislature clearly states that a\n threshold limitation on a statute’s scope shall\n count as jurisdictional, then courts and\n\f8 TIJERINO V. STETSON DESERT PROJECT\n\n litigants will be duly instructed and will not\n left to wrestle with the issue. But when\n Congress does not rank a statutory limitation\n on coverage as jurisdictional, courts should\n treat the restriction as nonjurisdictional in\n character.\n\nArbaugh v. Y & H Corp., 546 U.S. 500, 515–16 (2006)\n(internal citation omitted). A limitation on a statute’s scope\ncounts as a jurisdictional limitation only if Congress clearly\n“intended” it to be jurisdictional such that courts should raise\nthe issue “on their own motion.” Id. at 514. Otherwise courts\nshould treat the statutory threshold as nonjurisdictional. Id.\n\n In Arbaugh, a restaurant challenged its former waitress’\nTitle VII sexual harassment claim after the close of trial,\narguing that the district court lacked subject-matter\njurisdiction over the case because Title VII’s statutory\ndefinition of “employer” did not include restaurants of its\nsize. Id. at 504 (“Y & H asserted that it had fewer than\n15 employees on its payroll and therefore was not amenable\nto suit under Title VII.”). Noting the tendency of courts to\nconflate subject-matter jurisdiction with the “plaintiff’s need\nand ability to prove the defendant bound by the federal law\nasserted as the predicate for relief—a merits-related\ndetermination,” the Supreme Court set out to clarify this\n“dichotomy.” Id. at 511 (emphasis added). The Court held\nthat, “if subject-matter jurisdiction turns on contested facts,\nthe trial judge may be authorized to review the evidence and\nresolve the dispute on her own. If satisfaction of an essential\nelement of a claim for relief is at issue, however, the jury is\nthe proper trier of contested facts.” Id. at 514 (internal\ncitations omitted).\n\f TIJERINO V. STETSON DESERT PROJECT 9\n\n Subsequently, in Reed Elsevier, Inc. v. Muchnick,\n559 U.S. 154 (2010), the Court reasoned that courts should\ndetermine whether a statutory provision is jurisdictional by\nlooking at the provision’s “text and structure,” as well as\n“context, and relevant historical treatment.” Id. at 162,\n166–68 (applying the Arbaugh test to hold that the Copyright\nAct’s registration requirement was not a jurisdictional\nprecondition to filing an infringement claim). The Ninth\nCircuit has applied Arbaugh and Reed Elsevier to\nemployment-related statutory provisions. See Day v. AT & T\nDisability Income Plan, 698 F.3d 1091, 1099–1100 (9th Cir.\n2012); Leeson, 671 F.3d at 978. A straightforward\nframework has emerged from this line of cases: when\ndetermining whether a statutory threshold is a jurisdictional\nbarrier, we ask whether Congress clearly labeled the statutory\nrequirement as jurisdictional and whether the requirement is\nlocated in a jurisdiction-granting provision; if so, Congress\nlikely intended the provision to be jurisdictional. See Reed\nElsevier, 559 U.S. at 160–69; Day, 698 F.3d at 1099–1100;\nLeeson, 671 F.3d at 970–79. Also helpful to our analysis is\nwhether courts have historically treated the requirement as\njurisdictional absent an express designation from Congress.\nReed Elsevier, 559 U.S. aat 169. In this Circuit, we also\nensure that “no other reasons necessitate[] that the provision\nbe construed as jurisdictional.” Leeson, 671 F.3d at 977. All\nof these factors help guide our analysis of the employment\nstatus provision in the FLSA.\n\n Congress did not clearly label the FLSA employment\nstatus provision as jurisdictional. “As in all cases involving\nstatutory construction, our starting point must be the language\nemployed by Congress.” Am. Tobacco Co. v. Patterson,\n456 U.S. 63, 68 (1982) (internal quotation marks omitted).\nHere, the relevant provision of FLSA § 216(b) states:\n\f10 TIJERINO V. STETSON DESERT PROJECT\n\n Any employer who violates the provisions of\n section 206 or section 207 of this title shall be\n liable to the employee or employees affected\n in the amount of their unpaid minimum\n wages, or their unpaid overtime\n compensation, as the case may be, and in an\n additional equal amount as liquidated\n damages . . . . An action to recover the\n liability prescribed in the preceding sentences\n may be maintained against any employer\n (including a public agency) in any Federal or\n State court of competent jurisdiction by any\n one or more employees for and in behalf of\n himself or themselves and other employees\n similarly situated.\n\n29 U.S.C. § 216(b).3 Section 216(b)’s requirement that\nemployers be liable to employees for violations of the FLSA\nis not clearly labeled as jurisdictional. In fact, there is not\neven a mention of jurisdiction in this section (besides a\ngeneral reference to a “court of competent jurisdiction”), and\nthere is no indication whatsoever that employee status is a\njurisdictional limitation. Compare this to the statute in Reed\nElsevier, in which even “the presence of the word\n“jurisdiction” in the last sentence of [the statutory provision]”\ndid not render that requirement jurisdictional. Reed Elsevier,\n559 U.S. at 163. In the “Definitions” section of the FLSA,\nCongress defined the term “employee” to mean “any\nindividual employed by an employer,” while also setting\naside exceptions to the term “employee” for certain\n\n\n 3\n 29 U.S.C. § 206 requires that non-exempt employees be paid the\nminimum wage and § 207 requires that non-exempt employees be paid\novertime wages for hours worked in excess of forty hours in one week.\n\f TIJERINO V. STETSON DESERT PROJECT 11\n\nindividuals employed by governmental entities, individuals\nwho volunteer to perform services for governmental entities,\nand immediate family members of employers employed in\nagriculture. 29 U.S.C. § 203(e)(1)–(4)(A). Here too, there is\nno indication that the criterion of a plaintiff being an\n“employee” is in any way related to subject matter\njurisdiction, let alone clearly labeled as jurisdictional.\n\n Section 203(e) and § 216(b) are not located in\njurisdiction-granting provisions of the statute. In Arbaugh,\nthe Court concluded that Congress did not intend that Title\nVII’s definition of the term “employer” be jurisdictional\nbecause Congress placed the term’s definition in a section\ntitled “Definitions”—“a separate provision that ‘does not\nspeak in jurisdictional terms or refer in any way to the\njurisdiction of the district courts.’” Arbaugh, 546 U.S. at 515\n(internal citation omitted); see also Leeson, 671 F.3d\nat 978–79 (Plaintiff’s status as a plan “participant,” for\npurposes of the Employee Retirement Income Security Act,\nwas a substantive element of his claim—not a prerequisite for\nsubject-matter jurisdiction—because the definition of\n“participant” appears in a provision that “does not speak in\njurisdictional terms.”). Here, Congress also placed § 203(e),\nwhich defines the term “employee,” in a section titled\n“Definitions,” and § 216(b) in a section titled “Penalties,”\nneither of which speaks to jurisdiction. See 29 U.S.C.\n§§ 203(e), 216(b); see also Day, 698 F.3d at 1099–1100\n(statute’s minimum age requirement for plaintiffs alleging\ndiscrimination claims “appear[ed] in a separate provision that\nd[id] not speak in jurisdictional terms” and therefore was a\nsubstantive element of a claim rather than a jurisdictional\nlimitation). Congress did use jurisdictional language in other\nsections of the FLSA, which suggests that Congress clearly\ndid not intend the requirements of § 216(b) and § 203(e) to be\n\f12 TIJERINO V. STETSON DESERT PROJECT\n\njurisdictional. See 29 U.S.C. § 210(a) (stating that upon the\nfiling of a petition for judicial review of a § 208 order in a\nspecific court of appeals, “such court shall have exclusive\njurisdiction to affirm, modify, . . . or set aside such order”).\n\n Furthermore, courts have not historically treated § 203(e)\nand § 216(b)’s employment status provisions as\njurisdictional. Although the Court in Reed Elsevier found\nprevious treatment of the Copyright Act’s registration\nrequirement as jurisdictional, it still concluded “that\n§ 411(a)’s registration requirement [was] nonjurisdictional,\nnotwithstanding its prior jurisdictional treatment.” 559 U.S.\nat 169. Here, courts have not historically treated the\nemployment status provision as jurisdictional; in similar\ncases, both the Fourth and Fifth Circuits have treated this\nquestion as nonjurisdictional. See McFeeley v. Jackson St.\nEntm’t LLC, 825 F.3d 235 (4th Cir. 2016); Reich v. Circle C.\nInvs., Inc., 998 F.2d 324 (5th Cir. 1993). Thus, all factors\nweigh in favor of construing the statute’s employment status\nprovision as nonjurisdictional.\n\n There is no support for the district court’s theory that the\nemployment status provision constitutes an antecedent\njurisdictional issue, and we can find no reasons that\n“necessitate[] that the provision be construed as\njurisdictional.” Leeson, 671 F.3d at 977. The district court\nattempted to distinguish cases like Leeson and Day, writing\nthat “the controlling cases upon which Plaintiff relie[d] to\nrefute the Court’s decision that employee status is an\nantecedent issue [were] specific to Employee Retirement\nIncome Security Act and Title VII of the Civil Rights Act\ncases – not FLSA cases.” But, aside from this conclusory\nstatement, the district court did not provide any further\nanalysis. It provided no rationale as to why Arbaugh’s\n\f TIJERINO V. STETSON DESERT PROJECT 13\n\nframework is not controlling, other than to suggest that\nArbaugh does not control merely because that case involved\na different statute than the statute in this case. Instead, the\ndistrict court cited Rutherford Food Corp. v. McComb,\n331 U.S. 722 (1947), in support of its position that the\nexercise of jurisdiction over a FLSA complaint turned on\nwhether the plaintiff was an employee. Id. at 727 (“Whether\nor not the acts charged in this complaint violate the [FLSA]\ndepends, so far as the meat boners are concerned, upon a\ndetermination as to whether either of both respondents are\nemployers of the boners.”). The district court, however,\noverlooked that Rutherford was a decision reached on the\nmerits, not on jurisdictional grounds. See id.4\n\n In summary, under the Arbaugh test, it is clear that the\nFLSA’s employment status provision in § 203(e) and\n§ 216(b) should be construed as nonjurisdictional. Our\nconclusion is also supported by applications of this test to\nother employment-related statutory provisions. See Day,\n698 F.3d at 1099–1100; Leeson, 671 F.3d at 979. Most\nimportantly, Congress did not state that the statute’s\nemployment status provision was jurisdictional, and § 203(e)\nand § 216(b) are not located in jurisdiction-granting\nprovisions of the statute. Finally, courts have not historically\ntreated the employment status provision as jurisdictional.\nThe term “employee” in the FLSA “serves to identify those\nplaintiffs who may be entitled to relief, not to limit the\n\n\n\n 4\n The district court also cited to Dellinger v. Sci. Applications Int’l\nCorp., in support of this proposition. Dellinger, however, is\ndistinguishable in that the plaintiff was never hired by nor performed any\nwork for the defendant: a prospective employer. See 649 F.3d 226, 228\n(4th Cir. 2011).\n\f14 TIJERINO V. STETSON DESERT PROJECT\n\nauthority of federal courts to adjudicate claims” under the\nFLSA. Id. at 978.\n\n B. The allegations were not so patently without merit\n as to justify dismissal for lack of subject-matter\n jurisdiction.\n\n Having held that the FLSA’s employment status provision\nin § 203(e) and § 216(b) is nonjurisdictional, we now turn to\nthe Dancers’ complaints to ensure that they allege a\nsubstantial issue of federal law that is sufficient to invoke\nsubject-matter jurisdiction. A complaint for relief properly\ninvokes federal jurisdiction where its well-pleaded allegations\nestablish “either that federal law creates the cause of action or\nthat the plaintiff’s right to relief necessarily depends on\nresolution of a substantial question of federal law.”\nFranchise Tax Bd., 463 U.S. at 27–28. There is no federal\nquestion jurisdiction where the federal claim asserted is “too\ninsubstantial,” Cook v. Peter Kiewit Sons Co., 775 F.2d 1030,\n1035 (9th Cir. 1985), defined as so patently without merit that\nthe claim requires no meaningful consideration. See Yokeno\nv. Mafnas, 973 F.2d 803, 808 (9th Cir. 1992); see also\nCement Masons Health & Welfare Trust Fund for N. Cal. v.\nStone, 197 F.3d 1003, 1008 (9th Cir. 1999) (“Any non-\nfrivolous assertion of a federal claim suffices to establish\nfederal question jurisdiction, even if that claim is later\ndismissed on the merits.”).\n\n Here, the Dancers’ complaints facially allege substantial\nissues of federal law: whether the Dancers could collectively\nrecover damages under § 216(b) for the Club’s failure to pay\nthem at the required minimum wage and at the required\novertime rate, for hours worked in excess of 40 per week, in\nviolation of § 206 and § 207 of the FLSA. Moreover, the\n\f TIJERINO V. STETSON DESERT PROJECT 15\n\nDancers allege numerous facts supporting their claims that\nthe Club owes them wages because they are employees under\nthe FLSA, and relatedly, under Arizona state law. The claims\nin the Dancers’ complaints are not “wholly insubstantial and\nfrivolous,” and are not so patently without merit as to warrant\ndismissal for lack of subject-matter jurisdiction. Arbaugh,\n546 U.S. at 513 n.10. Therefore, there is federal question\nsubject-matter jurisdiction to hear the Dancers’ § 216(b)\nclaims pursuant to 28 U.S.C. § 1331, and there is\nsupplemental jurisdiction over the state law claims.\n\nII. The Dancers’ other arguments on appeal\n\n The Dancers make two additional arguments on appeal.\nFirst, they ask us to remand with instructions to grant the\nDancers’ motion for conditional class certification under the\nFLSA. The district court did not reach a merits-based\nanalysis of conditional certification because it erroneously\ndetermined that it lacked subject-matter jurisdiction over the\ncase. Our holding that employee status under the FLSA is\nnonjurisdictional necessarily reverses the reasoning of the\ndistrict court’s denial of conditional certification. On\nremand, the district court will have the opportunity to\ndetermine whether to conditionally certify an FLSA\ncollective action in accordance with our precedent. See\nCampbell v. City of Los Angeles, 903 F.3d 1090, 1100 (9th\nCir. 2018) (discussing the “near-universal practice” of the\ntwo-step approach to conditional certification under\n§ 216(b)). We deny the Dancers’ request to remand with\nspecific instructions to grant the Dancers’ motion for\nconditional certification.\n\n Second, the Dancers ask us to remand with instructions\n“to adjudicate the merits of the Dancers’ claims in accordance\n\f16 TIJERINO V. STETSON DESERT PROJECT\n\nwith well-established case law applying the FLSA’s\neconomic realities test.” The district court applied the\neconomic realities test in reaching its conclusion that it lacked\nsubject-matter jurisdiction. Because we reverse the district\ncourt’s conclusion that it lacked subject-matter jurisdiction,\non remand, it will have the opportunity to address the\neconomic realities test when it addresses the merits of the\nDancers’ FLSA claims. See Real v. Driscoll Strawberry\nAssocs., Inc., 603 F.2d 748, 750 (9th Cir. 1979) (evaluating\nthe employment status of the plaintiffs on summary\njudgment). We doubt that the experienced district court is in\nneed of any further instructions on remand.\n\n CONCLUSION\n\n For the reasons set forth above, we reverse the district\ncourt’s dismissal of these actions for lack of subject-matter\njurisdiction, and remand for further proceedings consistent\nwith this opinion.\n\n REVERSED and REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Josephine Tijerino v. Stetson Desert Project, LLC\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-16\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nJOSEPHINE TIJERINO; TAMICKA No. 18-16013\\nTOLIVER; JANE ROE DANCER,\\n Plaintiffs-Appellants, D.C. Nos.\\n CV 15-2563-SMM\\n v. 15-2564-SMM\\n 16-0408-SMM\\nSTETSON DESERT PROJECT, LLC,\\nDBA Lé Girls Cabaret; CORY J.\\nANDERSON; CARY ANDERSON, OPINION\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Stephen M. McNamee, District Judge, Presiding\\n\\n Argued and Submitted July 9, 2019\\n Portland, Oregon\\n\\n Filed August 16, 2019\\n\\n Before: Ferdinand F. Fernandez, A. Wallace Tashima,\\n and John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Tashima\\n\\f2 TIJERINO V. STETSON DESERT PROJECT\\n\\n SUMMARY*\\n\\n\\n Labor Law\\n\\n The panel reversed the district court’s dismissal for lack\\nof subject-matter jurisdiction of an action brought under the\\nFair Labor Standards Act and Arizona state law by a group of\\nexotic dancers against the club at which they worked.\\n\\n The district court held that it lacked jurisdiction because\\nthe dancers did not prove at the outset of the case that they\\nwere employees rather than independent contractors.\\nReversing, the panel held that the statutory requirement that\\nplaintiffs must be employees as defined in the FLSA is a\\nmerits-based determination, not a jurisdictional limitation.\\nFurther, the dancers’ complaints sufficiently alleged\\nsubstantial issues of federal law. The panel remanded the\\ncase for further proceedings.\\n\\n\\n COUNSEL\\n\\nAndrew Sterling (argued) and Michael J. Rusing, Rusing\\nLopez & Lizardi PLLC, Tucson, Arizona, for Plaintiff-\\nAppellant Jane Roe Dancer.\\n\\nClifford P. Bendau II, The Bendau Law Firm PLLC, Phoenix,\\nArizona, for Plaintiffs-Appellants Josephine Tijerino and\\nTamika Toliver.\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f TIJERINO V. STETSON DESERT PROJECT 3\\n\\nLaurent R.G. Badoux (argued), Buchalter APC, Scottsdale,\\nArizona, for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nTASHIMA, Circuit Judge:\\n\\n These are actions under the Fair Labor Standards Act\\n(“FLSA”) brought by a group of exotic dancers against the\\nclub at which they worked, for asserted violations of the\\nFLSA and Arizona state law. The district court dismissed the\\nsuit with prejudice for lack of subject-matter jurisdiction\\nbecause the Dancers did not prove at the outset of the case\\nthat they were employees rather than independent contractors.\\nAs we explain below, the district court erred in reaching that\\ndecision. We reverse and remand for further proceedings.\\n\\n BACKGROUND\\n\\n Plaintiffs-Appellants Josephine Tijerino, Tamicka\\nToliver, and Jane Roe Dancer (collectively, the “Dancers”) all\\nworked as exotic dancers at Lé Girls Gentlemen’s Club, a\\nstrip club in Phoenix, Arizona. The club is owned and\\noperated by Defendants-Appellees Stetson Desert Project,\\nLLC, dba Lé Girls Cabaret, Cory Anderson, and Cary\\nAnderson (collectively, the “Club”). Initially, each dancer\\nfiled a separate complaint against the Club alleging violations\\nof employment law. The Dancers all allege that the Club\\nfailed to pay them any wages in violation of the FLSA and\\nwillfully failed to pay them at the rate of Arizona’s minimum\\nwage in violation of the Arizona Minimum Wage Act. See\\n29 U.S.C. § 206(a); Ariz. Rev. Stat. § 23-363. Separately,\\nTijerino and Toliver allege that the Club failed to pay them\\n\\f4 TIJERINO V. STETSON DESERT PROJECT\\n\\novertime wages in violation of the FLSA and willfully failed\\nto pay their wages for labor in violation of the Arizona Wage\\nLaw. See 29 U.S.C § 207; Ariz. Rev. Stat. § 23-351. Jane\\nRoe Dancer separately alleges that the Club was unjustly\\nenriched in violation of Ariz. Rev. Stat. § 23-202.\\n\\n Jane Roe Dancer filed her complaint as a collective action\\non behalf of herself and all other similarly situated\\nindividuals who worked at the Club as dancers. See\\n29 U.S.C. § 216(b). She filed a motion to consolidate the\\nDancers’ cases, which the district court granted. She also\\nfiled a motion for conditional certification and to authorize\\nnotice to putative class members, and she included a\\ndeclaration in which she provided information suggesting that\\nthe exotic dancers who worked at the Club were similarly\\nsituated for purposes of the FLSA. Shortly after the cases\\nwere consolidated, the parties submitted a proposed joint case\\nmanagement order in which they agreed that “[t]he basis for\\nthe court’s subject-matter jurisdiction over plaintiff’s claims\\nis the Fair Labor Standards Act, 29 U.S.C. § 201 et seq. . . .”\\n\\n The district court denied the Dancers’ motion to\\nconditionally certify an opt-in class and dismissed the case\\nwith prejudice for lack of subject-matter jurisdiction. First,\\nin denying the motion to conditionally certify a FLSA opt-in\\nclass, the district court called such a motion “premature.” It\\nheld that Jane Roe “must satisfy her initial burden of alleging\\nspecific facts that permit an inference that she is an\\n‘employee’ (and the Defendants her ‘employer’) within the\\nmeaning of the FLSA.” It also noted that the Dancers’ status\\nas employees or independent contractors of the Club was\\n“unsettled,” and ordered the Dancers to file a brief “on\\nwhether Plaintiff and her putative class are employees within\\nthe meaning of FLSA under the economic realities test.”\\n\\f TIJERINO V. STETSON DESERT PROJECT 5\\n\\n The district court next denied the Dancers’ renewed\\nmotion for conditional class certification and held that the\\nDancers were independent contractors, not employees. It\\nheld that § 216(b)’s requirement that employees bring actions\\nunder the statute constituted an “antecedent issue” of\\njurisdiction and thus that, if the Dancers were not employees\\nunder the FLSA, there could be no violation of the FLSA and\\ntherefore no federal question jurisdiction. The district court\\nproceeded to evaluate the Dancers’ employment status under\\nthe FLSA’s economic realities test and declared them to be\\nindependent contractors. It then denied the Dancers’ motion\\nto conditionally certify a FLSA opt-in class based on a lack\\nof subject-matter jurisdiction.1\\n\\n Finally, the district court dismissed all three plaintiffs’\\ncases with prejudice for lack of subject-matter jurisdiction.2\\nThe district court “reiterate[d] its position that employee\\nstatus under the FLSA is an antecedent jurisdictional issue,”\\nand that plaintiffs had not “come forward with any evidence\\nfrom which the Court could find a reasonable indicia of\\nemployment.”\\n\\n The Dancers appeal from the final judgment, the order\\ndismissing the cases with prejudice for lack of subject-matter\\njurisdiction, and the order denying their renewed motion for\\nconditional certification of an FLSA opt-in class.\\n\\n 1\\n The district court later denied the Dancers’ motion to reconsider this\\norder.\\n 2\\n Although we need not examine the issue in detail because we\\nreverse the district court’s dismissal for other reasons, we note that, in\\ngeneral, dismissal for lack of subject matter jurisdiction should be without\\nprejudice. See Mo. ex rel. Koster v. Harris, 847 F.3d 646, 656 (9th Cir.\\n2017).\\n\\f6 TIJERINO V. STETSON DESERT PROJECT\\n\\nAPPELLATE JURISDICTION AND STANDARD OF\\n REVIEW\\n\\n We have appellate jurisdiction under 28 U.S.C. § 1291.\\nWe review de novo a dismissal for lack of subject-matter\\njurisdiction. Leeson v. Transamerica Disability Income Plan,\\n671 F.3d 969, 974 (9th Cir. 2012). A district court’s\\ninterpretation of federal law, such as the FLSA, is also\\nreviewed de novo. Id.\\n\\n DISCUSSION\\n\\nI. The district court erred in concluding that it lacked\\n subject-matter jurisdiction.\\n\\n Congress granted federal district courts “original\\njurisdiction of all civil actions arising under the Constitution,\\nlaws, or treaties of the United States.” 28 U.S.C. § 1331.\\nThe Supreme Court has held that “Congress has given the\\nlower federal courts jurisdiction to hear . . . only those cases\\nin which a well-pleaded complaint establishes either that\\nfederal law creates the cause of action or that the plaintiff’s\\nright to relief necessarily depends on resolution of a\\nsubstantial question of federal law.” Franchise Tax Bd. v.\\nConstr. Laborers Vacation Trust for S. Cal., 463 U.S. 1,\\n27–28 (1983). “[A] federally created claim for relief is\\ngenerally a ‘sufficient condition for federal-question\\njurisdiction.’” Mims v. Arrow Fin. Servs., LLC, 565 U.S. 368,\\n377 (2012) (quoting Grable & Sons Metal Prods., Inc. v.\\nDarue Eng’g & Mfg., 545 U.S. 308, 317 (2005)). The FLSA\\nprovides such a claim for aggrieved employees to sue to\\nrecover unpaid minimum wages or unpaid overtime\\ncompensation, either individually or collectively. 29 U.S.C.\\n§ 216(b).\\n\\f TIJERINO V. STETSON DESERT PROJECT 7\\n\\n In order to determine whether the district court had\\nsubject-matter jurisdiction over the Dancers’ claims, we ask\\ntwo questions. First, we ask whether the employment status\\nprovision—the statutory requirement that plaintiffs must be\\nemployees as defined in the FLSA, 20 U.S.C. § 203(e), in\\norder to prevail in a § 216(b) action—is a jurisdictional\\nlimitation on the statute’s coverage such that it prevents the\\nDancers from moving forward with their claims. Second, we\\nask whether the Dancers’ complaints survive the well-pleaded\\ncomplaint rule to invoke federal question subject-matter\\njurisdiction.\\n\\n A. The Dancers’ employment status is a merits-based\\n determination, not an antecedent jurisdictional\\n issue.\\n\\n If the employment status provision of the statute is\\njurisdictional, then district courts would be required to\\nanalyze whether plaintiffs are employees under the FLSA at\\nthe outset of the litigation under Federal Rule of Civil\\nProcedure 12(b)(1), as the district court did in this case. But\\nif the employment status provision is an essential ingredient\\nrelated to the merits of a claim, then whether plaintiffs are\\nemployees should be decided at summary judgment or trial,\\nassuming the complaint sufficiently alleges that plaintiffs are\\nemployees.\\n\\n In determining whether a statutory provision constitutes\\na jurisdictional limitation or an “essential ingredient” of a\\nclaim, the Supreme Court has set forth a bright-line test:\\n\\n If the Legislature clearly states that a\\n threshold limitation on a statute’s scope shall\\n count as jurisdictional, then courts and\\n\\f8 TIJERINO V. STETSON DESERT PROJECT\\n\\n litigants will be duly instructed and will not\\n left to wrestle with the issue. But when\\n Congress does not rank a statutory limitation\\n on coverage as jurisdictional, courts should\\n treat the restriction as nonjurisdictional in\\n character.\\n\\nArbaugh v. Y & H Corp., 546 U.S. 500, 515–16 (2006)\\n(internal citation omitted). A limitation on a statute’s scope\\ncounts as a jurisdictional limitation only if Congress clearly\\n“intended” it to be jurisdictional such that courts should raise\\nthe issue “on their own motion.” Id. at 514. Otherwise courts\\nshould treat the statutory threshold as nonjurisdictional. Id.\\n\\n In Arbaugh, a restaurant challenged its former waitress’\\nTitle VII sexual harassment claim after the close of trial,\\narguing that the district court lacked subject-matter\\njurisdiction over the case because Title VII’s statutory\\ndefinition of “employer” did not include restaurants of its\\nsize. Id. at 504 (“Y & H asserted that it had fewer than\\n15 employees on its payroll and therefore was not amenable\\nto suit under Title VII.”). Noting the tendency of courts to\\nconflate subject-matter jurisdiction with the “plaintiff’s need\\nand ability to prove the defendant bound by the federal law\\nasserted as the predicate for relief—a merits-related\\ndetermination,” the Supreme Court set out to clarify this\\n“dichotomy.” Id. at 511 (emphasis added). The Court held\\nthat, “if subject-matter jurisdiction turns on contested facts,\\nthe trial judge may be authorized to review the evidence and\\nresolve the dispute on her own. If satisfaction of an essential\\nelement of a claim for relief is at issue, however, the jury is\\nthe proper trier of contested facts.” Id. at 514 (internal\\ncitations omitted).\\n\\f TIJERINO V. STETSON DESERT PROJECT 9\\n\\n Subsequently, in Reed Elsevier, Inc. v. Muchnick,\\n559 U.S. 154 (2010), the Court reasoned that courts should\\ndetermine whether a statutory provision is jurisdictional by\\nlooking at the provision’s “text and structure,” as well as\\n“context, and relevant historical treatment.” Id. at 162,\\n166–68 (applying the Arbaugh test to hold that the Copyright\\nAct’s registration requirement was not a jurisdictional\\nprecondition to filing an infringement claim). The Ninth\\nCircuit has applied Arbaugh and Reed Elsevier to\\nemployment-related statutory provisions. See Day v. AT & T\\nDisability Income Plan, 698 F.3d 1091, 1099–1100 (9th Cir.\\n2012); Leeson, 671 F.3d at 978. A straightforward\\nframework has emerged from this line of cases: when\\ndetermining whether a statutory threshold is a jurisdictional\\nbarrier, we ask whether Congress clearly labeled the statutory\\nrequirement as jurisdictional and whether the requirement is\\nlocated in a jurisdiction-granting provision; if so, Congress\\nlikely intended the provision to be jurisdictional. See Reed\\nElsevier, 559 U.S. at 160–69; Day, 698 F.3d at 1099–1100;\\nLeeson, 671 F.3d at 970–79. Also helpful to our analysis is\\nwhether courts have historically treated the requirement as\\njurisdictional absent an express designation from Congress.\\nReed Elsevier, 559 U.S. aat 169. In this Circuit, we also\\nensure that “no other reasons necessitate[] that the provision\\nbe construed as jurisdictional.” Leeson, 671 F.3d at 977. All\\nof these factors help guide our analysis of the employment\\nstatus provision in the FLSA.\\n\\n Congress did not clearly label the FLSA employment\\nstatus provision as jurisdictional. “As in all cases involving\\nstatutory construction, our starting point must be the language\\nemployed by Congress.” Am. Tobacco Co. v. Patterson,\\n456 U.S. 63, 68 (1982) (internal quotation marks omitted).\\nHere, the relevant provision of FLSA § 216(b) states:\\n\\f10 TIJERINO V. STETSON DESERT PROJECT\\n\\n Any employer who violates the provisions of\\n section 206 or section 207 of this title shall be\\n liable to the employee or employees affected\\n in the amount of their unpaid minimum\\n wages, or their unpaid overtime\\n compensation, as the case may be, and in an\\n additional equal amount as liquidated\\n damages . . . . An action to recover the\\n liability prescribed in the preceding sentences\\n may be maintained against any employer\\n (including a public agency) in any Federal or\\n State court of competent jurisdiction by any\\n one or more employees for and in behalf of\\n himself or themselves and other employees\\n similarly situated.\\n\\n29 U.S.C. § 216(b).3 Section 216(b)’s requirement that\\nemployers be liable to employees for violations of the FLSA\\nis not clearly labeled as jurisdictional. In fact, there is not\\neven a mention of jurisdiction in this section (besides a\\ngeneral reference to a “court of competent jurisdiction”), and\\nthere is no indication whatsoever that employee status is a\\njurisdictional limitation. Compare this to the statute in Reed\\nElsevier, in which even “the presence of the word\\n“jurisdiction” in the last sentence of [the statutory provision]”\\ndid not render that requirement jurisdictional. Reed Elsevier,\\n559 U.S. at 163. In the “Definitions” section of the FLSA,\\nCongress defined the term “employee” to mean “any\\nindividual employed by an employer,” while also setting\\naside exceptions to the term “employee” for certain\\n\\n\\n 3\\n 29 U.S.C. § 206 requires that non-exempt employees be paid the\\nminimum wage and § 207 requires that non-exempt employees be paid\\novertime wages for hours worked in excess of forty hours in one week.\\n\\f TIJERINO V. STETSON DESERT PROJECT 11\\n\\nindividuals employed by governmental entities, individuals\\nwho volunteer to perform services for governmental entities,\\nand immediate family members of employers employed in\\nagriculture. 29 U.S.C. § 203(e)(1)–(4)(A). Here too, there is\\nno indication that the criterion of a plaintiff being an\\n“employee” is in any way related to subject matter\\njurisdiction, let alone clearly labeled as jurisdictional.\\n\\n Section 203(e) and § 216(b) are not located in\\njurisdiction-granting provisions of the statute. In Arbaugh,\\nthe Court concluded that Congress did not intend that Title\\nVII’s definition of the term “employer” be jurisdictional\\nbecause Congress placed the term’s definition in a section\\ntitled “Definitions”—“a separate provision that ‘does not\\nspeak in jurisdictional terms or refer in any way to the\\njurisdiction of the district courts.’” Arbaugh, 546 U.S. at 515\\n(internal citation omitted); see also Leeson, 671 F.3d\\nat 978–79 (Plaintiff’s status as a plan “participant,” for\\npurposes of the Employee Retirement Income Security Act,\\nwas a substantive element of his claim—not a prerequisite for\\nsubject-matter jurisdiction—because the definition of\\n“participant” appears in a provision that “does not speak in\\njurisdictional terms.”). Here, Congress also placed § 203(e),\\nwhich defines the term “employee,” in a section titled\\n“Definitions,” and § 216(b) in a section titled “Penalties,”\\nneither of which speaks to jurisdiction. See 29 U.S.C.\\n§§ 203(e), 216(b); see also Day, 698 F.3d at 1099–1100\\n(statute’s minimum age requirement for plaintiffs alleging\\ndiscrimination claims “appear[ed] in a separate provision that\\nd[id] not speak in jurisdictional terms” and therefore was a\\nsubstantive element of a claim rather than a jurisdictional\\nlimitation). Congress did use jurisdictional language in other\\nsections of the FLSA, which suggests that Congress clearly\\ndid not intend the requirements of § 216(b) and § 203(e) to be\\n\\f12 TIJERINO V. STETSON DESERT PROJECT\\n\\njurisdictional. See 29 U.S.C. § 210(a) (stating that upon the\\nfiling of a petition for judicial review of a § 208 order in a\\nspecific court of appeals, “such court shall have exclusive\\njurisdiction to affirm, modify, . . . or set aside such order”).\\n\\n Furthermore, courts have not historically treated § 203(e)\\nand § 216(b)’s employment status provisions as\\njurisdictional. Although the Court in Reed Elsevier found\\nprevious treatment of the Copyright Act’s registration\\nrequirement as jurisdictional, it still concluded “that\\n§ 411(a)’s registration requirement [was] nonjurisdictional,\\nnotwithstanding its prior jurisdictional treatment.” 559 U.S.\\nat 169. Here, courts have not historically treated the\\nemployment status provision as jurisdictional; in similar\\ncases, both the Fourth and Fifth Circuits have treated this\\nquestion as nonjurisdictional. See McFeeley v. Jackson St.\\nEntm’t LLC, 825 F.3d 235 (4th Cir. 2016); Reich v. Circle C.\\nInvs., Inc., 998 F.2d 324 (5th Cir. 1993). Thus, all factors\\nweigh in favor of construing the statute’s employment status\\nprovision as nonjurisdictional.\\n\\n There is no support for the district court’s theory that the\\nemployment status provision constitutes an antecedent\\njurisdictional issue, and we can find no reasons that\\n“necessitate[] that the provision be construed as\\njurisdictional.” Leeson, 671 F.3d at 977. The district court\\nattempted to distinguish cases like Leeson and Day, writing\\nthat “the controlling cases upon which Plaintiff relie[d] to\\nrefute the Court’s decision that employee status is an\\nantecedent issue [were] specific to Employee Retirement\\nIncome Security Act and Title VII of the Civil Rights Act\\ncases – not FLSA cases.” But, aside from this conclusory\\nstatement, the district court did not provide any further\\nanalysis. It provided no rationale as to why Arbaugh’s\\n\\f TIJERINO V. STETSON DESERT PROJECT 13\\n\\nframework is not controlling, other than to suggest that\\nArbaugh does not control merely because that case involved\\na different statute than the statute in this case. Instead, the\\ndistrict court cited Rutherford Food Corp. v. McComb,\\n331 U.S. 722 (1947), in support of its position that the\\nexercise of jurisdiction over a FLSA complaint turned on\\nwhether the plaintiff was an employee. Id. at 727 (“Whether\\nor not the acts charged in this complaint violate the [FLSA]\\ndepends, so far as the meat boners are concerned, upon a\\ndetermination as to whether either of both respondents are\\nemployers of the boners.”). The district court, however,\\noverlooked that Rutherford was a decision reached on the\\nmerits, not on jurisdictional grounds. See id.4\\n\\n In summary, under the Arbaugh test, it is clear that the\\nFLSA’s employment status provision in § 203(e) and\\n§ 216(b) should be construed as nonjurisdictional. Our\\nconclusion is also supported by applications of this test to\\nother employment-related statutory provisions. See Day,\\n698 F.3d at 1099–1100; Leeson, 671 F.3d at 979. Most\\nimportantly, Congress did not state that the statute’s\\nemployment status provision was jurisdictional, and § 203(e)\\nand § 216(b) are not located in jurisdiction-granting\\nprovisions of the statute. Finally, courts have not historically\\ntreated the employment status provision as jurisdictional.\\nThe term “employee” in the FLSA “serves to identify those\\nplaintiffs who may be entitled to relief, not to limit the\\n\\n\\n\\n 4\\n The district court also cited to Dellinger v. Sci. Applications Int’l\\nCorp., in support of this proposition. Dellinger, however, is\\ndistinguishable in that the plaintiff was never hired by nor performed any\\nwork for the defendant: a prospective employer. See 649 F.3d 226, 228\\n(4th Cir. 2011).\\n\\f14 TIJERINO V. STETSON DESERT PROJECT\\n\\nauthority of federal courts to adjudicate claims” under the\\nFLSA. Id. at 978.\\n\\n B. The allegations were not so patently without merit\\n as to justify dismissal for lack of subject-matter\\n jurisdiction.\\n\\n Having held that the FLSA’s employment status provision\\nin § 203(e) and § 216(b) is nonjurisdictional, we now turn to\\nthe Dancers’ complaints to ensure that they allege a\\nsubstantial issue of federal law that is sufficient to invoke\\nsubject-matter jurisdiction. A complaint for relief properly\\ninvokes federal jurisdiction where its well-pleaded allegations\\nestablish “either that federal law creates the cause of action or\\nthat the plaintiff’s right to relief necessarily depends on\\nresolution of a substantial question of federal law.”\\nFranchise Tax Bd., 463 U.S. at 27–28. There is no federal\\nquestion jurisdiction where the federal claim asserted is “too\\ninsubstantial,” Cook v. Peter Kiewit Sons Co., 775 F.2d 1030,\\n1035 (9th Cir. 1985), defined as so patently without merit that\\nthe claim requires no meaningful consideration. See Yokeno\\nv. Mafnas, 973 F.2d 803, 808 (9th Cir. 1992); see also\\nCement Masons Health & Welfare Trust Fund for N. Cal. v.\\nStone, 197 F.3d 1003, 1008 (9th Cir. 1999) (“Any non-\\nfrivolous assertion of a federal claim suffices to establish\\nfederal question jurisdiction, even if that claim is later\\ndismissed on the merits.”).\\n\\n Here, the Dancers’ complaints facially allege substantial\\nissues of federal law: whether the Dancers could collectively\\nrecover damages under § 216(b) for the Club’s failure to pay\\nthem at the required minimum wage and at the required\\novertime rate, for hours worked in excess of 40 per week, in\\nviolation of § 206 and § 207 of the FLSA. Moreover, the\\n\\f TIJERINO V. STETSON DESERT PROJECT 15\\n\\nDancers allege numerous facts supporting their claims that\\nthe Club owes them wages because they are employees under\\nthe FLSA, and relatedly, under Arizona state law. The claims\\nin the Dancers’ complaints are not “wholly insubstantial and\\nfrivolous,” and are not so patently without merit as to warrant\\ndismissal for lack of subject-matter jurisdiction. Arbaugh,\\n546 U.S. at 513 n.10. Therefore, there is federal question\\nsubject-matter jurisdiction to hear the Dancers’ § 216(b)\\nclaims pursuant to 28 U.S.C. § 1331, and there is\\nsupplemental jurisdiction over the state law claims.\\n\\nII. The Dancers’ other arguments on appeal\\n\\n The Dancers make two additional arguments on appeal.\\nFirst, they ask us to remand with instructions to grant the\\nDancers’ motion for conditional class certification under the\\nFLSA. The district court did not reach a merits-based\\nanalysis of conditional certification because it erroneously\\ndetermined that it lacked subject-matter jurisdiction over the\\ncase. Our holding that employee status under the FLSA is\\nnonjurisdictional necessarily reverses the reasoning of the\\ndistrict court’s denial of conditional certification. On\\nremand, the district court will have the opportunity to\\ndetermine whether to conditionally certify an FLSA\\ncollective action in accordance with our precedent. See\\nCampbell v. City of Los Angeles, 903 F.3d 1090, 1100 (9th\\nCir. 2018) (discussing the “near-universal practice” of the\\ntwo-step approach to conditional certification under\\n§ 216(b)). We deny the Dancers’ request to remand with\\nspecific instructions to grant the Dancers’ motion for\\nconditional certification.\\n\\n Second, the Dancers ask us to remand with instructions\\n“to adjudicate the merits of the Dancers’ claims in accordance\\n\\f16 TIJERINO V. STETSON DESERT PROJECT\\n\\nwith well-established case law applying the FLSA’s\\neconomic realities test.” The district court applied the\\neconomic realities test in reaching its conclusion that it lacked\\nsubject-matter jurisdiction. Because we reverse the district\\ncourt’s conclusion that it lacked subject-matter jurisdiction,\\non remand, it will have the opportunity to address the\\neconomic realities test when it addresses the merits of the\\nDancers’ FLSA claims. See Real v. Driscoll Strawberry\\nAssocs., Inc., 603 F.2d 748, 750 (9th Cir. 1979) (evaluating\\nthe employment status of the plaintiffs on summary\\njudgment). We doubt that the experienced district court is in\\nneed of any further instructions on remand.\\n\\n CONCLUSION\\n\\n For the reasons set forth above, we reverse the district\\ncourt’s dismissal of these actions for lack of subject-matter\\njurisdiction, and remand for further proceedings consistent\\nwith this opinion.\\n\\n REVERSED and REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4075,"cells":{"id":{"kind":"number","value":4649094,"string":"4,649,094"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"John O'Rourke v. No. California Electrical"},"decision_date":{"kind":"string","value":"2019-08-16"},"docket_number":{"kind":"string","value":"17-17419"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: J. Clifford Wallace, Eugene E. Siler, * and M. Margaret McKeown, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"WALLACE, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nJOHN J. O’ROURKE, No. 17-17419\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 3:16-cv-02007-\\n WHO\\nNORTHERN CALIFORNIA ELECTRICAL\\nWORKERS PENSION PLAN; BOARD OF\\nTRUSTEES OF THE NORTHERN OPINION\\nCALIFORNIA ELECTRICAL WORKERS\\nPENSION TRUST,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n William Horsley Orrick, District Judge, Presiding\\n\\n Argued and Submitted March 15, 2019\\n San Francisco, California\\n\\n Filed August 16, 2019\\n\\n Before: J. Clifford Wallace, Eugene E. Siler, *\\n and M. Margaret McKeown, Circuit Judges.\\n\\n Opinion by Judge Wallace\\n\\n\\n *\\n The Honorable Eugene E. Siler, United States Circuit Judge for\\nthe U.S. Court of Appeals for the Sixth Circuit, sitting by designation.\\n\\f2 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\n SUMMARY **\\n\\n\\n ERISA\\n\\n The panel affirmed the district court’s grant of summary\\njudgment in an ERISA action challenging the denial of\\nplaintiff’s request for early retirement benefits.\\n\\n Plaintiff accrued benefits through a multiemployer\\nERISA plan during his work as an electrician. When he left\\nthis position to work for an electrical workers’ union as an\\nadministrator, he sought early retirement benefits from the\\nplan. The plan’s board of trustees decided that plaintiff’s\\nunion work fell within the plan’s definition of “prohibited\\nemployment,” and so no benefits were due for any month in\\nwhich he engaged in that work.\\n\\n Reviewing the denial of benefits for an abuse of\\ndiscretion, the panel held that any procedural irregularities\\nin the actions of the board were minor and, at most, weighed\\nonly slightly and weakly in favor of holding that an abuse of\\ndiscretion occurred. The panel held that the board did not\\nabuse its discretion in interpreting the plan’s definition of\\nprohibited employment to include plaintiff’s union work\\nbecause the board’s interpretation did not clearly conflict\\nwith the plan’s plain language, did not render any other plan\\nprovision nugatory, and did not lack a rational nexus to the\\nplan’s purpose.\\n\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 3\\n\\n COUNSEL\\n\\nTeresa S. Renaker (argued) and Margo Hasselman\\nGreenough, Renaker Hasselman Scott LLP, San Francisco,\\nCalifornia, for Plaintiff-Appellant.\\n\\nClarissa A. Kang (argued), R. Bradford Huss, and Angel L.\\nGarrett, Trucker Huss, APC, San Francisco, California, for\\nDefendants-Appellees.\\n\\n\\n OPINION\\n\\nWALLACE, Circuit Judge:\\n\\n John O’Rourke is an electrician who accrued benefits\\nthrough a multiemployer ERISA pension plan and was a\\nmember of the plan’s board of trustees (Board). After\\nO’Rourke left this position to work for an electrical workers’\\nunion as an administrator, he sought early retirement\\nbenefits from the plan. The Board denied his request.\\nO’Rourke then filed this action for judicial review of the\\ndenial, and the district court entered summary judgment in\\nthe Board’s favor. We have jurisdiction over O’Rourke’s\\nappeal from that judgment under 28 U.S.C. § 1291, and we\\naffirm.\\n\\n I.\\n\\n A.\\n\\n The Northern California Electrical Workers Pension\\nPlan (Plan) is a multiemployer plan governed by the\\nEmployee Retirement Income Security Act (ERISA). See\\n29 U.S.C. § 1002(37)(A). The Plan is funded through the\\nNorthern California Electrical Workers Pension Trust\\n\\f4 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\n(Trust), to which participating employers contribute, and is\\nadministered by the six-member Board, containing three\\nlabor representatives and three management representatives.\\nThe Plan provides its participants with three types of benefit\\n— (1) normal pension, (2) early pension, and (3) disability\\npension — payable as a monthly annuity when participants\\nattain certain Plan-defined eligibility requirements. The Plan\\nprovides that the Board shall resolve “[a]ny dispute as to\\neligibility, type, amount or duration of benefits,” Plan Art.\\nIX, § A, and grants the Board “the exclusive power and\\ndiscretion to interpret the provisions of the Plan and any\\nrules issued under the Plan, and to determine all questions\\narising under the Plan including eligibility for benefits,” Plan\\nArt. XIII, § A.\\n\\n Relevant to this appeal, the Plan provides that\\nparticipants become eligible for early retirement and an early\\npension at age fifty-five, if they have also accumulated ten\\nor more years of covered employment. Plan Art. III,\\n§ (B)(1). However, no benefit payments are made for either\\nnormal pensions or early pensions “for any month in which\\nthe Participant works in Prohibited Employment.” Plan Art.\\nIII, § G(1). For participants who have not yet reached the age\\nof sixty-five, the Plan defines “Prohibited Employment” as\\n“the performance of services in any capacity in the Electrical\\nIndustry.” Plan Art. III, § G(3)(a). In turn, “Electrical\\nIndustry” is defined as “all branches of the Electrical Trade\\nin the United States.” Plan. Art. I. “Electrical Trade” is not\\ndefined.\\n\\n B.\\n\\n John O’Rourke is a participant in the Plan who began\\nworking as an electrician in 1979. In 1999, he was elected\\nbusiness manager of his local union, the International\\nBrotherhood of Electrical Workers (IBEW) Local 6. As\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 5\\n\\nbusiness manager, O’Rourke served as a Plan trustee on the\\nBoard.\\n\\n In September 2010, O’Rourke proposed that the Board\\nadopt a change in the Plan’s suspension of benefits rules to\\nexempt work for unions from the definition of prohibited\\nemployment. O’Rourke recommended that the Plan Counsel\\nreview the Plan and Trust and determine “what changes, if\\nany, are required to accomplish this.” The Board concurred\\nwith that recommendation, but no further action was taken\\nat that time. O’Rourke then left his Local 6 and Board\\nposition to join IBEW as an International Field\\nRepresentative, followed by service as Vice President for the\\nIBEW Ninth District. O’Rourke applied for an early\\nretirement pension from the Plan in June 2014, in\\nanticipation of his upcoming fifty-fifth birthday. It is\\nundisputed that once O’Rourke turned fifty-five, he met all\\nother eligibility requirements for the early pension, aside\\nfrom the suspension of benefits provision. It is also\\nundisputed that O’Rourke’s work for IBEW did not include\\ntraditional work as an electrician, such as wiring, repair,\\ninstallation, and maintenance.\\n\\n Meanwhile, Plan Counsel investigated the matter as\\nrequested. In March 2014, Plan Counsel submitted a\\nmemorandum to the Board opining that no changes to the\\nPlan were necessary because the Plan’s prohibited\\nemployment did not include work for a union. The Board\\nconsidered the opinion at a meeting the following week,\\nwhere two trustees expressed “strong disagreement with the\\nlegal opinion,” indicating that “employment for Local 6, or\\nany other IBEW for that matter, [should be] considered\\n‘covered employment’ . . . subject to suspension of\\nbenefits.” The Board did not resolve the matter at that\\n\\f6 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nmeeting; instead it requested that Plan Counsel conduct\\nfurther research on the matter.\\n\\n Plan Counsel submitted a second memorandum in\\nAugust 2014. After reviewing past Board meeting minutes,\\nPlan Counsel concluded that the Board had previously\\ndefined prohibited employment “broadly to include work\\nthat may not specifically bear upon or utilize skills and\\nexperience obtained by work in the electrical trade.” Plan\\nCounsel therefore opined that the Board should not adopt\\nany policy of categorical exemption, and instead make each\\ndetermination based on “all the facts and circumstances that\\nmay be submitted regarding the nature of the work in\\nquestion.”\\n\\n The Board discussed the second memorandum and\\nO’Rourke’s application at its next meeting, resolved to defer\\ndecision pending further information from O’Rourke, and\\nrequested an explanation from him “as to why he believes\\nthe work he is performing for [IBEW] is not prohibited under\\nthe Plan.” O’Rourke responded by letter, explaining that his\\nduties were “purely administrative” and did “not require,\\ndirectly or indirectly, the use of the same skills employed by\\nelectricians in an electrical trade or craft.”\\n\\n The Trustees discussed O’Rourke’s response via email\\nin anticipation of their next meeting in December, with one\\ntrustee stating that O’Rourke’s application was “a loser\\nbarring an amendment to the plan.” At the next meeting, the\\nBoard again discussed O’Rourke’s application and was\\nagain unable to reach consensus. Plan Counsel then opined\\nthat the suspension of benefits provision was ambiguous and\\nsuggested clarifying its meaning so that the Board could\\nreach consensus, to which the Board agreed.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 7\\n\\n The following month, in January 2015, O’Rourke’s\\ncounsel sent a demand letter to the Board stating O’Rourke’s\\nposition that any interpretation of the Plan to include his\\nIBEW employment would be an abuse of discretion.\\nTrustees discussed the letter via email, with one trustee\\nstating that O’Rourke’s description of his work as\\n“administrative” was not outcome determinative, with a\\npersonal reference to O’Rourke’s time as a trustee:\\n\\n First, to use the term “administrative”, as if\\n it’s [sic] use in and of itself it [sic] would\\n render his benefits payable, fails to address\\n how the Trustees have interpreted the Plan in\\n the past. Time and time again, Plan\\n participants sought to have their benefits paid\\n while working in and [sic] administrative\\n capacity. Mr. O’Rourke should know that\\n better than anyone since it was he that\\n championed that determination.\\n\\nPlan Counsel also stated that “management has not relented\\non the issue.”\\n\\n The Board met again in March 2015, and this time\\nagreed to deny O’Rourke’s application. In a letter to\\nO’Rourke’s attorney, the Board explained that O’Rourke’s\\nIBEW employment fell within prohibited employment\\nbecause it entailed “the performance of services of some\\ncapacity in the Electrical Industry,” and no exception\\napplied. O’Rourke appealed that decision. The Board met\\nagain in August and denied the appeal. The Board explained\\nthat it agreed there was no factual dispute over O’Rourke’s\\nIBEW duties or that the Plan defined “Electrical Industry”\\nas “Electrical Trade,” but that under its interpretation of\\n\\f8 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\n“trade,” O’Rourke’s work came within the prohibited\\nemployment definition.\\n\\n O’Rourke filed this action against the Plan and the Board\\nin April 2016. O’Rourke alleged that he was entitled to an\\nearly pension and claimed benefits pursuant to ERISA. See\\n29 U.S.C. § 1132(a)(1)(B). The parties cross-moved for\\nsummary judgment and the district court granted the Board’s\\nmotion and denied O’Rourke’s motion. The district court\\nreasoned that the Board did not abuse its discretion by\\ninterpreting the Plan to include O’Rourke’s IBEW position\\nas prohibited employment because both parties’\\ninterpretations were reasonable. O’Rourke timely appealed.\\n\\n II.\\n\\n “Where an ERISA Plan grants discretionary authority to\\ndetermine eligibility for benefits or to construe the terms of\\nthe plan, a plan administrator’s interpretation of a plan is\\nreviewed for abuse of discretion.” Lehman v. Nelson, 862\\nF.3d 1203, 1216 (9th Cir. 2017) (quoting Tapley v. Locals\\n302 & 612 of Int’l Union of Operating Eng’rs-Emp’rs Const.\\nIndus. Ret. Plan, 728 F.3d 1134, 1139 (9th Cir. 2013)). “We\\nreview the district court’s application of this standard and the\\ndistrict court’s . . . summary judgment de novo.” Id.\\n\\n III.\\n\\n O’Rourke argues that the Board incorrectly interpreted\\nthe Plan to deny his application for benefits. The Plan\\ncontains an express discretion provision, Plan Art. XIII, § A,\\nand therefore our review is for abuse of discretion. Lehman,\\n862 F.3d at 1216. Before we address the merits of this\\nargument, however, we must decide how alleged procedural\\nirregularities affect our analysis.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 9\\n\\n A.\\n\\n O’Rourke contends that the Board acted “as an adversary\\nbent on denying his claim,” see Friedrich v. Intel Corp., 181\\nF.3d 1105, 1110 (9th Cir. 1999) (internal quotation marks\\nomitted), rather than as a neutral arbiter acting in the Plan’s\\nbest interests. According to him, we should view the Board’s\\ndecision with suspicion, making it easier for us to conclude\\nthe Board abused its discretion.\\n\\n We previously addressed the issue of procedural\\nirregularity in Abatie v. Alta Health and Life Insurance, 458\\nF.3d 955 (9th Cir. 2006) (en banc). In that case, we explained\\nthat ERISA provides “for only two alternatives[: w]hen a\\nplan confers discretion, abuse of discretion review applies;\\nwhen it does not, de novo review applies.” Id. at 965. Thus,\\nprocedural irregularity does not alter the standard of review\\nexcept in “situations in which procedural irregularities are so\\nsubstantial” as to make doing so necessary, such as “[w]hen\\nan administrator engages in wholesale and flagrant\\nviolations of the procedural requirements of ERISA, and\\nthus acts in utter disregard of the underlying purpose of the\\nplan as well.” Id. at 971. O’Rourke agrees that this case does\\nnot present such a situation, and thus abuse of discretion\\nreview applies.\\n\\n The question remains, however, whether any procedural\\nirregularities here affect the abuse of discretion review itself.\\nIn Abatie, we rejected a “sliding scale” approach to this\\nproblem, where the greater the irregularity, the lesser the\\nplaintiff’s burden. Id. at 967–68. Instead, we explained that\\na “procedural irregularity, like a conflict of interest, is a\\nmatter to be weighed in deciding whether an administrator’s\\ndecision was an abuse of discretion.” Id. at 972. Thus,\\n“[w]hen an administrator can show that it has engaged in an\\nongoing, good faith exchange of information between the\\n\\f10 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nadministrator and the claimant, the court should give the\\nadministrator’s decision broad deference notwithstanding a\\nminor irregularity.” Id. (internal quotation marks and\\ncitation omitted). But when “[a] more serious procedural\\nirregularity” has occurred, it “may weigh more heavily.” Id.\\nat 972–73. We did not provide more guidance on the issue\\nbeyond these statements, explaining that abuse of discretion\\nreview is inherently “indefinite” and requires case-by-case\\nanalysis. Id. at 969.\\n\\n In this case, as the district court concluded, the Plan is\\nambiguous and both sides present plausible interpretations\\nof the Plan language. If we were to accept O’Rourke’s\\nassertions that the Board acted as an adversary against him,\\nwe might accordingly hold that an abuse of discretion\\noccurred where we otherwise might not. We must therefore\\nfirst decide how much weight to place on the alleged\\nprocedural irregularities in this case before proceeding.\\n\\n O’Rourke emphasizes four events as procedurally\\nirregular: (1) emails and meeting minutes showing “political\\nhostility” and personal animus towards him; (2) shifting\\nrationales for the denial; (3) the rejection of Plan Counsel’s\\n2014 opinion; and (4) the Board’s position reversal from the\\n2010 resolution. We address each in turn.\\n\\n First, the record does not support O’Rourke’s assertion\\nthat certain trustees harbored personal hostility towards him.\\nThe meeting minutes and emails reflect that certain trustees\\nstrongly disagreed with O’Rourke’s interpretation of the\\nPlan, but nothing in the record rises to the level of animus.\\nFor instance, calling an argument a “loser” is usually a\\nreflection on the argument’s merits, not its proponent’s, as\\nseems to be the case in the December email. Similarly,\\nsaying that O’Rourke should have known the Board would\\ninterpret the Plan in a way that did not favor him was\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 11\\n\\nirrelevant to the merits of O’Rourke’s demand letter, as an\\napplicant’s knowledge or intent under the Plan does not\\naffect their eligibility for benefits. Thus, we disagree with\\nO’Rourke’s argument that the Board acted based on animus\\nrather than in the Plan’s best interests.\\n\\n Second, the Board did not shift its rationales for denying\\nthe claim. O’Rourke argues that the Board initially denied\\nhis application because prohibited employment included\\nsome employment that did not require an electrician’s skills,\\nbut later denied his appeal because unions provide support\\nto workers in the electrical trade. There is no discrepancy. In\\ndenying the application, the Board explained that it was\\nrelying on a broad definition of “industry” which included\\nservices for IBEW. In denying the appeal, the Board\\nexplained that it was relying on a broad definition of “trade,”\\nwhich again included services for IBEW. The Plan defines\\n“Electrical Industry” as “all branches of the Electrical Trade\\nin the United States.” Plan. Art. I. Thus, the definition of\\n“industry” in the first decision and “trade” in the second\\ndecision is consistent with the Plan and the Board’s position\\nthroughout its decision-making and this litigation.\\n\\n Third, while disregarding the advice of counsel might\\nweigh in favor of holding that an abuse of discretion\\noccurred in a different case, here the record reflects that all\\nparties agreed the Plan was ambiguous and the disagreement\\nwas in good faith. In its first memorandum to the Board, Plan\\nCounsel opined that the Plan’s prohibited employment\\ndefinition related to work involving the skills used by an\\nelectrician. However, in the second memorandum, Plan\\nCounsel acknowledged that Plan Counsel had not reviewed\\nprevious determinations made by the Board when that\\nopinion was reached, and that the Board had previously\\ninterpreted prohibited employment “more broadly to include\\n\\f12 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nwork that may not specifically bear upon or utilize skills and\\nexperience obtained by work in the electrical trade.” Plan\\nCounsel accordingly revised the original opinion in light of\\nthe new information and recommended a case-by-case\\nanalysis based on the totality of the circumstances\\n“regarding the nature of the work in question.” Thus, while\\nO’Rourke is correct that the Board did not follow Plan\\nCounsel’s recommendation, a full picture of the record\\nshows that this fact weighs only slightly against the Board,\\nif at all.\\n\\n Fourth, the Board’s change in position from 2010 to\\n2014 demonstrates only a change in opinion, not a\\nprocedural irregularity. O’Rourke does not argue on appeal,\\nnor could he, that the 2010 resolution to ask Plan Counsel to\\ndetermine what changes were necessary to exempt union\\nservice from prohibited employment constituted binding\\naction. The motion was necessarily tentative because it\\nrequired further action. Instead, the change in position shows\\nthat the Board was once in favor of allowing early pension\\nbenefits for IBEW workers, and it then changed its mind.\\nSuch a change, in conjunction with other evidence showing\\ndisregard of the Plan’s terms, could be evidence of bad faith.\\nBut the only other evidence in the record supporting bad\\nfaith is the Board’s rejection of Plan Counsel’s opinion. As\\nexplained above, the Board’s rejection of Plan Counsel’s\\nopinion does not show bad faith under the circumstances.\\nThus, the Board’s change in position weighs only minimally\\nin favor of holding that an abuse of discretion occurred here.\\n\\n Finally, we point out that the Board kept O’Rourke\\ninformed at all stages of its decision-making and that none\\nof the putative irregularities prevented the administrative\\nrecord from being fully developed or prevented the Board or\\na court from knowing all relevant facts. This record does not\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 13\\n\\nreflect that the Board attempted to keep O’Rourke in the dark\\nor use procedural devices to prevent him from receiving\\nbenefits.\\n\\n Thus, O’Rourke’s procedural irregularity argument\\ndemonstrates at best only that the current Board held a\\ndifferent interpretation of the Plan from O’Rourke and from\\npast Boards, and that it appeared unwilling to change that\\ninterpretation on counsel’s advice or O’Rourke’s arguments.\\nWhile this could perhaps indicate some stubbornness beyond\\nthe platonic ideal of an open-minded plan administrator, it\\ndoes not reveal serious procedural irregularities that would\\nweigh heavily against the Board. At most, the Board’s\\nactions weigh only slightly and weakly in favor of holding\\nthat an abuse of discretion occurred here.\\n\\n B.\\n\\n Having concluded that any procedural irregularities were\\nminor, we turn to the Board’s interpretation of the Plan itself.\\nThe Plan defines “Prohibited Employment” as “the\\nperformance of services in any capacity in the Electrical\\nIndustry.” Plan Art. III, § G(3)(a). O’Rourke argues that by\\ndefining “Electrical Industry” to mean “all branches of the\\nElectrical Trade in the United States,” the Plan limits\\nprohibited employment to work involving the skills of an\\nelectrician. The Board responds that it does not matter\\nwhether the work involves an electrician’s skills, so long as\\nit involves any type of livelihood related to electrical work,\\nincluding administrative work. The question in this appeal is\\nthus whether the Board abused its discretion by interpreting\\n“performance of services in any capacity in the Electrical\\nIndustry” to include working for IBEW; i.e., in an\\nadministrative capacity for an electrical workers’ union. To\\nanswer that question, we ask whether the Board’s\\ninterpretation is unreasonable, “closely read[ing] contested\\n\\f14 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nterms and appl[ying] contract principles derived from state\\nlaw, guided by the policies expressed in ERISA and other\\nfederal labor laws.” Tapley, 728 F.3d at 1140 (internal\\nquotation marks and alterations omitted) (quoting\\nRichardson v. Pension Plan of Bethlehem Steel Corp., 112\\nF.3d 982, 985 (9th Cir. 1997)). We will accept the Board’s\\ninterpretation unless it is “not grounded on any reasonable\\nbasis.” Id. at 1139 (emphasis in original) (quoting Oster v.\\nBarco of Cal. Emps.’ Ret. Plan, 869 F.2d 1215, 1218 (9th\\nCir. 1988)). “The [Board’s] interpretation need not be the\\none this court would have reached, but only an interpretation\\nwhich has rational justifications.” Id. at 1139–40 (quoting\\nSmith v. CMTA-IAM Pension Trust, 654 F.2d 650, 655 (9th\\nCir. 1981)).\\n\\n In Tapley, we described three ways in which a Plan\\nadministrator’s interpretation might fail this test. First, it is\\nan abuse of discretion to “construe provisions of a plan in a\\nway that clearly conflicts with the plain language of the\\nPlan.” Id. at 1140 (alteration omitted) (quoting Johnson v.\\nTrs. of W. Conf. of Teamsters Pension Tr. Fund, 879 F.2d\\n651, 654 (9th Cir. 1989)). Second, it is an abuse of discretion\\nto interpret a provision in a way that “renders nugatory other\\nprovisions of the Plan.” Id. (internal quotation marks and\\nalterations omitted) (quoting Richardson, 112 F.3d at 985).\\nThird, it is an abuse of discretion to give an interpretation\\nthat “lacks any rational nexus to the primary purpose of the\\nPlan.” Id. (quotation marks omitted) (quoting Burditt v. W.\\nGrowers Pension Plan, 636 F. Supp. 1491, 1498 (C.D. Cal.\\n1986)). These three types of abuse correspond more or less\\nto basic principles of contract interpretation; i.e., (1) to\\ninterpret provisions based on their plain meaning, (2) to\\ninterpret provisions in context with other provisions, and\\n(3) to interpret provisions to reach a reasonable outcome that\\naccords with the expectations of the agreeing parties.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 15\\n\\n O’Rourke argues that all three Tapley errors are present\\nhere and require us to reject the Board’s interpretation. We\\ndisagree with each of his arguments.\\n\\n First, the Board’s interpretation does not clearly conflict\\nwith the Plan’s plain language. O’Rourke makes a strong\\nargument for why the best interpretation of “trade” in the\\nPlan is working as an electrician. However, he fails to\\nexclude the Board’s interpretation as a reasonable\\npossibility. The operative language in the Plan is very broad:\\nit defines prohibited employment as “the performance of\\nservices in any capacity in the Electrical Industry.” Plan Art.\\nIII, § G(3)(a) (emphasis added). Interpreting “industry” to\\nmean “trade,” and “trade” to mean “line of work,” it is\\nreasonable to conclude that providing administrative support\\nto electricians is the “performance of services in [some]\\ncapacity in the Electrical [line of work].” After all, modern\\nelectrical work on large projects often cannot be performed\\nalone; it relies on teams working together that include\\nelectricians, supervisors, and administrators. O’Rourke’s\\nunion work for IBEW, while admittedly further afield from\\ndirect administrative support to electricians, is still focused\\non supporting electricians, and therefore reasonably within\\nthe electrical line of work.\\n\\n Second, the Board’s interpretation does not render any\\nother Plan provision nugatory. O’Rourke argues that the\\nBoard’s interpretation of “Electrical Industry” in Plan\\nArticle III renders nugatory its definition in Plan Article I.\\nHowever, this argument assumes that “Electrical Trade” in\\nArticle I has the meaning O’Rourke assigns to it. As\\nexplained above, while plausible, that is not necessarily the\\ncase because the Board’s interpretation of “trade” is\\nreasonable. While “trade” means “industry” under the\\nBoard’s interpretation, that circularity does not make the\\n\\f16 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nArticle I definition superfluous. Definitions sections in\\ncontracts necessarily contain some redundancy. That is\\ninherent in the nature of definitions because they explain\\nwhat other obligations mean, rather than creating additional\\nobligations themselves. Thus, when a term appears in the\\ndefinition of another term, it is reasonable for the terms to be\\nsynonymous. The situation here might be different if the\\nBoard’s definition of “industry” in Article III rendered\\nnugatory “trade” in another operative section of the plan, but\\nthat is not the case here.\\n\\n Third, the Board’s interpretation does not lack a rational\\nnexus to the Plan’s purpose. O’Rourke argues that the\\npurpose behind the Plan’s suspension of early retirement\\nbenefits is to prevent retirees who already have incomes\\nfrom competing with younger workers in the same line of\\nwork, and that he is not doing so. Thus, O’Rourke argues, he\\nis not competing with anyone else participating in the Plan\\nand the Board’s interpretation fails to advance the Plan’s\\npurpose. O’Rourke’s argument makes sense when\\nconsidering his identified purpose. However, the Board\\narticulated additional purposes in its denial letters, including\\nthat the Plan should “avoid providing an incentive for\\nexperienced electricians to leave fieldwork for an\\nadministrative position in the industry and the substantial\\nbenefit costs that would go with it.” The Board thus\\nexplained that preserving assets for retirement through a\\nbroad interpretation of prohibited employment was\\nconsistent with the Plan’s purposes, and that explanation was\\nreasonable.\\n\\n In sum, none of the three situations we identified in\\nTapley are present here. We therefore conclude that the\\nBoard’s interpretation of the Plan was reasonable. The\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 17\\n\\ndistrict court correctly entered summary judgment in the\\nBoard’s favor.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426347/\", \"author_raw\": \"WALLACE, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"J CLIFFORD WALLACE"},"judge2":{"kind":"string","value":"EUGENE E SILER"},"judge3":{"kind":"string","value":"M MARGARET MCKEOWN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649094/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: John O'Rourke v. No. California Electrical\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-16\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nJOHN J. O’ROURKE, No. 17-17419\n Plaintiff-Appellant,\n D.C. No.\n v. 3:16-cv-02007-\n WHO\nNORTHERN CALIFORNIA ELECTRICAL\nWORKERS PENSION PLAN; BOARD OF\nTRUSTEES OF THE NORTHERN OPINION\nCALIFORNIA ELECTRICAL WORKERS\nPENSION TRUST,\n Defendants-Appellees.\n\n Appeal from the United States District Court\n for the Northern District of California\n William Horsley Orrick, District Judge, Presiding\n\n Argued and Submitted March 15, 2019\n San Francisco, California\n\n Filed August 16, 2019\n\n Before: J. Clifford Wallace, Eugene E. Siler, *\n and M. Margaret McKeown, Circuit Judges.\n\n Opinion by Judge Wallace\n\n\n *\n The Honorable Eugene E. Siler, United States Circuit Judge for\nthe U.S. Court of Appeals for the Sixth Circuit, sitting by designation.\n\f2 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\n SUMMARY **\n\n\n ERISA\n\n The panel affirmed the district court’s grant of summary\njudgment in an ERISA action challenging the denial of\nplaintiff’s request for early retirement benefits.\n\n Plaintiff accrued benefits through a multiemployer\nERISA plan during his work as an electrician. When he left\nthis position to work for an electrical workers’ union as an\nadministrator, he sought early retirement benefits from the\nplan. The plan’s board of trustees decided that plaintiff’s\nunion work fell within the plan’s definition of “prohibited\nemployment,” and so no benefits were due for any month in\nwhich he engaged in that work.\n\n Reviewing the denial of benefits for an abuse of\ndiscretion, the panel held that any procedural irregularities\nin the actions of the board were minor and, at most, weighed\nonly slightly and weakly in favor of holding that an abuse of\ndiscretion occurred. The panel held that the board did not\nabuse its discretion in interpreting the plan’s definition of\nprohibited employment to include plaintiff’s union work\nbecause the board’s interpretation did not clearly conflict\nwith the plan’s plain language, did not render any other plan\nprovision nugatory, and did not lack a rational nexus to the\nplan’s purpose.\n\n\n\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 3\n\n COUNSEL\n\nTeresa S. Renaker (argued) and Margo Hasselman\nGreenough, Renaker Hasselman Scott LLP, San Francisco,\nCalifornia, for Plaintiff-Appellant.\n\nClarissa A. Kang (argued), R. Bradford Huss, and Angel L.\nGarrett, Trucker Huss, APC, San Francisco, California, for\nDefendants-Appellees.\n\n\n OPINION\n\nWALLACE, Circuit Judge:\n\n John O’Rourke is an electrician who accrued benefits\nthrough a multiemployer ERISA pension plan and was a\nmember of the plan’s board of trustees (Board). After\nO’Rourke left this position to work for an electrical workers’\nunion as an administrator, he sought early retirement\nbenefits from the plan. The Board denied his request.\nO’Rourke then filed this action for judicial review of the\ndenial, and the district court entered summary judgment in\nthe Board’s favor. We have jurisdiction over O’Rourke’s\nappeal from that judgment under 28 U.S.C. § 1291, and we\naffirm.\n\n I.\n\n A.\n\n The Northern California Electrical Workers Pension\nPlan (Plan) is a multiemployer plan governed by the\nEmployee Retirement Income Security Act (ERISA). See\n29 U.S.C. § 1002(37)(A). The Plan is funded through the\nNorthern California Electrical Workers Pension Trust\n\f4 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\n(Trust), to which participating employers contribute, and is\nadministered by the six-member Board, containing three\nlabor representatives and three management representatives.\nThe Plan provides its participants with three types of benefit\n— (1) normal pension, (2) early pension, and (3) disability\npension — payable as a monthly annuity when participants\nattain certain Plan-defined eligibility requirements. The Plan\nprovides that the Board shall resolve “[a]ny dispute as to\neligibility, type, amount or duration of benefits,” Plan Art.\nIX, § A, and grants the Board “the exclusive power and\ndiscretion to interpret the provisions of the Plan and any\nrules issued under the Plan, and to determine all questions\narising under the Plan including eligibility for benefits,” Plan\nArt. XIII, § A.\n\n Relevant to this appeal, the Plan provides that\nparticipants become eligible for early retirement and an early\npension at age fifty-five, if they have also accumulated ten\nor more years of covered employment. Plan Art. III,\n§ (B)(1). However, no benefit payments are made for either\nnormal pensions or early pensions “for any month in which\nthe Participant works in Prohibited Employment.” Plan Art.\nIII, § G(1). For participants who have not yet reached the age\nof sixty-five, the Plan defines “Prohibited Employment” as\n“the performance of services in any capacity in the Electrical\nIndustry.” Plan Art. III, § G(3)(a). In turn, “Electrical\nIndustry” is defined as “all branches of the Electrical Trade\nin the United States.” Plan. Art. I. “Electrical Trade” is not\ndefined.\n\n B.\n\n John O’Rourke is a participant in the Plan who began\nworking as an electrician in 1979. In 1999, he was elected\nbusiness manager of his local union, the International\nBrotherhood of Electrical Workers (IBEW) Local 6. As\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 5\n\nbusiness manager, O’Rourke served as a Plan trustee on the\nBoard.\n\n In September 2010, O’Rourke proposed that the Board\nadopt a change in the Plan’s suspension of benefits rules to\nexempt work for unions from the definition of prohibited\nemployment. O’Rourke recommended that the Plan Counsel\nreview the Plan and Trust and determine “what changes, if\nany, are required to accomplish this.” The Board concurred\nwith that recommendation, but no further action was taken\nat that time. O’Rourke then left his Local 6 and Board\nposition to join IBEW as an International Field\nRepresentative, followed by service as Vice President for the\nIBEW Ninth District. O’Rourke applied for an early\nretirement pension from the Plan in June 2014, in\nanticipation of his upcoming fifty-fifth birthday. It is\nundisputed that once O’Rourke turned fifty-five, he met all\nother eligibility requirements for the early pension, aside\nfrom the suspension of benefits provision. It is also\nundisputed that O’Rourke’s work for IBEW did not include\ntraditional work as an electrician, such as wiring, repair,\ninstallation, and maintenance.\n\n Meanwhile, Plan Counsel investigated the matter as\nrequested. In March 2014, Plan Counsel submitted a\nmemorandum to the Board opining that no changes to the\nPlan were necessary because the Plan’s prohibited\nemployment did not include work for a union. The Board\nconsidered the opinion at a meeting the following week,\nwhere two trustees expressed “strong disagreement with the\nlegal opinion,” indicating that “employment for Local 6, or\nany other IBEW for that matter, [should be] considered\n‘covered employment’ . . . subject to suspension of\nbenefits.” The Board did not resolve the matter at that\n\f6 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\nmeeting; instead it requested that Plan Counsel conduct\nfurther research on the matter.\n\n Plan Counsel submitted a second memorandum in\nAugust 2014. After reviewing past Board meeting minutes,\nPlan Counsel concluded that the Board had previously\ndefined prohibited employment “broadly to include work\nthat may not specifically bear upon or utilize skills and\nexperience obtained by work in the electrical trade.” Plan\nCounsel therefore opined that the Board should not adopt\nany policy of categorical exemption, and instead make each\ndetermination based on “all the facts and circumstances that\nmay be submitted regarding the nature of the work in\nquestion.”\n\n The Board discussed the second memorandum and\nO’Rourke’s application at its next meeting, resolved to defer\ndecision pending further information from O’Rourke, and\nrequested an explanation from him “as to why he believes\nthe work he is performing for [IBEW] is not prohibited under\nthe Plan.” O’Rourke responded by letter, explaining that his\nduties were “purely administrative” and did “not require,\ndirectly or indirectly, the use of the same skills employed by\nelectricians in an electrical trade or craft.”\n\n The Trustees discussed O’Rourke’s response via email\nin anticipation of their next meeting in December, with one\ntrustee stating that O’Rourke’s application was “a loser\nbarring an amendment to the plan.” At the next meeting, the\nBoard again discussed O’Rourke’s application and was\nagain unable to reach consensus. Plan Counsel then opined\nthat the suspension of benefits provision was ambiguous and\nsuggested clarifying its meaning so that the Board could\nreach consensus, to which the Board agreed.\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 7\n\n The following month, in January 2015, O’Rourke’s\ncounsel sent a demand letter to the Board stating O’Rourke’s\nposition that any interpretation of the Plan to include his\nIBEW employment would be an abuse of discretion.\nTrustees discussed the letter via email, with one trustee\nstating that O’Rourke’s description of his work as\n“administrative” was not outcome determinative, with a\npersonal reference to O’Rourke’s time as a trustee:\n\n First, to use the term “administrative”, as if\n it’s [sic] use in and of itself it [sic] would\n render his benefits payable, fails to address\n how the Trustees have interpreted the Plan in\n the past. Time and time again, Plan\n participants sought to have their benefits paid\n while working in and [sic] administrative\n capacity. Mr. O’Rourke should know that\n better than anyone since it was he that\n championed that determination.\n\nPlan Counsel also stated that “management has not relented\non the issue.”\n\n The Board met again in March 2015, and this time\nagreed to deny O’Rourke’s application. In a letter to\nO’Rourke’s attorney, the Board explained that O’Rourke’s\nIBEW employment fell within prohibited employment\nbecause it entailed “the performance of services of some\ncapacity in the Electrical Industry,” and no exception\napplied. O’Rourke appealed that decision. The Board met\nagain in August and denied the appeal. The Board explained\nthat it agreed there was no factual dispute over O’Rourke’s\nIBEW duties or that the Plan defined “Electrical Industry”\nas “Electrical Trade,” but that under its interpretation of\n\f8 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\n“trade,” O’Rourke’s work came within the prohibited\nemployment definition.\n\n O’Rourke filed this action against the Plan and the Board\nin April 2016. O’Rourke alleged that he was entitled to an\nearly pension and claimed benefits pursuant to ERISA. See\n29 U.S.C. § 1132(a)(1)(B). The parties cross-moved for\nsummary judgment and the district court granted the Board’s\nmotion and denied O’Rourke’s motion. The district court\nreasoned that the Board did not abuse its discretion by\ninterpreting the Plan to include O’Rourke’s IBEW position\nas prohibited employment because both parties’\ninterpretations were reasonable. O’Rourke timely appealed.\n\n II.\n\n “Where an ERISA Plan grants discretionary authority to\ndetermine eligibility for benefits or to construe the terms of\nthe plan, a plan administrator’s interpretation of a plan is\nreviewed for abuse of discretion.” Lehman v. Nelson, 862\nF.3d 1203, 1216 (9th Cir. 2017) (quoting Tapley v. Locals\n302 & 612 of Int’l Union of Operating Eng’rs-Emp’rs Const.\nIndus. Ret. Plan, 728 F.3d 1134, 1139 (9th Cir. 2013)). “We\nreview the district court’s application of this standard and the\ndistrict court’s . . . summary judgment de novo.” Id.\n\n III.\n\n O’Rourke argues that the Board incorrectly interpreted\nthe Plan to deny his application for benefits. The Plan\ncontains an express discretion provision, Plan Art. XIII, § A,\nand therefore our review is for abuse of discretion. Lehman,\n862 F.3d at 1216. Before we address the merits of this\nargument, however, we must decide how alleged procedural\nirregularities affect our analysis.\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 9\n\n A.\n\n O’Rourke contends that the Board acted “as an adversary\nbent on denying his claim,” see Friedrich v. Intel Corp., 181\nF.3d 1105, 1110 (9th Cir. 1999) (internal quotation marks\nomitted), rather than as a neutral arbiter acting in the Plan’s\nbest interests. According to him, we should view the Board’s\ndecision with suspicion, making it easier for us to conclude\nthe Board abused its discretion.\n\n We previously addressed the issue of procedural\nirregularity in Abatie v. Alta Health and Life Insurance, 458\nF.3d 955 (9th Cir. 2006) (en banc). In that case, we explained\nthat ERISA provides “for only two alternatives[: w]hen a\nplan confers discretion, abuse of discretion review applies;\nwhen it does not, de novo review applies.” Id. at 965. Thus,\nprocedural irregularity does not alter the standard of review\nexcept in “situations in which procedural irregularities are so\nsubstantial” as to make doing so necessary, such as “[w]hen\nan administrator engages in wholesale and flagrant\nviolations of the procedural requirements of ERISA, and\nthus acts in utter disregard of the underlying purpose of the\nplan as well.” Id. at 971. O’Rourke agrees that this case does\nnot present such a situation, and thus abuse of discretion\nreview applies.\n\n The question remains, however, whether any procedural\nirregularities here affect the abuse of discretion review itself.\nIn Abatie, we rejected a “sliding scale” approach to this\nproblem, where the greater the irregularity, the lesser the\nplaintiff’s burden. Id. at 967–68. Instead, we explained that\na “procedural irregularity, like a conflict of interest, is a\nmatter to be weighed in deciding whether an administrator’s\ndecision was an abuse of discretion.” Id. at 972. Thus,\n“[w]hen an administrator can show that it has engaged in an\nongoing, good faith exchange of information between the\n\f10 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\nadministrator and the claimant, the court should give the\nadministrator’s decision broad deference notwithstanding a\nminor irregularity.” Id. (internal quotation marks and\ncitation omitted). But when “[a] more serious procedural\nirregularity” has occurred, it “may weigh more heavily.” Id.\nat 972–73. We did not provide more guidance on the issue\nbeyond these statements, explaining that abuse of discretion\nreview is inherently “indefinite” and requires case-by-case\nanalysis. Id. at 969.\n\n In this case, as the district court concluded, the Plan is\nambiguous and both sides present plausible interpretations\nof the Plan language. If we were to accept O’Rourke’s\nassertions that the Board acted as an adversary against him,\nwe might accordingly hold that an abuse of discretion\noccurred where we otherwise might not. We must therefore\nfirst decide how much weight to place on the alleged\nprocedural irregularities in this case before proceeding.\n\n O’Rourke emphasizes four events as procedurally\nirregular: (1) emails and meeting minutes showing “political\nhostility” and personal animus towards him; (2) shifting\nrationales for the denial; (3) the rejection of Plan Counsel’s\n2014 opinion; and (4) the Board’s position reversal from the\n2010 resolution. We address each in turn.\n\n First, the record does not support O’Rourke’s assertion\nthat certain trustees harbored personal hostility towards him.\nThe meeting minutes and emails reflect that certain trustees\nstrongly disagreed with O’Rourke’s interpretation of the\nPlan, but nothing in the record rises to the level of animus.\nFor instance, calling an argument a “loser” is usually a\nreflection on the argument’s merits, not its proponent’s, as\nseems to be the case in the December email. Similarly,\nsaying that O’Rourke should have known the Board would\ninterpret the Plan in a way that did not favor him was\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 11\n\nirrelevant to the merits of O’Rourke’s demand letter, as an\napplicant’s knowledge or intent under the Plan does not\naffect their eligibility for benefits. Thus, we disagree with\nO’Rourke’s argument that the Board acted based on animus\nrather than in the Plan’s best interests.\n\n Second, the Board did not shift its rationales for denying\nthe claim. O’Rourke argues that the Board initially denied\nhis application because prohibited employment included\nsome employment that did not require an electrician’s skills,\nbut later denied his appeal because unions provide support\nto workers in the electrical trade. There is no discrepancy. In\ndenying the application, the Board explained that it was\nrelying on a broad definition of “industry” which included\nservices for IBEW. In denying the appeal, the Board\nexplained that it was relying on a broad definition of “trade,”\nwhich again included services for IBEW. The Plan defines\n“Electrical Industry” as “all branches of the Electrical Trade\nin the United States.” Plan. Art. I. Thus, the definition of\n“industry” in the first decision and “trade” in the second\ndecision is consistent with the Plan and the Board’s position\nthroughout its decision-making and this litigation.\n\n Third, while disregarding the advice of counsel might\nweigh in favor of holding that an abuse of discretion\noccurred in a different case, here the record reflects that all\nparties agreed the Plan was ambiguous and the disagreement\nwas in good faith. In its first memorandum to the Board, Plan\nCounsel opined that the Plan’s prohibited employment\ndefinition related to work involving the skills used by an\nelectrician. However, in the second memorandum, Plan\nCounsel acknowledged that Plan Counsel had not reviewed\nprevious determinations made by the Board when that\nopinion was reached, and that the Board had previously\ninterpreted prohibited employment “more broadly to include\n\f12 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\nwork that may not specifically bear upon or utilize skills and\nexperience obtained by work in the electrical trade.” Plan\nCounsel accordingly revised the original opinion in light of\nthe new information and recommended a case-by-case\nanalysis based on the totality of the circumstances\n“regarding the nature of the work in question.” Thus, while\nO’Rourke is correct that the Board did not follow Plan\nCounsel’s recommendation, a full picture of the record\nshows that this fact weighs only slightly against the Board,\nif at all.\n\n Fourth, the Board’s change in position from 2010 to\n2014 demonstrates only a change in opinion, not a\nprocedural irregularity. O’Rourke does not argue on appeal,\nnor could he, that the 2010 resolution to ask Plan Counsel to\ndetermine what changes were necessary to exempt union\nservice from prohibited employment constituted binding\naction. The motion was necessarily tentative because it\nrequired further action. Instead, the change in position shows\nthat the Board was once in favor of allowing early pension\nbenefits for IBEW workers, and it then changed its mind.\nSuch a change, in conjunction with other evidence showing\ndisregard of the Plan’s terms, could be evidence of bad faith.\nBut the only other evidence in the record supporting bad\nfaith is the Board’s rejection of Plan Counsel’s opinion. As\nexplained above, the Board’s rejection of Plan Counsel’s\nopinion does not show bad faith under the circumstances.\nThus, the Board’s change in position weighs only minimally\nin favor of holding that an abuse of discretion occurred here.\n\n Finally, we point out that the Board kept O’Rourke\ninformed at all stages of its decision-making and that none\nof the putative irregularities prevented the administrative\nrecord from being fully developed or prevented the Board or\na court from knowing all relevant facts. This record does not\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 13\n\nreflect that the Board attempted to keep O’Rourke in the dark\nor use procedural devices to prevent him from receiving\nbenefits.\n\n Thus, O’Rourke’s procedural irregularity argument\ndemonstrates at best only that the current Board held a\ndifferent interpretation of the Plan from O’Rourke and from\npast Boards, and that it appeared unwilling to change that\ninterpretation on counsel’s advice or O’Rourke’s arguments.\nWhile this could perhaps indicate some stubbornness beyond\nthe platonic ideal of an open-minded plan administrator, it\ndoes not reveal serious procedural irregularities that would\nweigh heavily against the Board. At most, the Board’s\nactions weigh only slightly and weakly in favor of holding\nthat an abuse of discretion occurred here.\n\n B.\n\n Having concluded that any procedural irregularities were\nminor, we turn to the Board’s interpretation of the Plan itself.\nThe Plan defines “Prohibited Employment” as “the\nperformance of services in any capacity in the Electrical\nIndustry.” Plan Art. III, § G(3)(a). O’Rourke argues that by\ndefining “Electrical Industry” to mean “all branches of the\nElectrical Trade in the United States,” the Plan limits\nprohibited employment to work involving the skills of an\nelectrician. The Board responds that it does not matter\nwhether the work involves an electrician’s skills, so long as\nit involves any type of livelihood related to electrical work,\nincluding administrative work. The question in this appeal is\nthus whether the Board abused its discretion by interpreting\n“performance of services in any capacity in the Electrical\nIndustry” to include working for IBEW; i.e., in an\nadministrative capacity for an electrical workers’ union. To\nanswer that question, we ask whether the Board’s\ninterpretation is unreasonable, “closely read[ing] contested\n\f14 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\nterms and appl[ying] contract principles derived from state\nlaw, guided by the policies expressed in ERISA and other\nfederal labor laws.” Tapley, 728 F.3d at 1140 (internal\nquotation marks and alterations omitted) (quoting\nRichardson v. Pension Plan of Bethlehem Steel Corp., 112\nF.3d 982, 985 (9th Cir. 1997)). We will accept the Board’s\ninterpretation unless it is “not grounded on any reasonable\nbasis.” Id. at 1139 (emphasis in original) (quoting Oster v.\nBarco of Cal. Emps.’ Ret. Plan, 869 F.2d 1215, 1218 (9th\nCir. 1988)). “The [Board’s] interpretation need not be the\none this court would have reached, but only an interpretation\nwhich has rational justifications.” Id. at 1139–40 (quoting\nSmith v. CMTA-IAM Pension Trust, 654 F.2d 650, 655 (9th\nCir. 1981)).\n\n In Tapley, we described three ways in which a Plan\nadministrator’s interpretation might fail this test. First, it is\nan abuse of discretion to “construe provisions of a plan in a\nway that clearly conflicts with the plain language of the\nPlan.” Id. at 1140 (alteration omitted) (quoting Johnson v.\nTrs. of W. Conf. of Teamsters Pension Tr. Fund, 879 F.2d\n651, 654 (9th Cir. 1989)). Second, it is an abuse of discretion\nto interpret a provision in a way that “renders nugatory other\nprovisions of the Plan.” Id. (internal quotation marks and\nalterations omitted) (quoting Richardson, 112 F.3d at 985).\nThird, it is an abuse of discretion to give an interpretation\nthat “lacks any rational nexus to the primary purpose of the\nPlan.” Id. (quotation marks omitted) (quoting Burditt v. W.\nGrowers Pension Plan, 636 F. Supp. 1491, 1498 (C.D. Cal.\n1986)). These three types of abuse correspond more or less\nto basic principles of contract interpretation; i.e., (1) to\ninterpret provisions based on their plain meaning, (2) to\ninterpret provisions in context with other provisions, and\n(3) to interpret provisions to reach a reasonable outcome that\naccords with the expectations of the agreeing parties.\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 15\n\n O’Rourke argues that all three Tapley errors are present\nhere and require us to reject the Board’s interpretation. We\ndisagree with each of his arguments.\n\n First, the Board’s interpretation does not clearly conflict\nwith the Plan’s plain language. O’Rourke makes a strong\nargument for why the best interpretation of “trade” in the\nPlan is working as an electrician. However, he fails to\nexclude the Board’s interpretation as a reasonable\npossibility. The operative language in the Plan is very broad:\nit defines prohibited employment as “the performance of\nservices in any capacity in the Electrical Industry.” Plan Art.\nIII, § G(3)(a) (emphasis added). Interpreting “industry” to\nmean “trade,” and “trade” to mean “line of work,” it is\nreasonable to conclude that providing administrative support\nto electricians is the “performance of services in [some]\ncapacity in the Electrical [line of work].” After all, modern\nelectrical work on large projects often cannot be performed\nalone; it relies on teams working together that include\nelectricians, supervisors, and administrators. O’Rourke’s\nunion work for IBEW, while admittedly further afield from\ndirect administrative support to electricians, is still focused\non supporting electricians, and therefore reasonably within\nthe electrical line of work.\n\n Second, the Board’s interpretation does not render any\nother Plan provision nugatory. O’Rourke argues that the\nBoard’s interpretation of “Electrical Industry” in Plan\nArticle III renders nugatory its definition in Plan Article I.\nHowever, this argument assumes that “Electrical Trade” in\nArticle I has the meaning O’Rourke assigns to it. As\nexplained above, while plausible, that is not necessarily the\ncase because the Board’s interpretation of “trade” is\nreasonable. While “trade” means “industry” under the\nBoard’s interpretation, that circularity does not make the\n\f16 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\n\nArticle I definition superfluous. Definitions sections in\ncontracts necessarily contain some redundancy. That is\ninherent in the nature of definitions because they explain\nwhat other obligations mean, rather than creating additional\nobligations themselves. Thus, when a term appears in the\ndefinition of another term, it is reasonable for the terms to be\nsynonymous. The situation here might be different if the\nBoard’s definition of “industry” in Article III rendered\nnugatory “trade” in another operative section of the plan, but\nthat is not the case here.\n\n Third, the Board’s interpretation does not lack a rational\nnexus to the Plan’s purpose. O’Rourke argues that the\npurpose behind the Plan’s suspension of early retirement\nbenefits is to prevent retirees who already have incomes\nfrom competing with younger workers in the same line of\nwork, and that he is not doing so. Thus, O’Rourke argues, he\nis not competing with anyone else participating in the Plan\nand the Board’s interpretation fails to advance the Plan’s\npurpose. O’Rourke’s argument makes sense when\nconsidering his identified purpose. However, the Board\narticulated additional purposes in its denial letters, including\nthat the Plan should “avoid providing an incentive for\nexperienced electricians to leave fieldwork for an\nadministrative position in the industry and the substantial\nbenefit costs that would go with it.” The Board thus\nexplained that preserving assets for retirement through a\nbroad interpretation of prohibited employment was\nconsistent with the Plan’s purposes, and that explanation was\nreasonable.\n\n In sum, none of the three situations we identified in\nTapley are present here. We therefore conclude that the\nBoard’s interpretation of the Plan was reasonable. The\n\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 17\n\ndistrict court correctly entered summary judgment in the\nBoard’s favor.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: John O'Rourke v. No. California Electrical\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-16\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nJOHN J. O’ROURKE, No. 17-17419\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 3:16-cv-02007-\\n WHO\\nNORTHERN CALIFORNIA ELECTRICAL\\nWORKERS PENSION PLAN; BOARD OF\\nTRUSTEES OF THE NORTHERN OPINION\\nCALIFORNIA ELECTRICAL WORKERS\\nPENSION TRUST,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n William Horsley Orrick, District Judge, Presiding\\n\\n Argued and Submitted March 15, 2019\\n San Francisco, California\\n\\n Filed August 16, 2019\\n\\n Before: J. Clifford Wallace, Eugene E. Siler, *\\n and M. Margaret McKeown, Circuit Judges.\\n\\n Opinion by Judge Wallace\\n\\n\\n *\\n The Honorable Eugene E. Siler, United States Circuit Judge for\\nthe U.S. Court of Appeals for the Sixth Circuit, sitting by designation.\\n\\f2 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\n SUMMARY **\\n\\n\\n ERISA\\n\\n The panel affirmed the district court’s grant of summary\\njudgment in an ERISA action challenging the denial of\\nplaintiff’s request for early retirement benefits.\\n\\n Plaintiff accrued benefits through a multiemployer\\nERISA plan during his work as an electrician. When he left\\nthis position to work for an electrical workers’ union as an\\nadministrator, he sought early retirement benefits from the\\nplan. The plan’s board of trustees decided that plaintiff’s\\nunion work fell within the plan’s definition of “prohibited\\nemployment,” and so no benefits were due for any month in\\nwhich he engaged in that work.\\n\\n Reviewing the denial of benefits for an abuse of\\ndiscretion, the panel held that any procedural irregularities\\nin the actions of the board were minor and, at most, weighed\\nonly slightly and weakly in favor of holding that an abuse of\\ndiscretion occurred. The panel held that the board did not\\nabuse its discretion in interpreting the plan’s definition of\\nprohibited employment to include plaintiff’s union work\\nbecause the board’s interpretation did not clearly conflict\\nwith the plan’s plain language, did not render any other plan\\nprovision nugatory, and did not lack a rational nexus to the\\nplan’s purpose.\\n\\n\\n\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 3\\n\\n COUNSEL\\n\\nTeresa S. Renaker (argued) and Margo Hasselman\\nGreenough, Renaker Hasselman Scott LLP, San Francisco,\\nCalifornia, for Plaintiff-Appellant.\\n\\nClarissa A. Kang (argued), R. Bradford Huss, and Angel L.\\nGarrett, Trucker Huss, APC, San Francisco, California, for\\nDefendants-Appellees.\\n\\n\\n OPINION\\n\\nWALLACE, Circuit Judge:\\n\\n John O’Rourke is an electrician who accrued benefits\\nthrough a multiemployer ERISA pension plan and was a\\nmember of the plan’s board of trustees (Board). After\\nO’Rourke left this position to work for an electrical workers’\\nunion as an administrator, he sought early retirement\\nbenefits from the plan. The Board denied his request.\\nO’Rourke then filed this action for judicial review of the\\ndenial, and the district court entered summary judgment in\\nthe Board’s favor. We have jurisdiction over O’Rourke’s\\nappeal from that judgment under 28 U.S.C. § 1291, and we\\naffirm.\\n\\n I.\\n\\n A.\\n\\n The Northern California Electrical Workers Pension\\nPlan (Plan) is a multiemployer plan governed by the\\nEmployee Retirement Income Security Act (ERISA). See\\n29 U.S.C. § 1002(37)(A). The Plan is funded through the\\nNorthern California Electrical Workers Pension Trust\\n\\f4 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\n(Trust), to which participating employers contribute, and is\\nadministered by the six-member Board, containing three\\nlabor representatives and three management representatives.\\nThe Plan provides its participants with three types of benefit\\n— (1) normal pension, (2) early pension, and (3) disability\\npension — payable as a monthly annuity when participants\\nattain certain Plan-defined eligibility requirements. The Plan\\nprovides that the Board shall resolve “[a]ny dispute as to\\neligibility, type, amount or duration of benefits,” Plan Art.\\nIX, § A, and grants the Board “the exclusive power and\\ndiscretion to interpret the provisions of the Plan and any\\nrules issued under the Plan, and to determine all questions\\narising under the Plan including eligibility for benefits,” Plan\\nArt. XIII, § A.\\n\\n Relevant to this appeal, the Plan provides that\\nparticipants become eligible for early retirement and an early\\npension at age fifty-five, if they have also accumulated ten\\nor more years of covered employment. Plan Art. III,\\n§ (B)(1). However, no benefit payments are made for either\\nnormal pensions or early pensions “for any month in which\\nthe Participant works in Prohibited Employment.” Plan Art.\\nIII, § G(1). For participants who have not yet reached the age\\nof sixty-five, the Plan defines “Prohibited Employment” as\\n“the performance of services in any capacity in the Electrical\\nIndustry.” Plan Art. III, § G(3)(a). In turn, “Electrical\\nIndustry” is defined as “all branches of the Electrical Trade\\nin the United States.” Plan. Art. I. “Electrical Trade” is not\\ndefined.\\n\\n B.\\n\\n John O’Rourke is a participant in the Plan who began\\nworking as an electrician in 1979. In 1999, he was elected\\nbusiness manager of his local union, the International\\nBrotherhood of Electrical Workers (IBEW) Local 6. As\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 5\\n\\nbusiness manager, O’Rourke served as a Plan trustee on the\\nBoard.\\n\\n In September 2010, O’Rourke proposed that the Board\\nadopt a change in the Plan’s suspension of benefits rules to\\nexempt work for unions from the definition of prohibited\\nemployment. O’Rourke recommended that the Plan Counsel\\nreview the Plan and Trust and determine “what changes, if\\nany, are required to accomplish this.” The Board concurred\\nwith that recommendation, but no further action was taken\\nat that time. O’Rourke then left his Local 6 and Board\\nposition to join IBEW as an International Field\\nRepresentative, followed by service as Vice President for the\\nIBEW Ninth District. O’Rourke applied for an early\\nretirement pension from the Plan in June 2014, in\\nanticipation of his upcoming fifty-fifth birthday. It is\\nundisputed that once O’Rourke turned fifty-five, he met all\\nother eligibility requirements for the early pension, aside\\nfrom the suspension of benefits provision. It is also\\nundisputed that O’Rourke’s work for IBEW did not include\\ntraditional work as an electrician, such as wiring, repair,\\ninstallation, and maintenance.\\n\\n Meanwhile, Plan Counsel investigated the matter as\\nrequested. In March 2014, Plan Counsel submitted a\\nmemorandum to the Board opining that no changes to the\\nPlan were necessary because the Plan’s prohibited\\nemployment did not include work for a union. The Board\\nconsidered the opinion at a meeting the following week,\\nwhere two trustees expressed “strong disagreement with the\\nlegal opinion,” indicating that “employment for Local 6, or\\nany other IBEW for that matter, [should be] considered\\n‘covered employment’ . . . subject to suspension of\\nbenefits.” The Board did not resolve the matter at that\\n\\f6 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nmeeting; instead it requested that Plan Counsel conduct\\nfurther research on the matter.\\n\\n Plan Counsel submitted a second memorandum in\\nAugust 2014. After reviewing past Board meeting minutes,\\nPlan Counsel concluded that the Board had previously\\ndefined prohibited employment “broadly to include work\\nthat may not specifically bear upon or utilize skills and\\nexperience obtained by work in the electrical trade.” Plan\\nCounsel therefore opined that the Board should not adopt\\nany policy of categorical exemption, and instead make each\\ndetermination based on “all the facts and circumstances that\\nmay be submitted regarding the nature of the work in\\nquestion.”\\n\\n The Board discussed the second memorandum and\\nO’Rourke’s application at its next meeting, resolved to defer\\ndecision pending further information from O’Rourke, and\\nrequested an explanation from him “as to why he believes\\nthe work he is performing for [IBEW] is not prohibited under\\nthe Plan.” O’Rourke responded by letter, explaining that his\\nduties were “purely administrative” and did “not require,\\ndirectly or indirectly, the use of the same skills employed by\\nelectricians in an electrical trade or craft.”\\n\\n The Trustees discussed O’Rourke’s response via email\\nin anticipation of their next meeting in December, with one\\ntrustee stating that O’Rourke’s application was “a loser\\nbarring an amendment to the plan.” At the next meeting, the\\nBoard again discussed O’Rourke’s application and was\\nagain unable to reach consensus. Plan Counsel then opined\\nthat the suspension of benefits provision was ambiguous and\\nsuggested clarifying its meaning so that the Board could\\nreach consensus, to which the Board agreed.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 7\\n\\n The following month, in January 2015, O’Rourke’s\\ncounsel sent a demand letter to the Board stating O’Rourke’s\\nposition that any interpretation of the Plan to include his\\nIBEW employment would be an abuse of discretion.\\nTrustees discussed the letter via email, with one trustee\\nstating that O’Rourke’s description of his work as\\n“administrative” was not outcome determinative, with a\\npersonal reference to O’Rourke’s time as a trustee:\\n\\n First, to use the term “administrative”, as if\\n it’s [sic] use in and of itself it [sic] would\\n render his benefits payable, fails to address\\n how the Trustees have interpreted the Plan in\\n the past. Time and time again, Plan\\n participants sought to have their benefits paid\\n while working in and [sic] administrative\\n capacity. Mr. O’Rourke should know that\\n better than anyone since it was he that\\n championed that determination.\\n\\nPlan Counsel also stated that “management has not relented\\non the issue.”\\n\\n The Board met again in March 2015, and this time\\nagreed to deny O’Rourke’s application. In a letter to\\nO’Rourke’s attorney, the Board explained that O’Rourke’s\\nIBEW employment fell within prohibited employment\\nbecause it entailed “the performance of services of some\\ncapacity in the Electrical Industry,” and no exception\\napplied. O’Rourke appealed that decision. The Board met\\nagain in August and denied the appeal. The Board explained\\nthat it agreed there was no factual dispute over O’Rourke’s\\nIBEW duties or that the Plan defined “Electrical Industry”\\nas “Electrical Trade,” but that under its interpretation of\\n\\f8 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\n“trade,” O’Rourke’s work came within the prohibited\\nemployment definition.\\n\\n O’Rourke filed this action against the Plan and the Board\\nin April 2016. O’Rourke alleged that he was entitled to an\\nearly pension and claimed benefits pursuant to ERISA. See\\n29 U.S.C. § 1132(a)(1)(B). The parties cross-moved for\\nsummary judgment and the district court granted the Board’s\\nmotion and denied O’Rourke’s motion. The district court\\nreasoned that the Board did not abuse its discretion by\\ninterpreting the Plan to include O’Rourke’s IBEW position\\nas prohibited employment because both parties’\\ninterpretations were reasonable. O’Rourke timely appealed.\\n\\n II.\\n\\n “Where an ERISA Plan grants discretionary authority to\\ndetermine eligibility for benefits or to construe the terms of\\nthe plan, a plan administrator’s interpretation of a plan is\\nreviewed for abuse of discretion.” Lehman v. Nelson, 862\\nF.3d 1203, 1216 (9th Cir. 2017) (quoting Tapley v. Locals\\n302 & 612 of Int’l Union of Operating Eng’rs-Emp’rs Const.\\nIndus. Ret. Plan, 728 F.3d 1134, 1139 (9th Cir. 2013)). “We\\nreview the district court’s application of this standard and the\\ndistrict court’s . . . summary judgment de novo.” Id.\\n\\n III.\\n\\n O’Rourke argues that the Board incorrectly interpreted\\nthe Plan to deny his application for benefits. The Plan\\ncontains an express discretion provision, Plan Art. XIII, § A,\\nand therefore our review is for abuse of discretion. Lehman,\\n862 F.3d at 1216. Before we address the merits of this\\nargument, however, we must decide how alleged procedural\\nirregularities affect our analysis.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 9\\n\\n A.\\n\\n O’Rourke contends that the Board acted “as an adversary\\nbent on denying his claim,” see Friedrich v. Intel Corp., 181\\nF.3d 1105, 1110 (9th Cir. 1999) (internal quotation marks\\nomitted), rather than as a neutral arbiter acting in the Plan’s\\nbest interests. According to him, we should view the Board’s\\ndecision with suspicion, making it easier for us to conclude\\nthe Board abused its discretion.\\n\\n We previously addressed the issue of procedural\\nirregularity in Abatie v. Alta Health and Life Insurance, 458\\nF.3d 955 (9th Cir. 2006) (en banc). In that case, we explained\\nthat ERISA provides “for only two alternatives[: w]hen a\\nplan confers discretion, abuse of discretion review applies;\\nwhen it does not, de novo review applies.” Id. at 965. Thus,\\nprocedural irregularity does not alter the standard of review\\nexcept in “situations in which procedural irregularities are so\\nsubstantial” as to make doing so necessary, such as “[w]hen\\nan administrator engages in wholesale and flagrant\\nviolations of the procedural requirements of ERISA, and\\nthus acts in utter disregard of the underlying purpose of the\\nplan as well.” Id. at 971. O’Rourke agrees that this case does\\nnot present such a situation, and thus abuse of discretion\\nreview applies.\\n\\n The question remains, however, whether any procedural\\nirregularities here affect the abuse of discretion review itself.\\nIn Abatie, we rejected a “sliding scale” approach to this\\nproblem, where the greater the irregularity, the lesser the\\nplaintiff’s burden. Id. at 967–68. Instead, we explained that\\na “procedural irregularity, like a conflict of interest, is a\\nmatter to be weighed in deciding whether an administrator’s\\ndecision was an abuse of discretion.” Id. at 972. Thus,\\n“[w]hen an administrator can show that it has engaged in an\\nongoing, good faith exchange of information between the\\n\\f10 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nadministrator and the claimant, the court should give the\\nadministrator’s decision broad deference notwithstanding a\\nminor irregularity.” Id. (internal quotation marks and\\ncitation omitted). But when “[a] more serious procedural\\nirregularity” has occurred, it “may weigh more heavily.” Id.\\nat 972–73. We did not provide more guidance on the issue\\nbeyond these statements, explaining that abuse of discretion\\nreview is inherently “indefinite” and requires case-by-case\\nanalysis. Id. at 969.\\n\\n In this case, as the district court concluded, the Plan is\\nambiguous and both sides present plausible interpretations\\nof the Plan language. If we were to accept O’Rourke’s\\nassertions that the Board acted as an adversary against him,\\nwe might accordingly hold that an abuse of discretion\\noccurred where we otherwise might not. We must therefore\\nfirst decide how much weight to place on the alleged\\nprocedural irregularities in this case before proceeding.\\n\\n O’Rourke emphasizes four events as procedurally\\nirregular: (1) emails and meeting minutes showing “political\\nhostility” and personal animus towards him; (2) shifting\\nrationales for the denial; (3) the rejection of Plan Counsel’s\\n2014 opinion; and (4) the Board’s position reversal from the\\n2010 resolution. We address each in turn.\\n\\n First, the record does not support O’Rourke’s assertion\\nthat certain trustees harbored personal hostility towards him.\\nThe meeting minutes and emails reflect that certain trustees\\nstrongly disagreed with O’Rourke’s interpretation of the\\nPlan, but nothing in the record rises to the level of animus.\\nFor instance, calling an argument a “loser” is usually a\\nreflection on the argument’s merits, not its proponent’s, as\\nseems to be the case in the December email. Similarly,\\nsaying that O’Rourke should have known the Board would\\ninterpret the Plan in a way that did not favor him was\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 11\\n\\nirrelevant to the merits of O’Rourke’s demand letter, as an\\napplicant’s knowledge or intent under the Plan does not\\naffect their eligibility for benefits. Thus, we disagree with\\nO’Rourke’s argument that the Board acted based on animus\\nrather than in the Plan’s best interests.\\n\\n Second, the Board did not shift its rationales for denying\\nthe claim. O’Rourke argues that the Board initially denied\\nhis application because prohibited employment included\\nsome employment that did not require an electrician’s skills,\\nbut later denied his appeal because unions provide support\\nto workers in the electrical trade. There is no discrepancy. In\\ndenying the application, the Board explained that it was\\nrelying on a broad definition of “industry” which included\\nservices for IBEW. In denying the appeal, the Board\\nexplained that it was relying on a broad definition of “trade,”\\nwhich again included services for IBEW. The Plan defines\\n“Electrical Industry” as “all branches of the Electrical Trade\\nin the United States.” Plan. Art. I. Thus, the definition of\\n“industry” in the first decision and “trade” in the second\\ndecision is consistent with the Plan and the Board’s position\\nthroughout its decision-making and this litigation.\\n\\n Third, while disregarding the advice of counsel might\\nweigh in favor of holding that an abuse of discretion\\noccurred in a different case, here the record reflects that all\\nparties agreed the Plan was ambiguous and the disagreement\\nwas in good faith. In its first memorandum to the Board, Plan\\nCounsel opined that the Plan’s prohibited employment\\ndefinition related to work involving the skills used by an\\nelectrician. However, in the second memorandum, Plan\\nCounsel acknowledged that Plan Counsel had not reviewed\\nprevious determinations made by the Board when that\\nopinion was reached, and that the Board had previously\\ninterpreted prohibited employment “more broadly to include\\n\\f12 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nwork that may not specifically bear upon or utilize skills and\\nexperience obtained by work in the electrical trade.” Plan\\nCounsel accordingly revised the original opinion in light of\\nthe new information and recommended a case-by-case\\nanalysis based on the totality of the circumstances\\n“regarding the nature of the work in question.” Thus, while\\nO’Rourke is correct that the Board did not follow Plan\\nCounsel’s recommendation, a full picture of the record\\nshows that this fact weighs only slightly against the Board,\\nif at all.\\n\\n Fourth, the Board’s change in position from 2010 to\\n2014 demonstrates only a change in opinion, not a\\nprocedural irregularity. O’Rourke does not argue on appeal,\\nnor could he, that the 2010 resolution to ask Plan Counsel to\\ndetermine what changes were necessary to exempt union\\nservice from prohibited employment constituted binding\\naction. The motion was necessarily tentative because it\\nrequired further action. Instead, the change in position shows\\nthat the Board was once in favor of allowing early pension\\nbenefits for IBEW workers, and it then changed its mind.\\nSuch a change, in conjunction with other evidence showing\\ndisregard of the Plan’s terms, could be evidence of bad faith.\\nBut the only other evidence in the record supporting bad\\nfaith is the Board’s rejection of Plan Counsel’s opinion. As\\nexplained above, the Board’s rejection of Plan Counsel’s\\nopinion does not show bad faith under the circumstances.\\nThus, the Board’s change in position weighs only minimally\\nin favor of holding that an abuse of discretion occurred here.\\n\\n Finally, we point out that the Board kept O’Rourke\\ninformed at all stages of its decision-making and that none\\nof the putative irregularities prevented the administrative\\nrecord from being fully developed or prevented the Board or\\na court from knowing all relevant facts. This record does not\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 13\\n\\nreflect that the Board attempted to keep O’Rourke in the dark\\nor use procedural devices to prevent him from receiving\\nbenefits.\\n\\n Thus, O’Rourke’s procedural irregularity argument\\ndemonstrates at best only that the current Board held a\\ndifferent interpretation of the Plan from O’Rourke and from\\npast Boards, and that it appeared unwilling to change that\\ninterpretation on counsel’s advice or O’Rourke’s arguments.\\nWhile this could perhaps indicate some stubbornness beyond\\nthe platonic ideal of an open-minded plan administrator, it\\ndoes not reveal serious procedural irregularities that would\\nweigh heavily against the Board. At most, the Board’s\\nactions weigh only slightly and weakly in favor of holding\\nthat an abuse of discretion occurred here.\\n\\n B.\\n\\n Having concluded that any procedural irregularities were\\nminor, we turn to the Board’s interpretation of the Plan itself.\\nThe Plan defines “Prohibited Employment” as “the\\nperformance of services in any capacity in the Electrical\\nIndustry.” Plan Art. III, § G(3)(a). O’Rourke argues that by\\ndefining “Electrical Industry” to mean “all branches of the\\nElectrical Trade in the United States,” the Plan limits\\nprohibited employment to work involving the skills of an\\nelectrician. The Board responds that it does not matter\\nwhether the work involves an electrician’s skills, so long as\\nit involves any type of livelihood related to electrical work,\\nincluding administrative work. The question in this appeal is\\nthus whether the Board abused its discretion by interpreting\\n“performance of services in any capacity in the Electrical\\nIndustry” to include working for IBEW; i.e., in an\\nadministrative capacity for an electrical workers’ union. To\\nanswer that question, we ask whether the Board’s\\ninterpretation is unreasonable, “closely read[ing] contested\\n\\f14 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nterms and appl[ying] contract principles derived from state\\nlaw, guided by the policies expressed in ERISA and other\\nfederal labor laws.” Tapley, 728 F.3d at 1140 (internal\\nquotation marks and alterations omitted) (quoting\\nRichardson v. Pension Plan of Bethlehem Steel Corp., 112\\nF.3d 982, 985 (9th Cir. 1997)). We will accept the Board’s\\ninterpretation unless it is “not grounded on any reasonable\\nbasis.” Id. at 1139 (emphasis in original) (quoting Oster v.\\nBarco of Cal. Emps.’ Ret. Plan, 869 F.2d 1215, 1218 (9th\\nCir. 1988)). “The [Board’s] interpretation need not be the\\none this court would have reached, but only an interpretation\\nwhich has rational justifications.” Id. at 1139–40 (quoting\\nSmith v. CMTA-IAM Pension Trust, 654 F.2d 650, 655 (9th\\nCir. 1981)).\\n\\n In Tapley, we described three ways in which a Plan\\nadministrator’s interpretation might fail this test. First, it is\\nan abuse of discretion to “construe provisions of a plan in a\\nway that clearly conflicts with the plain language of the\\nPlan.” Id. at 1140 (alteration omitted) (quoting Johnson v.\\nTrs. of W. Conf. of Teamsters Pension Tr. Fund, 879 F.2d\\n651, 654 (9th Cir. 1989)). Second, it is an abuse of discretion\\nto interpret a provision in a way that “renders nugatory other\\nprovisions of the Plan.” Id. (internal quotation marks and\\nalterations omitted) (quoting Richardson, 112 F.3d at 985).\\nThird, it is an abuse of discretion to give an interpretation\\nthat “lacks any rational nexus to the primary purpose of the\\nPlan.” Id. (quotation marks omitted) (quoting Burditt v. W.\\nGrowers Pension Plan, 636 F. Supp. 1491, 1498 (C.D. Cal.\\n1986)). These three types of abuse correspond more or less\\nto basic principles of contract interpretation; i.e., (1) to\\ninterpret provisions based on their plain meaning, (2) to\\ninterpret provisions in context with other provisions, and\\n(3) to interpret provisions to reach a reasonable outcome that\\naccords with the expectations of the agreeing parties.\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 15\\n\\n O’Rourke argues that all three Tapley errors are present\\nhere and require us to reject the Board’s interpretation. We\\ndisagree with each of his arguments.\\n\\n First, the Board’s interpretation does not clearly conflict\\nwith the Plan’s plain language. O’Rourke makes a strong\\nargument for why the best interpretation of “trade” in the\\nPlan is working as an electrician. However, he fails to\\nexclude the Board’s interpretation as a reasonable\\npossibility. The operative language in the Plan is very broad:\\nit defines prohibited employment as “the performance of\\nservices in any capacity in the Electrical Industry.” Plan Art.\\nIII, § G(3)(a) (emphasis added). Interpreting “industry” to\\nmean “trade,” and “trade” to mean “line of work,” it is\\nreasonable to conclude that providing administrative support\\nto electricians is the “performance of services in [some]\\ncapacity in the Electrical [line of work].” After all, modern\\nelectrical work on large projects often cannot be performed\\nalone; it relies on teams working together that include\\nelectricians, supervisors, and administrators. O’Rourke’s\\nunion work for IBEW, while admittedly further afield from\\ndirect administrative support to electricians, is still focused\\non supporting electricians, and therefore reasonably within\\nthe electrical line of work.\\n\\n Second, the Board’s interpretation does not render any\\nother Plan provision nugatory. O’Rourke argues that the\\nBoard’s interpretation of “Electrical Industry” in Plan\\nArticle III renders nugatory its definition in Plan Article I.\\nHowever, this argument assumes that “Electrical Trade” in\\nArticle I has the meaning O’Rourke assigns to it. As\\nexplained above, while plausible, that is not necessarily the\\ncase because the Board’s interpretation of “trade” is\\nreasonable. While “trade” means “industry” under the\\nBoard’s interpretation, that circularity does not make the\\n\\f16 O’ROURKE V. N. CAL. ELEC. WORKERS PENSION\\n\\nArticle I definition superfluous. Definitions sections in\\ncontracts necessarily contain some redundancy. That is\\ninherent in the nature of definitions because they explain\\nwhat other obligations mean, rather than creating additional\\nobligations themselves. Thus, when a term appears in the\\ndefinition of another term, it is reasonable for the terms to be\\nsynonymous. The situation here might be different if the\\nBoard’s definition of “industry” in Article III rendered\\nnugatory “trade” in another operative section of the plan, but\\nthat is not the case here.\\n\\n Third, the Board’s interpretation does not lack a rational\\nnexus to the Plan’s purpose. O’Rourke argues that the\\npurpose behind the Plan’s suspension of early retirement\\nbenefits is to prevent retirees who already have incomes\\nfrom competing with younger workers in the same line of\\nwork, and that he is not doing so. Thus, O’Rourke argues, he\\nis not competing with anyone else participating in the Plan\\nand the Board’s interpretation fails to advance the Plan’s\\npurpose. O’Rourke’s argument makes sense when\\nconsidering his identified purpose. However, the Board\\narticulated additional purposes in its denial letters, including\\nthat the Plan should “avoid providing an incentive for\\nexperienced electricians to leave fieldwork for an\\nadministrative position in the industry and the substantial\\nbenefit costs that would go with it.” The Board thus\\nexplained that preserving assets for retirement through a\\nbroad interpretation of prohibited employment was\\nconsistent with the Plan’s purposes, and that explanation was\\nreasonable.\\n\\n In sum, none of the three situations we identified in\\nTapley are present here. We therefore conclude that the\\nBoard’s interpretation of the Plan was reasonable. The\\n\\f O’ROURKE V. N. CAL. ELEC. WORKERS PENSION 17\\n\\ndistrict court correctly entered summary judgment in the\\nBoard’s favor.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4076,"cells":{"id":{"kind":"number","value":4649095,"string":"4,649,095"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"East Bay Sanctuary Covenant v. William Barr"},"decision_date":{"kind":"string","value":"2019-08-16"},"docket_number":{"kind":"string","value":"19-16487"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"type\": \"010combined\", \"text\": \"FOR PUBLICATION FILED\\n UNITED STATES COURT OF APPEALS AUG 16 2019\\n MOLLY C. DWYER, CLERK\\n U.S. COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\nEAST BAY SANCTUARY COVENANT; No. 19-16487\\nAL OTRO LADO; INNOVATION LAW\\nLAB; CENTRAL AMERICAN D.C. No. 3:19-cv-04073-JST\\nRESOURCE CENTER, Northern District of California,\\n San Francisco\\n Plaintiffs-Appellees,\\n ORDER\\n v.\\n\\nWILLIAM P. BARR, Attorney General;\\nUNITED STATES DEPARTMENT OF\\nJUSTICE; JAMES MCHENRY, Director of\\nthe Executive Office for Immigration\\nReview, in his official capacity;\\nEXECUTIVE OFFICE FOR\\nIMMIGRATION REVIEW; KEVIN K.\\nMCALEENAN, Acting Secretary of\\nHomeland Security, in his official capacity;\\nU.S. DEPARTMENT OF HOMELAND\\nSECURITY; KENNETH T. CUCCINELLI,\\nActing Director of the U.S. Citizenship and\\nImmigration Services, in his official\\ncapacity; JOHN P. SANDERS, Acting\\nCommissioner of U.S. Customs and Border\\nProtection, in his official capacity; UNITED\\nSTATES CITIZENSHIP AND\\nIMMIGRATION SERVICES; U.S.\\nCUSTOMS AND BORDER\\nPROTECTION; MATTHEW ALBENCE,\\nActing Director of Immigration and\\nCustoms Enforcement, in his official\\ncapacity; IMMIGRATION AND\\nCUSTOMS ENFORCEMENT,\\n\\n Defendants-Appellants.\\n\\nBefore: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\\n\\n Appellants seek a stay pending appeal of the district court’s July 24, 2019\\n\\norder preliminarily enjoining the Department of Justice and Department of\\n\\nHomeland Security’s joint interim final rule, “Asylum Eligibility and Procedural\\n\\nModifications” (the “Rule”), 84 Fed. Reg. 33,829 (July 16, 2019).1\\n\\n The district court found that the Rule likely did not comply with the\\n\\nAdministrative Procedure Act’s (APA) notice-and-comment and 30-day grace\\n\\nperiod requirements because Appellants did not adequately support invocation of\\n\\nthe “good cause” and “foreign affairs” exemptions under the APA. See 5 U.S.C.\\n\\n§ 553(a)(1), (b)(1)(B), (d)(3); Buschmann v. Schweiker, 676 F.2d 352, 357 (9th\\n\\nCir. 1982) (good cause exemption “should be interpreted narrowly so that the\\n\\nexception will not swallow the rule” (internal citations omitted)); Yassini v.\\n\\nCrosland, 618 F.2d 1356, 1360 n.4 (9th Cir. 1980) (foreign affairs exemption\\n\\n“would become distended” if applied to immigration rules generally and requires\\n\\nshowing that ordinary public noticing would “provoke definitely undesirable\\n\\n\\n1\\n The State of Arizona’s amicus brief in support of Appellants’ motion has\\nbeen filed. The Professors of Immigration Law’s motion for leave to file an\\namicus brief in opposition to Appellants’ motion (Docket Entry No. 28) is granted,\\nand the brief is filed.\\n\\n\\n 2 19-16487\\n\\ninternational consequences”). We conclude that Appellants have not made the\\n\\nrequired “strong showing” that they are likely to succeed on the merits on this\\n\\nissue. Hilton v. Braunskill, 481 U.S. 770, 776 (1987).2\\n\\n Consequently, we deny the motion for stay pending appeal (Docket Entry\\n\\nNo. 3) insofar as the injunction applies within the Ninth Circuit.3\\n\\n We grant the motion for stay pending appeal insofar as the injunction applies\\n\\noutside the Ninth Circuit, because the nationwide scope of the injunction is not\\n\\n\\n\\n\\n2\\n Our finding that Appellants have not made a “strong showing” does not bind\\nthe merits panel in reviewing this aspect of the merits, as that is not the standard\\nthe merits panel will apply. See Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7,\\n20 (2008).\\n3\\n We do not assess Appellants’ remaining arguments as to likelihood of\\nsuccess on the merits and do not reach the remaining Hilton factors. See Nken v.\\nHolder, 556 U.S. 418, 435 (2009) (stating that the likelihood of success on the\\nmerits factor is one of the “most critical” and must be established before\\nconsidering the last two stay factors); cf. California v. Azar, 911 F.3d 558, 575 (9th\\nCir. 2018) (“Likelihood of success on the merits is the most important factor; if a\\nmovant fails to meet this threshold inquiry, we need not consider the other factors.”\\n(internal quotation marks omitted) (quoting Disney Enters., Inc. v. VidAngel, Inc.,\\n869 F.3d 848, 856 (9th Cir. 2017))).\\n\\n\\n 3 19-16487\\n\\nsupported by the record as it stands.4 Cf. City and County of San Francisco v.\\n\\nTrump, 897 F.3d 1225, 1243–45 (9th Cir. 2018).5\\n\\n An injunction must be “narrowly tailored to remedy the specific harm\\n\\nshown.” Id. at 1244 (quoting Bresgal v. Brock, 843 F.2d 1163, 1170 (9th Cir.\\n\\n1987)). We have upheld nationwide injunctions where such breadth was necessary\\n\\nto remedy a plaintiff’s harm. See, e.g., id.; California v. Azar, 911 F.3d 558, 582\\n\\n(9th Cir. 2018) (“Although there is no bar against nationwide relief in federal\\n\\ndistrict court . . . such broad relief must be necessary to give prevailing parties the\\n\\n\\n4\\n The dissent, without citing any authority, argues that “it is [not] within a\\nmotions panel’s province to parse the record for error at this stage” and accuses us\\nof “[going] beyond the recognized authority of a motions panel” by granting the\\nmotion for a stay pending appeal insofar as the injunction applies outside the Ninth\\nCircuit. We have two responses.\\n First, we did not have to “parse” the record for error. Appellants’ stay\\nmotion specifically argues that the district court erred in imposing a nationwide\\ninjunction. Moreover, the three sentences that the district court provided to\\nsupport the imposition of a nationwide injunction—none of which explains why it\\nbelieved a nationwide injunction was necessary in this case—make clear that it\\nfailed to undertake the analysis necessary before granting such broad relief.\\n Second, other motions panels of our court have reviewed the scope of\\ninjunctive relief granted by district courts. See, e.g., E. Bay Sanctuary Covenant v.\\nTrump, No. 18-17274, 2018 WL 8807133, at *24 (9th Cir. Dec. 7, 2018); Hawaii\\nv. Trump, No. 17-17168, 2017 WL 5343014, at *1 (9th Cir. Nov. 13, 2017). We\\nthink these decisions illustrate that it is indeed within our province—our duty,\\neven—to review whether the district court abused its discretion in granting a\\nnationwide injunction.\\n5\\n The dissent criticizes our reliance on Trump, 897 F.3d 1225, because the\\nprocedural posture in this case is different. We recognize this difference as we cite\\nTrump as an analogous case supporting our decision because, notwithstanding the\\ndifferent procedural posture, the issue in that case—whether the scope of the\\ninjunction was appropriate—is the same issue before us. See id. at 1244–45.\\n\\n 4 19-16487\\n\\nrelief to which they are entitled.” (internal quotation marks and alterations omitted)\\n\\n(quoting Bresgal, 843 F.2d at 1170–71)). These are, however, “exceptional cases.”\\n\\nTrump, 897 F.3d at 1244. To permit such broad injunctions as a general rule,\\n\\nwithout an articulated connection to a plaintiff’s particular harm, would\\n\\nunnecessarily “stymie novel legal challenges and robust debate” arising in different\\n\\njudicial districts. Id.; see also Azar, 911 F.3d at 583 (“The Supreme Court has\\n\\nrepeatedly emphasized that nationwide injunctions have detrimental consequences\\n\\nto the development of law and deprive appellate courts of a wider range of\\n\\nperspectives.”).\\n\\n Here, the district court failed to discuss whether a nationwide injunction is\\n\\nnecessary to remedy Plaintiffs’ alleged harm. Instead, in conclusory fashion, the\\n\\ndistrict court stated that nationwide relief is warranted simply because district\\n\\ncourts have the authority to impose such relief in some cases and because such\\n\\nrelief has been applied in the immigration context. The district court clearly erred\\n\\nby failing to consider whether nationwide relief is necessary to remedy Plaintiffs’\\n\\nalleged harms. And, based on the limited record before us, we do not believe a\\n\\nnationwide injunction is justified.\\n\\n Our dissenting colleague believes that a nationwide injunction is appropriate\\n\\nsimply because this case presents a rule that applies nationwide. That view,\\n\\nhowever, ignores our well-established rule that injunctive relief “must be tailored\\n\\n\\n 5 19-16487\\n\\nto remedy the specific harm alleged.” Lamb-Weston, Inc. v. McCain Foods, Ltd.,\\n\\n941 F.2d 970, 974 (9th Cir. 1991) (citations omitted). Indeed, were we to adopt\\n\\nthe dissent’s view, a nationwide injunction would result any time an enjoined\\n\\naction has potential nationwide effects. Such an approach would turn broad\\n\\ninjunctions into the rule rather than the exception. Under our case law, however,\\n\\nall injunctions—even ones involving national policies—must be “narrowly tailored\\n\\nto remedy the specific harm shown.” Trump, 897 F.3d at 1244 (quoting Bresgal,\\n\\n843 F.2d at 1170).\\n\\n We agree with our dissenting colleague that “time does not permit a full\\n\\nexploration of the merits of the ‘nationwide’ issue.” But whereas he believes that\\n\\nsuch a factor supports the granting of a nationwide injunction until a merits panel\\n\\ncan address the case, we reach precisely the opposite conclusion. “National\\n\\ninjunctions interfere with good decisionmaking by the federal judiciary.” Samuel\\n\\nL. Bray, Multiple Chancellors: Reforming the National Injunction, 131 Harv. L.\\n\\nRev. 417, 461 (2017). They “deprive” other parties of “the right to litigate in other\\n\\nforums.” Azar, 911 F.3d at 583. Based on the briefing and limited record before\\n\\nus, and absent an explanation by the district court as to why a nationwide\\n\\ninjunction is necessary to remedy Plaintiffs’ alleged harm in this case, we must\\n\\ngrant the motion for stay pending appeal insofar as the injunction applies outside\\n\\nthe Ninth Circuit.\\n\\n\\n 6 19-16487\"}, {\"type\": \"dissent\", \"author\": \"Our\", \"text\": \"Our dissenting colleague also argues that it is “perplexing” that the\\n\\ngovernment’s failure to demonstrate a strong showing of likelihood of success on\\n\\nthe merits “does not . . . require that a stay of the nationwide aspect of the\\n\\ninjunction [] be denied.” That contention misses the mark, however, by conflating\\n\\nthe merits of the government’s position with district court’s authority to issue a\\n\\nnationwide injunction. Whether Appellants have made a strong showing of\\n\\nlikelihood of success on the merits entitling them to a stay of the preliminary\\n\\ninjunction is a separate question from whether the scope of the injunction is\\n\\nappropriate. In Azar, for example, we affirmed the preliminary injunction because,\\n\\namong other things, we found that the plaintiffs were likely to succeed on their\\n\\nclaim that the rules were invalid. 911 F.3d at 575–81. Despite our conclusion that\\n\\nthe rules were likely invalid, however, we also determined that the injunction’s\\n\\nnationwide scope was not supported by the record. Id. at 584–85. Azar illustrates\\n\\nthat, beyond examining the merits of a party’s arguments, a district court must\\n\\nseparately analyze whether nationwide relief is “necessary to give prevailing\\n\\nparties the relief to which they are entitled” before issuing such an injunction. Id.\\n\\nat 582 (quoting Bresgal, 843 F.2d at 1170–71).\\n\\n\\n\\n\\n 7 19-16487\\n\\n Our approach—granting a more limited injunction—allows other litigants\\n\\nwishing to challenge the Rule to do so. Indeed, several already have.6 Litigation\\n\\nover the Rule’s lawfulness will promote “the development of the law and the\\n\\npercolation of legal issues in the lower courts” and allow the Supreme Court, if it\\n\\nchooses to address the Rule, to do so “[with] the benefit of additional viewpoints\\n\\nfrom other lower federal courts and [with] a fully developed factual record.”\\n\\nAmanda Frost, In Defense of Nationwide Injunctions, 93 N.Y.U. L. REV. 1065,\\n\\n1107–08 (2018).7\\n\\n In sum, our decision to partially grant the stay simply upholds the law of our\\n\\ncircuit by ensuring that injunctive relief is properly tailored to the alleged harm.8\\n\\n While this appeal proceeds, the district court retains jurisdiction to further\\n\\ndevelop the record in support of a preliminary injunction extending beyond the\\n\\n\\n6\\n As Appellants point out, hours before the district court ruled here, a District\\nof Columbia district court, presented with the same Rule, denied materially\\nidentical relief to organizations similar to the Plaintiffs here. See CAIR v. Trump,\\nNo. 1:19-CV-02117-TJK, 2019 WL 3436501 (D.D.C. July 24, 2019).\\n7\\n Accord United States v. Mendoza, 464 U.S. 154, 160 (1984) (“[O]nly one\\nfinal adjudication would deprive this Court of the benefit it receives from\\npermitting several courts of appeals to explore a difficult question.”).\\n8\\n Contrary to the dissent’s position, the fact that injunctive relief may\\ntemporarily cause the Rule to be administered inconsistently in different locations\\nis not a sound reason for imposing relief that is broader than necessary. As we\\nexplain above, our law requires that injunctive relief be narrowly tailored to\\nremedy the plaintiffs’ alleged harm, and it may only be broadened “if such breadth\\nis necessary to give prevailing parties the relief to which they are entitled.”\\nBresgal, 843 F.2d at 1170–71 (emphasis added).\\n\\n\\n 8 19-16487\\n\\nNinth Circuit. Cf. Trump, 897 F.3d at 1245 (“Because the record is insufficiently\\n\\ndeveloped as to the question of the national scope of the injunction, we vacate the\\n\\ninjunction to the extent that it applies outside California and remand to the district\\n\\ncourt for a more searching inquiry into whether this case justifies the breadth of the\\n\\ninjunction imposed.”).\\n\\n The opening brief is due September 3, 2019; the answering brief is due\\n\\nOctober 1, 2019; and the optional reply is due within 21 days after service of the\\n\\nanswering brief. This case will be placed on a December 2019 argument calendar.\\n\\n\\n\\n\\n 9 19-16487\\n\\n FILED\\nEast Bay Sanctuary Covenant v. Barr, No. 19-16487\\n AUG 16 2019\"}, {\"author\": \"TASHIMA, Circuit Judge, concurring\", \"type\": \"concurrence\", \"text\": \"TASHIMA, Circuit Judge, concurring in part and dissenting in part: MOLLY C. DWYER, CLERK\\n U.S. COURT OF APPEALS\\n\\n\\n I concur in the portion of the order denying the motion for stay pending\\n\\nappeal [Dkt. 3] insofar as the injunction applies within the Ninth Circuit, but\\n\\ndissent from the balance of the order.\\n\\n Acting as a motions panel, all we have before us is the government’s motion\\n\\nfor a stay. I do not believe that it is within a motions panel’s province to parse the\\n\\nrecord for error at this stage, which is what the majority does in concluding that\\n\\n“the nationwide scope of the injunction is not supported by the record as it stands.”\\n\\n(Citation omitted.) But the majority then goes beyond the recognized authority of\\n\\na motions panel by concluding that “[t]he district court clearly erred by failing to\\n\\nconsider whether nationwide relief is necessary to remedy Plaintiffs’ alleged\\n\\nharms,” and, on that basis “grant[s] the motion for stay pending appeal insofar as\\n\\nthe injunction applies outside the Ninth Circuit.” It then, in the penultimate\\n\\nparagraph of the Order, in effect, remands the case to the district court for a partial\\n\\ndo-over:\\n\\n While this appeal proceeds, the district court retains jurisdiction to\\n further develop the record in support of a preliminary injunction extending\\n beyond the Ninth Circuit.1\\n\\n 1\\n The majority relies on City & County of San Francisco v. Trump, 897\\nF.3d 1225, 1245 (9th Cir. 2018), but the citation is completely inapposite. That\\n (continued...)\\n\\nBut vacating and remanding it to the district court for a more searching inquiry into\\n\\nwhether this case justifies the breadth of the injunction is indubitably an action\\n\\nwithin the province of a merits panel—not a motions panel.2\\n\\n At the same time, the order places the merits briefing (of this appeal) on an\\n\\nexpedited schedule for placement “on a December 2019 argument calendar.” What\\n\\nissues are the parties expected to brief, assuming that parallel proceedings in the\\n\\ndistrict court are still ongoing? And if the district court completes its second-look\\n\\nremand proceedings within the next few weeks or months and issues a modified\\n\\ninjunction, or issues the same nationwide injunction, but one which is supported by\\n\\nsupplemental findings of fact, should the parties seek to file supplemental briefs on\\n\\nthe newly-raised and newly-decided issues in this appeal to the merits panel\\n\\nassigned to this appeal, or should a new notice of appeal be filed, giving rise to a\\n\\n\\n 1\\n (...continued)\\nwas an opinion by a merits panel charged with deciding the appeal, not a motions\\npanel charged with deciding a stay motion, and the merits panel did exactly what it\\nwas charged with, i.e., it decided the appeal; it “AFFIRMED in part;VACATED in\\npari; and REMANDED.” Id. We, as a motions panel, have no equivalent charge.\\n 2\\n Because the issue has been decided, applying the clear error standard\\nof review, the injunction vacated and remanded to the district court, the merits\\npanel, presumably has been deprived of deciding this issue. The majority’s\\nassertion, in footnote 2, that its action “does not bind the merits panel,” is an empty\\npromise. Deciding the case on the merits, vacating and remanding the injunction is\\nnot in accord with the dictates of Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7,\\n20 (2008).\\n\\n Page -2-\\n\\nnew appeal? These are some of the new and difficult questions raised by the\\n\\nmajority’s split-decision.\\n\\n While time does not permit a full exploration of the merits of the\\n\\n“nationwide” issue, some problems posed by the majority’s Ninth Circuit-only\\n\\ninjunction are apparent. Perhaps, the district court did not make detailed findings\\n\\nin support of a nationwide injunction because the need for one in the circumstances\\n\\nof this case is obvious. For starters, the joint interim final rule, “Asylum Eligibility\\n\\nand Procedural Modifications,” will affect asylum applications across the breadth\\n\\nof the southern border. Should asylum law be administered differently in Texas\\n\\nthan in California? These issues and problems illustrate why tinkering with the\\n\\nmerits on a limited stay motion record can be risky. And it is why such issues are\\n\\nreserved for the more deliberate examination that a merits panel can give them.\\n\\n There is also a glaring inconsistency—a contradiction—in the majority’s\\n\\nsplit-the-baby approach. If, as the majority and I agree, the government’s failure to\\n\\nmeet the first Hilton v. Braunskill, 481 U.S. 770, 776 (1987), factor —likelihood of\\n\\nsuccess on the merits, because of its failure to comply with the APA—means that\\n\\nits stay motion with respect to the preliminary injunction’s application within the\\n\\nNinth Circuit fails, it is perplexing to me why that failure does not infect the\\n\\nbalance of its stay motion and require that a stay of the nationwide aspect of the\\n\\n Page -3-\\n\\ninjunction also be denied.3 The majority, in its rush to address the merits of the\\n\\nnationwide aspect of the injunction, simply elides this contradiction.\\n\\n Because I would not peel off part of the preliminary injunction and remand\\n\\nthat portion to the district court, “[b]ecause the record is insufficiently developed\\n\\nas to the question of the national scope of the injunction” (quoting San Francisco\\n\\nv. Trump, 897 F.3d at 1245), while retaining jurisdiction over the remainder, I\\n\\ndissent from the remand4 of the nationwide scope of the preliminary injunction to\\n\\nthe district court.\\n\\n I would simply deny the stay motion.\\n\\n\\n\\n\\n 3\\n The majority’s answer to this point is to state that “Whether\\nAppellants have made a strong showing of likelihood of success on the merits\\nentitling them to a stay of the preliminary injunction is a separate question from\\nwhether the scope of the injunction is appropriate.” But that doesn’t answer (or\\neven try to answer) my question of why the government’s failure to meet the\\nlikelihood-of-success factor doesn’t doom its motion to stay the nationwide portion\\nof the injunction, as well as the California portion.\\n 4\\n The Order does not use the word “remand,” but the majority does not\\nquarrel with the obvious inference from its statement that “the district court retains\\njurisdiction to further develop the record in support of a preliminary injunction\\nextending beyond the Ninth Circuit,” is, in substance, a remand.\\n\\n Page -4-\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426348/\", \"author_raw\": \"TASHIMA, Circuit Judge, concurring\"}]}"},"judge1":{"kind":"string","value":"TASHIMA"},"judge2":{"kind":"string","value":"M SMITH"},"judge3":{"kind":"string","value":"BENNETT"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{\"TASHIMA\": \", Circuit\", \"M SMITH\": \", Circuit\", \"BENNETT\": \", Circuit\"}"},"opinions_count":{"kind":"number","value":3,"string":"3"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":1,"string":"1"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649095/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: East Bay Sanctuary Covenant v. William Barr\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-16\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: Yes\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION FILED\n UNITED STATES COURT OF APPEALS AUG 16 2019\n MOLLY C. DWYER, CLERK\n U.S. COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\nEAST BAY SANCTUARY COVENANT; No. 19-16487\nAL OTRO LADO; INNOVATION LAW\nLAB; CENTRAL AMERICAN D.C. No. 3:19-cv-04073-JST\nRESOURCE CENTER, Northern District of California,\n San Francisco\n Plaintiffs-Appellees,\n ORDER\n v.\n\nWILLIAM P. BARR, Attorney General;\nUNITED STATES DEPARTMENT OF\nJUSTICE; JAMES MCHENRY, Director of\nthe Executive Office for Immigration\nReview, in his official capacity;\nEXECUTIVE OFFICE FOR\nIMMIGRATION REVIEW; KEVIN K.\nMCALEENAN, Acting Secretary of\nHomeland Security, in his official capacity;\nU.S. DEPARTMENT OF HOMELAND\nSECURITY; KENNETH T. CUCCINELLI,\nActing Director of the U.S. Citizenship and\nImmigration Services, in his official\ncapacity; JOHN P. SANDERS, Acting\nCommissioner of U.S. Customs and Border\nProtection, in his official capacity; UNITED\nSTATES CITIZENSHIP AND\nIMMIGRATION SERVICES; U.S.\nCUSTOMS AND BORDER\nPROTECTION; MATTHEW ALBENCE,\nActing Director of Immigration and\nCustoms Enforcement, in his official\ncapacity; IMMIGRATION AND\nCUSTOMS ENFORCEMENT,\n\n Defendants-Appellants.\n\nBefore: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\n\n Appellants seek a stay pending appeal of the district court’s July 24, 2019\n\norder preliminarily enjoining the Department of Justice and Department of\n\nHomeland Security’s joint interim final rule, “Asylum Eligibility and Procedural\n\nModifications” (the “Rule”), 84 Fed. Reg. 33,829 (July 16, 2019).1\n\n The district court found that the Rule likely did not comply with the\n\nAdministrative Procedure Act’s (APA) notice-and-comment and 30-day grace\n\nperiod requirements because Appellants did not adequately support invocation of\n\nthe “good cause” and “foreign affairs” exemptions under the APA. See 5 U.S.C.\n\n§ 553(a)(1), (b)(1)(B), (d)(3); Buschmann v. Schweiker, 676 F.2d 352, 357 (9th\n\nCir. 1982) (good cause exemption “should be interpreted narrowly so that the\n\nexception will not swallow the rule” (internal citations omitted)); Yassini v.\n\nCrosland, 618 F.2d 1356, 1360 n.4 (9th Cir. 1980) (foreign affairs exemption\n\n“would become distended” if applied to immigration rules generally and requires\n\nshowing that ordinary public noticing would “provoke definitely undesirable\n\n\n1\n The State of Arizona’s amicus brief in support of Appellants’ motion has\nbeen filed. The Professors of Immigration Law’s motion for leave to file an\namicus brief in opposition to Appellants’ motion (Docket Entry No. 28) is granted,\nand the brief is filed.\n\n\n 2 19-16487\n\ninternational consequences”). We conclude that Appellants have not made the\n\nrequired “strong showing” that they are likely to succeed on the merits on this\n\nissue. Hilton v. Braunskill, 481 U.S. 770, 776 (1987).2\n\n Consequently, we deny the motion for stay pending appeal (Docket Entry\n\nNo. 3) insofar as the injunction applies within the Ninth Circuit.3\n\n We grant the motion for stay pending appeal insofar as the injunction applies\n\noutside the Ninth Circuit, because the nationwide scope of the injunction is not\n\n\n\n\n2\n Our finding that Appellants have not made a “strong showing” does not bind\nthe merits panel in reviewing this aspect of the merits, as that is not the standard\nthe merits panel will apply. See Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7,\n20 (2008).\n3\n We do not assess Appellants’ remaining arguments as to likelihood of\nsuccess on the merits and do not reach the remaining Hilton factors. See Nken v.\nHolder, 556 U.S. 418, 435 (2009) (stating that the likelihood of success on the\nmerits factor is one of the “most critical” and must be established before\nconsidering the last two stay factors); cf. California v. Azar, 911 F.3d 558, 575 (9th\nCir. 2018) (“Likelihood of success on the merits is the most important factor; if a\nmovant fails to meet this threshold inquiry, we need not consider the other factors.”\n(internal quotation marks omitted) (quoting Disney Enters., Inc. v. VidAngel, Inc.,\n869 F.3d 848, 856 (9th Cir. 2017))).\n\n\n 3 19-16487\n\nsupported by the record as it stands.4 Cf. City and County of San Francisco v.\n\nTrump, 897 F.3d 1225, 1243–45 (9th Cir. 2018).5\n\n An injunction must be “narrowly tailored to remedy the specific harm\n\nshown.” Id. at 1244 (quoting Bresgal v. Brock, 843 F.2d 1163, 1170 (9th Cir.\n\n1987)). We have upheld nationwide injunctions where such breadth was necessary\n\nto remedy a plaintiff’s harm. See, e.g., id.; California v. Azar, 911 F.3d 558, 582\n\n(9th Cir. 2018) (“Although there is no bar against nationwide relief in federal\n\ndistrict court . . . such broad relief must be necessary to give prevailing parties the\n\n\n4\n The dissent, without citing any authority, argues that “it is [not] within a\nmotions panel’s province to parse the record for error at this stage” and accuses us\nof “[going] beyond the recognized authority of a motions panel” by granting the\nmotion for a stay pending appeal insofar as the injunction applies outside the Ninth\nCircuit. We have two responses.\n First, we did not have to “parse” the record for error. Appellants’ stay\nmotion specifically argues that the district court erred in imposing a nationwide\ninjunction. Moreover, the three sentences that the district court provided to\nsupport the imposition of a nationwide injunction—none of which explains why it\nbelieved a nationwide injunction was necessary in this case—make clear that it\nfailed to undertake the analysis necessary before granting such broad relief.\n Second, other motions panels of our court have reviewed the scope of\ninjunctive relief granted by district courts. See, e.g., E. Bay Sanctuary Covenant v.\nTrump, No. 18-17274, 2018 WL 8807133, at *24 (9th Cir. Dec. 7, 2018); Hawaii\nv. Trump, No. 17-17168, 2017 WL 5343014, at *1 (9th Cir. Nov. 13, 2017). We\nthink these decisions illustrate that it is indeed within our province—our duty,\neven—to review whether the district court abused its discretion in granting a\nnationwide injunction.\n5\n The dissent criticizes our reliance on Trump, 897 F.3d 1225, because the\nprocedural posture in this case is different. We recognize this difference as we cite\nTrump as an analogous case supporting our decision because, notwithstanding the\ndifferent procedural posture, the issue in that case—whether the scope of the\ninjunction was appropriate—is the same issue before us. See id. at 1244–45.\n\n 4 19-16487\n\nrelief to which they are entitled.” (internal quotation marks and alterations omitted)\n\n(quoting Bresgal, 843 F.2d at 1170–71)). These are, however, “exceptional cases.”\n\nTrump, 897 F.3d at 1244. To permit such broad injunctions as a general rule,\n\nwithout an articulated connection to a plaintiff’s particular harm, would\n\nunnecessarily “stymie novel legal challenges and robust debate” arising in different\n\njudicial districts. Id.; see also Azar, 911 F.3d at 583 (“The Supreme Court has\n\nrepeatedly emphasized that nationwide injunctions have detrimental consequences\n\nto the development of law and deprive appellate courts of a wider range of\n\nperspectives.”).\n\n Here, the district court failed to discuss whether a nationwide injunction is\n\nnecessary to remedy Plaintiffs’ alleged harm. Instead, in conclusory fashion, the\n\ndistrict court stated that nationwide relief is warranted simply because district\n\ncourts have the authority to impose such relief in some cases and because such\n\nrelief has been applied in the immigration context. The district court clearly erred\n\nby failing to consider whether nationwide relief is necessary to remedy Plaintiffs’\n\nalleged harms. And, based on the limited record before us, we do not believe a\n\nnationwide injunction is justified.\n\n Our dissenting colleague believes that a nationwide injunction is appropriate\n\nsimply because this case presents a rule that applies nationwide. That view,\n\nhowever, ignores our well-established rule that injunctive relief “must be tailored\n\n\n 5 19-16487\n\nto remedy the specific harm alleged.” Lamb-Weston, Inc. v. McCain Foods, Ltd.,\n\n941 F.2d 970, 974 (9th Cir. 1991) (citations omitted). Indeed, were we to adopt\n\nthe dissent’s view, a nationwide injunction would result any time an enjoined\n\naction has potential nationwide effects. Such an approach would turn broad\n\ninjunctions into the rule rather than the exception. Under our case law, however,\n\nall injunctions—even ones involving national policies—must be “narrowly tailored\n\nto remedy the specific harm shown.” Trump, 897 F.3d at 1244 (quoting Bresgal,\n\n843 F.2d at 1170).\n\n We agree with our dissenting colleague that “time does not permit a full\n\nexploration of the merits of the ‘nationwide’ issue.” But whereas he believes that\n\nsuch a factor supports the granting of a nationwide injunction until a merits panel\n\ncan address the case, we reach precisely the opposite conclusion. “National\n\ninjunctions interfere with good decisionmaking by the federal judiciary.” Samuel\n\nL. Bray, Multiple Chancellors: Reforming the National Injunction, 131 Harv. L.\n\nRev. 417, 461 (2017). They “deprive” other parties of “the right to litigate in other\n\nforums.” Azar, 911 F.3d at 583. Based on the briefing and limited record before\n\nus, and absent an explanation by the district court as to why a nationwide\n\ninjunction is necessary to remedy Plaintiffs’ alleged harm in this case, we must\n\ngrant the motion for stay pending appeal insofar as the injunction applies outside\n\nthe Ninth Circuit.\n\n\n 6 19-16487\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: East Bay Sanctuary Covenant v. William Barr\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-16\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: Yes\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION FILED\\n UNITED STATES COURT OF APPEALS AUG 16 2019\\n MOLLY C. DWYER, CLERK\\n U.S. COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\nEAST BAY SANCTUARY COVENANT; No. 19-16487\\nAL OTRO LADO; INNOVATION LAW\\nLAB; CENTRAL AMERICAN D.C. No. 3:19-cv-04073-JST\\nRESOURCE CENTER, Northern District of California,\\n San Francisco\\n Plaintiffs-Appellees,\\n ORDER\\n v.\\n\\nWILLIAM P. BARR, Attorney General;\\nUNITED STATES DEPARTMENT OF\\nJUSTICE; JAMES MCHENRY, Director of\\nthe Executive Office for Immigration\\nReview, in his official capacity;\\nEXECUTIVE OFFICE FOR\\nIMMIGRATION REVIEW; KEVIN K.\\nMCALEENAN, Acting Secretary of\\nHomeland Security, in his official capacity;\\nU.S. DEPARTMENT OF HOMELAND\\nSECURITY; KENNETH T. CUCCINELLI,\\nActing Director of the U.S. Citizenship and\\nImmigration Services, in his official\\ncapacity; JOHN P. SANDERS, Acting\\nCommissioner of U.S. Customs and Border\\nProtection, in his official capacity; UNITED\\nSTATES CITIZENSHIP AND\\nIMMIGRATION SERVICES; U.S.\\nCUSTOMS AND BORDER\\nPROTECTION; MATTHEW ALBENCE,\\nActing Director of Immigration and\\nCustoms Enforcement, in his official\\ncapacity; IMMIGRATION AND\\nCUSTOMS ENFORCEMENT,\\n\\n Defendants-Appellants.\\n\\nBefore: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\\n\\n Appellants seek a stay pending appeal of the district court’s July 24, 2019\\n\\norder preliminarily enjoining the Department of Justice and Department of\\n\\nHomeland Security’s joint interim final rule, “Asylum Eligibility and Procedural\\n\\nModifications” (the “Rule”), 84 Fed. Reg. 33,829 (July 16, 2019).1\\n\\n The district court found that the Rule likely did not comply with the\\n\\nAdministrative Procedure Act’s (APA) notice-and-comment and 30-day grace\\n\\nperiod requirements because Appellants did not adequately support invocation of\\n\\nthe “good cause” and “foreign affairs” exemptions under the APA. See 5 U.S.C.\\n\\n§ 553(a)(1), (b)(1)(B), (d)(3); Buschmann v. Schweiker, 676 F.2d 352, 357 (9th\\n\\nCir. 1982) (good cause exemption “should be interpreted narrowly so that the\\n\\nexception will not swallow the rule” (internal citations omitted)); Yassini v.\\n\\nCrosland, 618 F.2d 1356, 1360 n.4 (9th Cir. 1980) (foreign affairs exemption\\n\\n“would become distended” if applied to immigration rules generally and requires\\n\\nshowing that ordinary public noticing would “provoke definitely undesirable\\n\\n\\n1\\n The State of Arizona’s amicus brief in support of Appellants’ motion has\\nbeen filed. The Professors of Immigration Law’s motion for leave to file an\\namicus brief in opposition to Appellants’ motion (Docket Entry No. 28) is granted,\\nand the brief is filed.\\n\\n\\n 2 19-16487\\n\\ninternational consequences”). We conclude that Appellants have not made the\\n\\nrequired “strong showing” that they are likely to succeed on the merits on this\\n\\nissue. Hilton v. Braunskill, 481 U.S. 770, 776 (1987).2\\n\\n Consequently, we deny the motion for stay pending appeal (Docket Entry\\n\\nNo. 3) insofar as the injunction applies within the Ninth Circuit.3\\n\\n We grant the motion for stay pending appeal insofar as the injunction applies\\n\\noutside the Ninth Circuit, because the nationwide scope of the injunction is not\\n\\n\\n\\n\\n2\\n Our finding that Appellants have not made a “strong showing” does not bind\\nthe merits panel in reviewing this aspect of the merits, as that is not the standard\\nthe merits panel will apply. See Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7,\\n20 (2008).\\n3\\n We do not assess Appellants’ remaining arguments as to likelihood of\\nsuccess on the merits and do not reach the remaining Hilton factors. See Nken v.\\nHolder, 556 U.S. 418, 435 (2009) (stating that the likelihood of success on the\\nmerits factor is one of the “most critical” and must be established before\\nconsidering the last two stay factors); cf. California v. Azar, 911 F.3d 558, 575 (9th\\nCir. 2018) (“Likelihood of success on the merits is the most important factor; if a\\nmovant fails to meet this threshold inquiry, we need not consider the other factors.”\\n(internal quotation marks omitted) (quoting Disney Enters., Inc. v. VidAngel, Inc.,\\n869 F.3d 848, 856 (9th Cir. 2017))).\\n\\n\\n 3 19-16487\\n\\nsupported by the record as it stands.4 Cf. City and County of San Francisco v.\\n\\nTrump, 897 F.3d 1225, 1243–45 (9th Cir. 2018).5\\n\\n An injunction must be “narrowly tailored to remedy the specific harm\\n\\nshown.” Id. at 1244 (quoting Bresgal v. Brock, 843 F.2d 1163, 1170 (9th Cir.\\n\\n1987)). We have upheld nationwide injunctions where such breadth was necessary\\n\\nto remedy a plaintiff’s harm. See, e.g., id.; California v. Azar, 911 F.3d 558, 582\\n\\n(9th Cir. 2018) (“Although there is no bar against nationwide relief in federal\\n\\ndistrict court . . . such broad relief must be necessary to give prevailing parties the\\n\\n\\n4\\n The dissent, without citing any authority, argues that “it is [not] within a\\nmotions panel’s province to parse the record for error at this stage” and accuses us\\nof “[going] beyond the recognized authority of a motions panel” by granting the\\nmotion for a stay pending appeal insofar as the injunction applies outside the Ninth\\nCircuit. We have two responses.\\n First, we did not have to “parse” the record for error. Appellants’ stay\\nmotion specifically argues that the district court erred in imposing a nationwide\\ninjunction. Moreover, the three sentences that the district court provided to\\nsupport the imposition of a nationwide injunction—none of which explains why it\\nbelieved a nationwide injunction was necessary in this case—make clear that it\\nfailed to undertake the analysis necessary before granting such broad relief.\\n Second, other motions panels of our court have reviewed the scope of\\ninjunctive relief granted by district courts. See, e.g., E. Bay Sanctuary Covenant v.\\nTrump, No. 18-17274, 2018 WL 8807133, at *24 (9th Cir. Dec. 7, 2018); Hawaii\\nv. Trump, No. 17-17168, 2017 WL 5343014, at *1 (9th Cir. Nov. 13, 2017). We\\nthink these decisions illustrate that it is indeed within our province—our duty,\\neven—to review whether the district court abused its discretion in granting a\\nnationwide injunction.\\n5\\n The dissent criticizes our reliance on Trump, 897 F.3d 1225, because the\\nprocedural posture in this case is different. We recognize this difference as we cite\\nTrump as an analogous case supporting our decision because, notwithstanding the\\ndifferent procedural posture, the issue in that case—whether the scope of the\\ninjunction was appropriate—is the same issue before us. See id. at 1244–45.\\n\\n 4 19-16487\\n\\nrelief to which they are entitled.” (internal quotation marks and alterations omitted)\\n\\n(quoting Bresgal, 843 F.2d at 1170–71)). These are, however, “exceptional cases.”\\n\\nTrump, 897 F.3d at 1244. To permit such broad injunctions as a general rule,\\n\\nwithout an articulated connection to a plaintiff’s particular harm, would\\n\\nunnecessarily “stymie novel legal challenges and robust debate” arising in different\\n\\njudicial districts. Id.; see also Azar, 911 F.3d at 583 (“The Supreme Court has\\n\\nrepeatedly emphasized that nationwide injunctions have detrimental consequences\\n\\nto the development of law and deprive appellate courts of a wider range of\\n\\nperspectives.”).\\n\\n Here, the district court failed to discuss whether a nationwide injunction is\\n\\nnecessary to remedy Plaintiffs’ alleged harm. Instead, in conclusory fashion, the\\n\\ndistrict court stated that nationwide relief is warranted simply because district\\n\\ncourts have the authority to impose such relief in some cases and because such\\n\\nrelief has been applied in the immigration context. The district court clearly erred\\n\\nby failing to consider whether nationwide relief is necessary to remedy Plaintiffs’\\n\\nalleged harms. And, based on the limited record before us, we do not believe a\\n\\nnationwide injunction is justified.\\n\\n Our dissenting colleague believes that a nationwide injunction is appropriate\\n\\nsimply because this case presents a rule that applies nationwide. That view,\\n\\nhowever, ignores our well-established rule that injunctive relief “must be tailored\\n\\n\\n 5 19-16487\\n\\nto remedy the specific harm alleged.” Lamb-Weston, Inc. v. McCain Foods, Ltd.,\\n\\n941 F.2d 970, 974 (9th Cir. 1991) (citations omitted). Indeed, were we to adopt\\n\\nthe dissent’s view, a nationwide injunction would result any time an enjoined\\n\\naction has potential nationwide effects. Such an approach would turn broad\\n\\ninjunctions into the rule rather than the exception. Under our case law, however,\\n\\nall injunctions—even ones involving national policies—must be “narrowly tailored\\n\\nto remedy the specific harm shown.” Trump, 897 F.3d at 1244 (quoting Bresgal,\\n\\n843 F.2d at 1170).\\n\\n We agree with our dissenting colleague that “time does not permit a full\\n\\nexploration of the merits of the ‘nationwide’ issue.” But whereas he believes that\\n\\nsuch a factor supports the granting of a nationwide injunction until a merits panel\\n\\ncan address the case, we reach precisely the opposite conclusion. “National\\n\\ninjunctions interfere with good decisionmaking by the federal judiciary.” Samuel\\n\\nL. Bray, Multiple Chancellors: Reforming the National Injunction, 131 Harv. L.\\n\\nRev. 417, 461 (2017). They “deprive” other parties of “the right to litigate in other\\n\\nforums.” Azar, 911 F.3d at 583. Based on the briefing and limited record before\\n\\nus, and absent an explanation by the district court as to why a nationwide\\n\\ninjunction is necessary to remedy Plaintiffs’ alleged harm in this case, we must\\n\\ngrant the motion for stay pending appeal insofar as the injunction applies outside\\n\\nthe Ninth Circuit.\\n\\n\\n 6 19-16487\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4077,"cells":{"id":{"kind":"number","value":4649096,"string":"4,649,096"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Aaron Senne v. Kansas City Royals Baseball"},"decision_date":{"kind":"string","value":"2019-08-16"},"docket_number":{"kind":"string","value":"17-16245"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Michael R. Murphy, * Richard A. Paez, and Sandra S. Ikuta, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PAEZ, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nAARON SENNE; MICHAEL LIBERTO; Nos. 17-16245\\nOLIVER ODLE; BRAD MCATEE; CRAIG 17-16267\\nBENNIGSON; MATT LAWSON; KYLE 17-16276\\nWOODRUFF; RYAN KIEL; KYLE\\nNICHOLSON; BRAD STONE; MATT D.C. No.\\nDALY; AARON MEADE; JUSTIN 3:14-cv-00608-\\nMURRAY; JAKE KAHAULELIO; RYAN JCS\\nKHOURY; DUSTIN PEASE; JEFF\\nNADEAU; JON GASTON; BRANDON\\nHENDERSON; TIM PAHUTA; LEE OPINION\\nSMITH; JOSEPH NEWBY; RYAN\\nHUTSON; MATT FREVERT; ROBERTO\\nORTIZ; WITER JIMENEZ; KRIS WATTS;\\nMITCH HILLIGOSS; DANIEL BRITT;\\nYADEL MARTI; HELDER VELAQUEZ;\\nJORGE JIMENEZ; JORGE MINYETY;\\nEDWIN MAYSONET; JOSE DIAZ; NICK\\nGIARRAPUTO; LAUREN GAGNIER;\\nLEONARD DAVIS; GASPAR SANTIAGO;\\nGRANT DUFF; OMAR AGUILAR; MARK\\nWAGNER; DAVID QUINOWSKI;\\nBRANDON PINCKNEY, Individually\\nand on Behalf of All Those Similarly\\nSituated; JAKE OPITZ; BRETT\\nNEWSOME,\\n Plaintiffs-Appellants,\\n\\n v.\\n\\n2 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n\\nKANSAS CITY ROYALS BASEBALL\\nCORP.; MARLINS TEAMCO LLC; SAN\\nFRANCISCO BASEBALL ASSOCIATES,\\nLLC; OFFICE OF THE COMMISSIONER\\nOF BASEBALL, DBA Major League\\nBaseball, an unincorporated\\nassociation; ALLAN HUBER SELIG,\\n“BUD”; ANGELS BASEBALL LP; ST.\\nLOUIS CARDINALS, LLC; COLORADO\\nROCKIES BASEBALL CLUB, LTD.;\\nCINCINNATI REDS, LLC; HOUSTON\\nBASEBALL PARTNERS LLC;\\nATHLETICS INVESTMENT GROUP,\\nLLC; ROGERS BLUE JAYS BASEBALL\\nPARTNERSHIP; PADRES L.P.; SAN\\nDIEGO PADRES BASEBALL CLUB, L.P.;\\nMINNESOTA TWINS, LLC; DETROIT\\nTIGERS, INC.; LOS ANGELES\\nDODGERS LLC; STERLING METS L.P.;\\nAZPB L.P.; NEW YORK YANKEES\\nP’SHIP; RANGERS BASEBALL EXPRESS,\\nLLC; MILWAUKEE BREWERS\\nBASEBALL CLUB, INC.; CHICAGO\\nCUBS BASEBALL CLUB, LLC;\\nPITTSBURGH ASSOCIATES, LP;\\nBASEBALL CLUB OF SEATTLE, LLP;\\nLOS ANGELES DODGERS HOLDING\\nCOMPANY LLC; RANGERS BASEBALL,\\nLLC,\\n Defendants-Appellees.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 3\\n\\n Appeals from the United States District Court\\n for the Northern District of California\\n Joseph C. Spero, Magistrate Judge, Presiding\\n\\n Argued and Submitted June 13, 2018\\n San Francisco, California\\n\\n Filed August 16, 2019\\n\\n Before: Michael R. Murphy, * Richard A. Paez,\\n and Sandra S. Ikuta, Circuit Judges.\\n\\n Opinion by Judge Paez;\\n Dissent by Judge Ikuta\\n\\n\\n\\n\\n *\\n The Honorable Michael R. Murphy, United States Circuit Judge\\nfor the U.S. Court of Appeals for the Tenth Circuit, sitting by\\ndesignation.\\n\\n4 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n SUMMARY **\\n\\n\\n Labor Law / Class and Collective Certification\\n\\n The panel affirmed in part and reversed in part the\\ndistrict court’s orders certifying a class and a collective\\naction for wage-and-hour claims brought by minor league\\nbaseball players under the Fair Labor Standards Act and\\nstate law.\\n\\n The district court certified a California class under\\nFederal Rule of Civil Procedure 23(b)(3) but denied\\ncertification for Arizona and Florida classes and for a Rule\\n23(b)(2) class. The district court also certified an FLSA\\ncollective.\\n\\n The panel held that, as to the state law claims, California\\nchoice-of-law rules applied. The panel held that under\\nSullivan v. Oracle Corp., 254 P.3d 237 (Cal. 2011),\\nCalifornia law applied to the Rule 23(b)(3) California class.\\nThe panel reversed the district court’s determination that\\nchoice-of-law considerations defeated the predominance and\\nadequacy requirements for the proposed Arizona and Florida\\nRule 23(b)(3) classes. Applying California’s three-step\\ngovernmental interest analysis for choice-of-law questions,\\nthe panel concluded that Arizona law should apply to the\\nwork performed in Arizona, and Florida law to the work\\nperformed in Florida.\\n\\n The panel reversed the district court’s refusal to certify a\\nRule 23(b)(2) class for unpaid work at defendants’ training\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 5\\n\\nfacilities in Arizona and Florida on the basis that choice-of-\\nlaw issues undermined “cohesiveness” and therefore made\\ninjunctive and declaratory relief inappropriate. The panel\\nconcluded that the district court’s errors in its choice-of-law\\nanalysis relating to the proposed Arizona and Florida Rule\\n23(b)(3) classes applied equally to its refusal to certify the\\nRule 23(b)(2) class. The panel further held that the district\\ncourt erred in imposing a “cohesiveness” requirement for the\\nproposed Rule 23(b)(2) class. The panel remanded for the\\ndistrict court to consider anew whether to certify the Rule\\n23(b)(2) class.\\n\\n The panel held that plaintiffs could meet the\\npredominance requirement for the proposed California,\\nFlorida, and Arizona Rule 23(b)(3) classes through a\\ncombination of representative evidence and application of\\nthe “continuous workday” rule. The panel applied the Mt.\\nClemens burden-shifting framework and the holding of\\nTyson Foods v. Bouaphakeo, 136 S. Ct. 1036 (2016), that\\nrepresentative evidence may be used at the class certification\\nstage and may be used to establish liability in addition to\\ndamages. The panel explained that the continuous workday\\nrule presumes that once the beginning of the workday is\\ntriggered, an employee performs compensable work\\nthroughout the rest of the day until the employee completes\\ntheir last principal activity. Any activity that is “integral and\\nindispensable” to principal activities triggers the beginning\\nof the workday. As to the Arizona and Florida classes,\\ncovering alleged minimum wage violations in the lack of any\\npay for time spent participating in spring training, extended\\nspring training, and instructional leagues, the panel affirmed\\nthe determination that the predominance requirement was\\nmet. As to the California class, covering overtime and\\nminimum wage claims relating to work performed during the\\nchampionship season, the panel held that the district court\\n\\n6 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\ndid not abuse its discretion in concluding that defendants’\\nuniform pay policy, the team schedules, and representative\\nevidence, including an expert survey known as the “Main\\nSurvey,” established predominance. The panel held that the\\ndistrict court was not required to “rigorously analyze” the\\nMain Survey, rather than evaluating its admissibility under\\nDaubert and its appropriateness for meeting class\\ncertification requirements under Tyson.\\n\\n Affirming the district court’s certification of the FLSA\\ncollective action, the panel applied the standard set forth in\\nCampbell v. City of L.A., 903 F.3d 1090 (9th Cir. 2018),\\nwhich postdated the district court’s ruling, and held that the\\ndistrict court’s use of the ad hoc approach was harmless\\nerror. The panel concluded that collective certification was\\nproper because plaintiffs shared similar issues of law or fact\\nmaterial to the disposition of their FLSA claims and thus\\nwere similarly situated.\\n\\n The panel affirmed the district court’s certification of the\\nCalifornia Rule 23(b)(3) class and the FLSA collective\\naction, reversed the district court’s refusal to certify Arizona\\nand Florida classes and a Rule 23(b)(2) class, and remanded\\nfor further proceedings.\\n\\n Dissenting, Judge Ikuta wrote that the district court\\ncorrectly concluded that consideration of plaintiffs’ claims\\non a classwide basis would be overwhelmed by\\nindividualized choice-of-law inquiries. She wrote that the\\nmajority’s rule, applying the law of the jurisdiction where\\nthe work took place, was contrary to the court’s framework\\nfor analyzing the intersection of class action and choice-of-\\nlaw issues, overlooked the complexity of California’s\\nchoice-of-law rules, and created significant practical and\\nlogistical problems.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 7\\n\\n COUNSEL\\n\\nRobert L. King (argued) and Garrett R. Broshuis, Korein\\nTillery LLC, St. Louis, Missouri; Bruce L. Simon and\\nBenjamin E. Shiftan, Pearson Simon & Warshaw LLP, San\\nFrancisco, California; Daniel L. Warshaw and Bobby Pouya,\\nPearson Simon & Warshaw LLP, Sherman Oaks, California;\\nfor Plaintiffs-Appellants.\\n\\nElise M. Bloom (argued), Adam M. Lupion, and Mark D.\\nHarris, Proskauer Rose LLP, New York, New York; John E.\\nRoberts, Proskauer Rose LLP, Boston, Massachusetts; for\\nDefendants-Apellees.\\n\\nDavid C. Frederick and Jeffrey A. Love, Kellogg Hansen\\nTodd Figel & Frederick PLLC, Washington, D.C., for Amici\\nCuriae Professors Peter Hay and Patrick J. Borchers.\\n\\nAllan Steyer and Donald Scott MacRae, Steyer Lowenthal\\nBoodrookas Alvarez & Smith LLP, San Francisco,\\nCalifornia, for Amici Curiae Professional Hockey Players\\nAssocation; Association of Minor League Umpires; Office\\nand Professional Employee’s International Union; and\\nUnited Steel, Paper and Forestry, Rubber, Manufacturing,\\nEnergy, Allied and Industrial Service Workers International\\nUnion.\\n\\n8 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n It is often said that baseball is America’s pastime. In this\\ncase, current and former minor league baseball players\\nallege that the American tradition of baseball collides with a\\ntradition far less benign: the exploitation of workers. We are\\ntasked with deciding whether these minor league players\\nmay properly bring their wage-and-hour claims on a\\ncollective and classwide basis.\\n\\n BACKGROUND\\n\\n I.\\n\\n Most major professional sports in America have their\\nown “farm system” for developing talent: for the National\\nBasketball Association, it’s the G-League; for the National\\nHockey League, it’s the American Hockey League; and for\\nMajor League Baseball (MLB), it’s Minor League Baseball.\\nMLB and its thirty franchise teams rely heavily on this\\nextensive minor league system, which has nearly 200\\naffiliates across the country and employs approximately\\n6,000 minor league players. Nearly all MLB players begin\\ntheir careers in the minor leagues. Each minor league club\\nis associated with one of the thirty franchise MLB teams.\\n\\n The minor league system is governed by the Major\\nLeague Rules (MLRs), which dictate the terms of\\nemployment and compensation for both minor and major\\nleague players. Under the MLRs, all minor league players\\nare required to sign a seven-year Uniform Player Contract\\n(UPC). Ostensibly, players are required to sign the UPC for\\n“morale” and “to produce the similarity of conditions\\nnecessary for keen competition.”\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 9\\n\\n The UPC “obligates Player[s] to perform professional\\nservices on a calendar year basis, regardless of the fact that\\nsalary payments are to be made only during the actual\\nchampionship playing season.” It describes its scope as\\nsetting “the terms and conditions of employment during all\\nperiods in which Player is employed by Club as a Minor\\nLeague Player.” Players are paid by the MLB franchise\\naffiliated with the minor league team for which they play.\\nUnder the UPC, first-year players are paid a fixed salary of\\n$1,100 per month during the regular (“championship”)\\nseason that runs from April through September. In addition\\nto their salaries during the championship season, some\\nplayers receive signing or performance-related bonuses and\\ncollege scholarships.\\n\\n Beginning in early March each year, the minor league\\naffiliates conduct spring training in Arizona and Florida;\\nevery MLB franchise operates a minor league training\\ncomplex in one of these two states. The parties dispute\\nwhether spring training is required, but the UPC strongly\\nindicates that it is mandatory. 1 Virtually all players are\\nunpaid during spring training.\\n\\n Spring training lasts approximately four weeks, until the\\nchampionship season begins in April. Some players attest\\nthat spring training entails working seven days a week, with\\nno days off. During spring training, teams typically have\\n\\n 1\\n The UPC provides that “Player’s duties and obligations under [the\\nUPC] continue in full force and effect throughout the calendar year,\\nincluding Club’s championship playing season, Club’s training season,\\nClub’s exhibition games, Club’s instructional, post-season training or\\nwinter league games, any official play-off series, any other official post-\\nseason series in which Player shall be required to participate . . . and any\\nremaining portions of the calendar year.”\\n\\n10 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nscheduled activities in the morning prior to playing games in\\nthe afternoon. For example, a team spring training schedule\\nfor one of the San Francisco Giants’ affiliates describes that\\nat 6:30 AM, there was an “Early Van for Treatment and\\nEarly Work” 2; at 7:00 AM, the “Regular Van” departed; at\\n7:45 AM, the “Early Work” began; and then between 9:00\\nAM and 11:00 AM, the team would perform activities such\\nas “Stretch,” “Throwing Program,” and “Batting Practice.”\\nLunch was to be at 11:00 AM, before a 12:10 PM bus to a\\nneighboring city for a 1:00 PM away game.\\n\\n At the conclusion of spring training in early April, some\\nplayers are assigned to minor league affiliates, and begin\\nplaying games in the championship season. During the\\nchampionship season, minor league teams play games either\\nsix or seven days per week. The championship season lasts\\naround five months, beginning in April and ending in\\nSeptember. One of the regular season leagues within minor\\nleague baseball is the California League, which—as the\\nname implies—plays games exclusively within California.\\n\\n Players who are not assigned to play for affiliates in the\\nchampionship season stay at the Arizona or Florida facilities\\nfor “extended spring training.” Extended spring training\\ncontinues until June, and involves similar activities to spring\\ntraining. Although most players do not get paid during\\nextended spring training, as many as seven MLB clubs do\\npay for work during extended spring training due to an\\nambiguity in the MLRs over when players are permitted to\\nbe paid.\\n\\n\\n\\n 2\\n The schedule instructed players to “CHECK [the] BOARD FOR\\nEARLY WORK.”\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 11\\n\\n After the championship season ends in September, some\\nplayers participate in the “instructional leagues,” which run\\nfrom approximately mid-September to mid-October. The\\nparties dispute whether participation in the instructional\\nleagues is mandatory for the players involved, although as\\nwith spring training, the UPC strongly implies that\\nparticipation is required. Activities and schedules during the\\ninstructional league are similar to spring training. And just\\nas with spring training, players are virtually never paid for\\nparticipation in the instructional league.\\n\\n II.\\n\\n Plaintiffs are forty-five current and former minor league\\nbaseball players who bring claims under the federal Fair\\nLabor Standards Act (FLSA) and the wage-and-hour laws of\\nCalifornia, Arizona, and Florida against MLB, MLB\\nCommissioner Bud Selig, and a number 3 of MLB franchises.\\nPlaintiffs allege that defendants do not pay the players at all\\nduring spring training, extended spring training, or the\\ninstructional leagues. They further allege that because\\nplayers are “employees” and the activities the players\\nperform during those periods constitute compensable work,\\ndefendants have unlawfully failed to pay them at least\\nminimum wage. And according to plaintiffs, while the\\nplayers are paid—albeit not much—during the\\nchampionship season, they routinely work overtime, for\\nwhich they are never compensated as a matter of policy.\\n\\n In May 2015, plaintiffs filed their Second Amended\\nConsolidated Class Action Complaint, which alleged wage-\\n\\n 3\\n Plaintiffs originally named all 30 MLB franchises as defendants,\\nbut eight of the franchises were subsequently dismissed for lack of\\npersonal jurisdiction.\\n\\n12 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nand-hour claims under the laws of eight states and the FLSA;\\nplaintiffs also sought certification of a FLSA collective\\naction. The district court preliminarily certified the FLSA\\ncollective in October 2015. Notice was sent to\\napproximately 15,000 current and former minor league\\nplayers, of which more than 2,200 opted in.\\n\\n In 2016, defendants moved to decertify the FLSA\\ncollective, while plaintiffs moved to certify a Rule 23(b)(2)\\nclass as well as Rule 23(b)(3) classes under the laws of eight\\nstates. The district court denied certification for all proposed\\nRule 23(b)(3) classes, concluding that predominance was not\\nsatisfied for two primary reasons. Senne v. Kansas City\\nRoyals Baseball Corp., 315 F.R.D. 523, 572, 577–84 (N.D.\\nCal. 2016). First, the court concluded that predominance\\nwas defeated by the choice-of-law issues presented by the\\nproposed classes, given that (1) the winter off-season\\ntraining claims entailed work performed in dozens of\\ndifferent states with no common schedule or situs; and\\n(2) the championship season claims involved frequent travel\\nbetween state lines for away games. Id. at 580–81. The\\ndistrict court also determined that the inclusion of claims for\\nwinter off-season work fatally undermined predominance, as\\nthe court would be required to undertake an overwhelming\\nnumber of individualized inquiries to determine which\\nactivities constituted compensable “work” and how much\\ntime was spent doing “work.” Id. at 577–84. For similar\\nreasons, the court held that plaintiffs were not “similarly\\nsituated” and therefore decertified the FLSA collective. Id.\\nat 585–86. The court also granted the defendants’ motion to\\nexclude an expert survey (the “Pilot Survey”) submitted by\\nplaintiffs, finding that its methodology and results did not\\nsatisfy the requirements of Federal Rule of Evidence 702 and\\nDaubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S.\\n579 (1993). Id. at 586–90. The court further refused to\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 13\\n\\ncertify the proposed Rule 23(b)(2) class, concluding that\\nbecause the plaintiffs were all former—rather than current—\\nplayers, they lacked standing to represent a (b)(2) class. Id.\\nat 584–85.\\n\\n Plaintiffs moved for reconsideration, narrowing their\\nproposed classes significantly in response to the concerns the\\ndistrict court expressed in its initial certification order.\\nPlaintiffs requested Rule 23(b)(3) certification of an Arizona\\nclass and a Florida class for work performed during spring\\ntraining, extended spring training, and the instructional\\nleagues in those states. Plaintiffs also moved for\\ncertification of a 23(b)(3) California class, covering players\\nwho participated in the California League during the\\nchampionship season. Additionally, plaintiffs sought to\\ncertify a reworked FLSA collective of players who\\nparticipated in the California League or in spring training,\\nextended spring training, and the instructional leagues. In\\naddition to the 23(b)(3) classes and FLSA collective,\\nplaintiffs requested certification of a Rule 23(b)(2)\\ninjunctive relief class consisting of current minor league\\nplayers who participate in spring training, extended spring\\ntraining, or the instructional leagues in Florida or Arizona.\\nTo cure the court’s earlier concerns about standing, four\\ncurrent minor league players moved to intervene to represent\\nthe proposed (b)(2) class.\\n\\n On reconsideration, plaintiffs argued that they could\\nmeet Rule 23(b)(3)’s predominance requirement and\\nFLSA’s “similarly situated” requirement through a\\ncombination of the use of representative evidence and\\napplication of the so-called “continuous workday” rule. 4\\n\\n 4\\n As we shall explain, the continuous workday rule presumes that\\nonce the beginning of the workday is triggered, an employee performs\\n\\n14 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nPlaintiffs’ representative evidence took a variety of forms,\\nincluding an expert survey (the “Main Survey”), hundreds of\\nteam schedules, payroll data, and testimony from both\\nplayers and league officials. The most controversial piece of\\nevidence was the Main Survey, which plaintiffs argued\\nserved as representative evidence of hours worked,\\nparticularly when used in concert with a continuous workday\\ntheory.\\n\\n The Main Survey asked players to report the times they\\n“most often” arrived and departed from the ballpark or\\ntraining facility during the championship season, spring\\ntraining, extended spring training, and the instructional\\nleagues, and asked players to estimate how much time they\\nspent eating meals while at the ballpark. The survey did not,\\nhowever, ask players about the kinds of activities they\\nperformed at the facilities, or how much time they spent\\nperforming particular activities. Given these purported\\nshortcomings, defendants moved to exclude the Main\\nSurvey, and further argued that even if the survey were\\nadmissible under Daubert, it still could not be used to meet\\nthe predominance and “similarly situated” requirements due\\nto its alleged flaws. The district court denied defendants’\\nmotion to exclude the Main Survey, finding it admissible\\nunder Daubert and concluding that defendants’ challenges\\nwent “to the weight of the Survey and not its admissibility”\\nand were “better left to a jury to evaluate.” The district court\\nfurther concluded that the Main Survey could be used in\\ncombination with other evidence—such as team schedules,\\ntestimony, and payroll data—to meet Rule 23(b)(3)’s\\n\\ncompensable work throughout the rest of the day until the employee\\ncompletes their last principal activity or the last activity which is\\n“integral and indispensable” to the employee’s principal activities. IBP,\\nInc. v. Alvarez, 546 U.S. 21, 28, 32–37 (2005).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 15\\n\\npredominance and FLSA’s “similarly situated”\\nrequirements, observing that certifying the classes and the\\nFLSA collective “will not preclude Defendants from\\nchallenging the sufficiency of the Main Survey and\\nPlaintiffs’ damages model on summary judgment and/or at\\ntrial.”\\n\\n Because it concluded that the predominance and\\n“similarly situated” requirements could be met with the use\\nof representative evidence and application of the continuous\\nworkday rule, the district court recertified the narrowed\\nFLSA collective and certified a California (b)(3) class.\\nHowever, the district court denied certification for the\\nArizona, Florida, and (b)(2) classes, holding that choice-of-\\nlaw concerns defeated predominance for the Arizona and\\nFlorida classes and undermined “cohesiveness” for the\\n(b)(2) class.\\n\\n At defendants’ request, the district court certified the\\nFLSA collective certification order for interlocutory review\\nunder 28 U.S.C. § 1292. Plaintiffs petitioned us for\\npermission to appeal the denial of certification for the\\nArizona, Florida, and Rule 23(b)(2) classes, and defendants\\nlikewise petitioned to appeal the certification of the\\nCalifornia class; we granted both petitions, consolidating\\nthose cross-appeals with the FLSA collective appeal.\\n\\n STANDARD OF REVIEW\\n\\n We review for abuse of discretion the district court’s\\nclass certification rulings, and review for clear error any\\nfindings of fact the district court relied upon in its\\ncertification order. Parsons v. Ryan, 754 F.3d 657, 673 (9th\\nCir. 2014). A district court’s choice of law determinations,\\nhowever, are reviewed de novo. Mazza v. Am. Honda Motor\\nCo., 666 F.3d 581, 589 (9th Cir. 2012). A district court\\n\\n16 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nabuses its discretion where it commits an error of law, relies\\non an improper factor, omits a substantial factor, or engages\\nin a clear error of judgment in weighing the correct mix of\\nfactors. Stockwell v. City & County of San Francisco, 749\\nF.3d 1107, 1113 (9th Cir. 2014) (citing Bateman v. Am.\\nMulti-Cinema, Inc., 623 F.3d 708, 712 (9th Cir. 2010)).\\nWhen we review a grant of class certification, “we accord\\nthe district court noticeably more deference than when we\\nreview a denial.” Abdullah v. U.S. Sec. Assocs., Inc., 731\\nF.3d 952, 956 (9th Cir. 2013) (quoting Wolin v. Jaguar Land\\nRover N. Am., LLC, 617 F.3d 1168, 1171 (9th Cir. 2010)).\\n\\n ANALYSIS\\n\\n To paraphrase the Chief Justice, these complex appeals\\nrequire us to call a great number of balls and strikes, as both\\nparties raise numerous challenges to the district court’s\\ncertification order. For their part, plaintiffs challenge the\\ndistrict court’s decision to deny certification for the Arizona\\nand Florida Rule 23(b)(3) classes and the Rule 23(b)(2) class\\non the grounds that choice-of-law issues defeated the\\npredominance requirement for the Arizona and Florida\\n(b)(3) classes and also thwarted “cohesiveness” for the\\nproposed (b)(2) class. Defendants, on the other hand, contest\\nthe district court’s certification of the California (b)(3) class,\\narguing first that choice-of-law issues defeat both\\npredominance and adequacy, and second, that plaintiffs\\ncannot meet the predominance requirement through the use\\nof their proffered representative evidence: the Main Survey,\\nteam schedules, payroll records, deposition testimony, and\\ndeclarations. Defendants further charge that the district\\ncourt erred in certifying the FLSA collective because\\nplaintiffs’ representative evidence does not show that the\\ncollective members are “similarly situated.” Defendants\\nalso contend that the district court erred by not “rigorously\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 17\\n\\nanalyzing” plaintiffs’ expert evidence at the class and\\ncollective certification stage. We address each argument in\\nturn.\\n\\n I.\\n\\n Class certification is governed by Federal Rule of Civil\\nProcedure 23. As a threshold matter, a party seeking class\\ncertification must satisfy the four requirements of Rule\\n23(a): (1) numerosity; (2) commonality; (3) typicality; and\\n(4) adequacy of representation. 5 “Class certification is\\nproper only if the trial court has concluded, after a ‘rigorous\\nanalysis,’ that Rule 23(a) has been satisfied.” Parsons, 754\\nF.3d at 674 (quoting Wang v. Chinese Daily News, Inc., 737\\nF.3d 538, 542–43 (9th Cir. 2013)).\\n\\n In addition to the requirements of Rule 23(a), a proposed\\nclass must also meet the requirements of one or more of the\\n“three different types of classes” set forth in Rule 23(b).\\nLeyva v. Medline Industries, Inc., 716 F.3d 510, 512 (9th Cir.\\n2013). Here, plaintiffs proposed classes under two of Rule\\n23(b)’s class types: Rule 23(b)(3) and 23(b)(2). A class may\\nbe certified under Rule 23(b)(3) only if the district court\\n“finds that the questions of law or fact common to class\\nmembers predominate over any questions affecting only\\nindividual members, and that a class action is superior to\\nother available methods for fairly and efficiently\\nadjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). Of\\nthese two requirements—predominance and superiority—\\nonly predominance is at issue on appeal. “The\\npredominance inquiry focuses on ‘the relationship between\\n\\n 5\\n Of Rule 23(a)’s four requirements, defendants contest only\\nadequacy on appeal; their arguments pertaining to adequacy have to do\\nwith choice-of-law issues.\\n\\n18 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nthe common and individual issues’ and ‘tests whether\\nproposed classes are sufficiently cohesive to warrant\\nadjudication by representation.’” Vinole v. Countrywide\\nHome Loans, Inc., 571 F.3d 935, 944 (9th Cir. 2009)\\n(quoting Hanlon v. Chrysler Corp., 150 F.3d 1011, 1022 (9th\\nCir. 1998)). In determining whether the predominance\\nrequirement is met, courts have a “duty to take a close look\\nat whether common questions predominate over individual\\nones” to ensure that individual questions do not “overwhelm\\nquestions common to the class.” Comcast Corp. v. Behrend,\\n569 U.S. 27, 34 (2013) (internal quotation marks and citation\\nomitted).\\n\\n Rule 23(b)(2), on the other hand, requires only that “the\\nparty opposing the class ha[ve] acted or refused to act on\\ngrounds that apply generally to the class, so that final\\ninjunctive relief or corresponding declaratory relief is\\nappropriate respecting the class as a whole.” Fed. R. Civ. P.\\n23(b)(2). Although 23(b)(2) classes are most common in the\\ncivil rights context, “we have certified many different kinds\\nof Rule 23(b)(2) classes.” Parsons, 754 F.3d at 686.\\n\\n II.\\n\\n We first address whether choice-of-law issues fatally\\nundermine plaintiffs’ proposed Rule 23 classes. The district\\ncourt’s decision was split on the impact of choice-of-law\\nquestions: as to the proposed Rule 23(b)(3) California class,\\nthe court held that choice-of-law concerns defeated neither\\nRule 23(b)(3)’s predominance requirement nor Rule 23(a)’s\\nadequacy requirement. Yet as to the proposed 23(b)(3)\\nArizona and Florida classes, the district court held the\\nopposite: that choice-of-law issues posed an insurmountable\\nhurdle to meeting both predominance and adequacy.\\nSimilarly, the court determined that choice-of-law questions\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 19\\n\\nmade certification of the proposed Rule 23(b)(2) class\\ninappropriate.\\n\\n Concerns over which state’s laws apply to a proposed\\nclass “do not necessarily preclude a 23(b)(3) action.”\\nHanlon, 150 F.3d at 1022. But “[u]nderstanding which law\\nwill apply before making a predominance determination is\\nimportant when there are variations in applicable state law,”\\nand potentially varying state laws may defeat predominance\\nin certain circumstances. Zinser v. Accufix Research Inst.,\\nInc., 253 F.3d 1180, 1189 (9th Cir. 2001) opinion amended\\non denial of reh’g, 273 F.3d 1266 (9th Cir. 2001). We have\\nbeen particularly concerned about the impact of choice-of-\\nlaw inquiries in nationwide consumer class actions and\\nproducts liability cases. See, e.g., Mazza, 666 F.3d at 585,\\n591–94; Zinser, 253 F.3d at 1184–90.\\n\\n A district court considering state law claims brought in\\nfederal court must utilize the choice-of-law rules of the\\nforum state—here, California. Klaxon Co. v. Stentor\\nElectric Mfg. Co., 313 U.S. 487, 496–97 (1941). “By\\ndefault, California courts apply California law unless a party\\nlitigant timely invokes the law of a foreign state, in which\\ncase it is the foreign law proponent who must shoulder the\\nburden of demonstrating that foreign law, rather than\\nCalifornia law, should apply to class claims.” In re Hyundai\\n& Kia Fuel Econ. Litig., 926 F.3d 539, 561 (9th Cir. 2019)\\n(en banc) (internal quotation marks and citations omitted).\\nTo meet their burden, the objectors must satisfy California’s\\nthree-step governmental interest test, used to resolve choice\\nof law issues. Id.\\n\\n First, the court determines whether the\\n relevant law of each of the potentially\\n affected jurisdictions with regard to the\\n particular issue in question is the same or\\n\\n20 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n different. Second, if there is a difference, the\\n court examines each jurisdiction’s interest in\\n the application of its own law under the\\n circumstances of the particular case to\\n determine whether a true conflict exists.\\n Third, if the court finds that there is a true\\n conflict, it carefully evaluates and compares\\n the nature and strength of the interest of each\\n jurisdiction in the application of its own law\\n to determine which state’s interest would be\\n more impaired if its policy were subordinated\\n to the policy of the other state, and then\\n ultimately applies the law of the state whose\\n interest would be the more impaired if its law\\n were not applied.\\n\\nKearney v. Salomon Smith Barney, Inc., 137 P.3d 914, 922\\n(Cal. 2006) (internal quotation marks and citations omitted);\\nsee also Hairu Chen v. Los Angeles Truck Centers, LLC, No.\\nS240245, 2019 WL 3281346, at *3 (Cal. July 22, 2019).\\n\\n In making its choice-of-law determinations, the district\\ncourt relied heavily on the California Supreme Court’s\\ndecision in Sullivan v. Oracle Corp., 254 P.3d 237 (Cal.\\n2011), and the parties do not dispute that Sullivan provides\\nthe most helpful guidance for the choice-of-law questions\\nbefore us. In Sullivan, the California Supreme Court\\nanswered a certified question from this court regarding\\nwhether California overtime law applied to non-resident\\nemployees of a California corporation who worked primarily\\nin their home states of Colorado and Arizona, but also\\nworked in California (and several other states) for “entire\\ndays or weeks” at a time. Id. at 239, 243. Sullivan first\\nconcluded that as a matter of statutory construction,\\nCalifornia law applied to all work performed for days or\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 21\\n\\nweeks at a time within the state’s borders, regardless of\\nwhether it was performed by residents or non-residents. Id.\\nat 241–43. Next, Sullivan undertook California’s three-step\\ngovernmental interest analysis for choice-of-law questions.\\nId. at 244–47. At the first step of the analysis—whether the\\nrelevant laws differed—the court noted that California’s\\novertime law “clearly” differed from the laws of the\\nplaintiffs’ home states. Id. at 245.\\n\\n At the second step—whether a “true” conflict existed—\\nthe court held that the existence of a true conflict was\\n“doubtful, at best.” Id. The court explained that the second\\nstep involves examining “each jurisdiction’s interest in the\\napplication of its own law under the circumstances of the\\nparticular case,” noting that a court “may make [its] own\\ndetermination of the relevant policies and interests, without\\ntaking ‘evidence’ as such on the matter.” Id. (internal\\nquotation marks, alterations, and citations omitted). Sullivan\\nobserved that “California has, and has unambiguously\\nasserted, a strong interest in applying its overtime law to all\\nnonexempt workers, and all work performed, within its\\nborders.” Id. at 245. The court concluded that “neither\\nColorado nor Arizona has a legitimate interest in shielding\\nOracle from the requirements of California wage law as to\\nwork performed here.” Id. at 246.\\n\\n In so holding, the court rejected two specific arguments\\nadvanced by Oracle. First, Oracle contended that because\\nArizona and Colorado have workers’ compensation statutes\\nwith express extraterritorial application, those statutes\\nindicate an interest in extending the protection of their\\nemployment laws to their residents working outside the\\nstate. Id. Not so, Sullivan held. While “a state has such an\\ninterest, at least in the abstract, when the traveling, resident\\nemployee of a domestic employer would otherwise be left\\n\\n22 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nwithout the protection of another state’s law,” the states had\\n“expressed no interest in disabling their residents from\\nreceiving the full protection of California overtime law when\\nworking here, or in requiring their residents to work side-by-\\nside with California residents in California for lower pay.”\\nId.\\n\\n Second, Oracle argued that Arizona and Colorado “have\\nan interest in providing hospitable regulatory environments\\nfor their own businesses” and thus “also have an interest in\\nshielding their own businesses from more costly and\\nburdensome regulatory environments in other states.” Id.\\nRelying on principles of federalism, Sullivan dismissed this\\nargument. While “a state can properly choose to create a\\nbusiness-friendly environment within its own boundaries,”\\nthe federal Constitution does not require a state to substitute\\n“‘the conflicting statute of another state’” for its own laws\\nthat are “‘applicable to persons and events’” within that\\nstate. Id. (quoting Phillips Petrol. Co. v. Shutts, 472 U.S.\\n797, 822 (1985)). Nor does the Constitution “permit one\\nstate to project its regulatory regime into the jurisdiction of\\nanother state.” Id. (citing Healy v. Beer Insti., Inc., 491 U.S.\\n324, 336–37 (1989)).\\n\\n Finally, although Sullivan held that there was almost\\ncertainly no true conflict because neither Arizona nor\\nColorado had a “legitimate interest” in blocking the\\napplication of California law to the work performed in\\nCalifornia, the court nonetheless proceeded to the third step\\nof the analysis “for the sake of argument.” Id. at 246–47.\\nSullivan concluded that the analysis at the third step—\\ndetermining which state’s interest would be more impaired\\nif its policy were subordinated to the policy of the other\\nstate—yielded a straightforward answer:\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 23\\n\\n [T]o subordinate California’s interests to\\n those of Colorado and Arizona\\n unquestionably would bring about the greater\\n impairment. To permit nonresidents to work\\n in California without the protection of our\\n overtime law would completely sacrifice, as\\n to those employees, the state’s important\\n public policy goals of protecting health and\\n safety and preventing the evils associated\\n with overwork. Not to apply California law\\n would also encourage employers to substitute\\n lower paid temporary employees from other\\n states for California employees, thus\\n threatening California’s legitimate interest in\\n expanding the job market. By way of\\n comparison, not to apply the overtime laws of\\n Colorado and Arizona would impact those\\n states’ interests negligibly, or not at all . . .\\n Alternatively, viewing Colorado’s and\\n Arizona's overtime regimens as expressions\\n of a general interest in providing hospitable\\n regulatory environments to businesses within\\n their own boundaries, that interest is not\\n perceptibly impaired by requiring a\\n California employer to comply with\\n California overtime law for work performed\\n here.\\n\\nId. at 247 (citations omitted) (emphasis added).\\n\\n A.\\n\\n We first conclude that the district court did not err in\\nholding that under Sullivan, California law should apply to\\n\\n24 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nthe (b)(3) California class. 6 Although defendants correctly\\npoint out that Sullivan is not precisely analogous to the case\\nat hand, the two principal differences on which defendants\\nrely are unpersuasive. Specifically, defendants first rely on\\nthe fact that while Sullivan involved a California\\ncorporation, “most of the MLB Club Defendants with\\naffiliates in the California League are located outside\\nCalifornia.” But a close reading of Sullivan indicates that\\nCalifornia law should apply to the California class, even\\nthough many of the employers are not headquartered in\\nCalifornia. For example, Sullivan expressly contemplated\\nthat California’s overtime laws may not apply to non-\\nresident employees of an out-of-state business who enter\\nCalifornia only “temporarily during the course of a\\nworkday,” but contrasted such a scenario with employees\\nwho work in California for “entire days and weeks,” who are\\ncovered by California law. Id. at 243 (emphases, internal\\nquotation marks, and citations omitted).\\n\\n Similarly, Sullivan specifically left open the possibility\\nthat other California employment laws, such as pay stub\\nrequirements, may not apply to non-resident employees of\\nout-of-state employers—with the clear implication that\\novertime laws would apply to such employees. See id. at\\n243–44. Likewise, far from limiting its holding only to non-\\nresident employees of in-state employers, Sullivan merely\\nemphasized that employees of in-state employers would\\nespecially be covered by California law. See id. at 243.\\n\\n 6\\n Contrary to the dissent’s criticism, Dissent at 74, we do not\\nshortcut the governmental interest analysis. As we explain in the text,\\nwe believe that Sullivan mandates application of California law to the\\nCalifornia class. Rather than repeating Sullivan’s choice of law analysis,\\nwe focus on several additional considerations that further support our\\ndecision to affirm the district court’s reliance on Sullivan.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 25\\n\\n Second, defendants characterize Sullivan as resting on\\nthe court’s determination that “neither Arizona nor Colorado\\n. . . has asserted an interest in regulating overtime work\\nperformed in other states.” Defendants argue that here, by\\ncontrast, “numerous” states have a competing interest in\\nregulating work performed in California. But defendants\\nmisread Sullivan by erroneously presuming that its\\nconclusion at the third step—that subordinating\\n“California’s interests to those of Colorado and Arizona\\nunquestionably would bring about the greater\\nimpairment”—hinged entirely on whether Arizona or\\nColorado law had asserted an interest in extraterritorial\\napplication of their wage laws. Id. at 247. It is certainly\\naccurate to say that Sullivan’s holding was influenced by the\\nfact that neither Arizona nor Colorado law purported to\\napply extraterritorially. Yet the court’s discussion at step\\nthree cannot fairly be read to support the argument that\\nCalifornia’s “strong interest in applying its overtime law to\\n. . . all work performed within its borders,” id. at 245, would\\nsuddenly become the lesser-impaired interest in the event\\nanother state expressed a clear interest in applying its wage\\nlaws to work performed in California. Rather, Sullivan\\nstrongly indicates that California’s interest in applying its\\nlaws to work performed within its borders for days or weeks\\nat a time would reign supreme regardless of whether another\\nstate expressed an interest in applying its own wage laws\\ninstead of California’s.\\n\\n26 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n Although we read Sullivan as clearly mandating the\\napplication of California law to the California class, two\\nadditional considerations support our conclusion today. 7\\n\\n First, because the district court found that plaintiffs had\\nmet their burden of showing that California law could\\nconstitutionally be applied—a determination defendants do\\nnot contest on appeal—the burden shifted to defendants “to\\ndemonstrate ‘that foreign law, rather than California law,\\nshould apply to class claims.’” Mazza, 666 F.3d at 590\\n(quoting Wash. Mut. Bank v. Superior Court, 15 P.3d 1071,\\n1081 (Cal. 2001)). The district court held that defendants\\nfailed to meet this burden, because they had “not gone\\nbeyond speculating in a general manner that the claims of\\nsome members of the putative California Class might be\\nsubject to the law of another state and that the interests of\\nanother state might be more impaired by application of\\nCalifornia law.”\\n\\n Defendants specifically point to one of the named\\nplaintiffs—Mitch Hilligoss—as an example of the alleged\\n“need to conduct choice of law inquiries as to every member\\nof the California class.” The district court found this\\nexample unpersuasive for several reasons, and we agree.\\nThe defendants argued that Illinois law should apply to\\nHilligoss’ work in California because the time he spent in\\nCalifornia was a small proportion of his overall career\\n(around two months out of a six-year career). The district\\ncourt, however, correctly read Sullivan as indicating that\\nCalifornia law should nonetheless apply to Hilligoss’\\nCalifornia work. Indeed, the proportion of time the non-\\n\\n 7\\n Moreover, as the dissent acknowledges, the California Supreme\\nCourt has expressed a strong interest in regulating wage and hour claims\\nwithin its borders. Dissent at 77.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 27\\n\\nresident employees in Sullivan worked in California was\\nquite small (and in one case, even less than the proportion of\\nHilligoss’ career spent in California): during the relevant\\nthree-year period, one worked 20 days, another 74 days, and\\nthe third 110 days. 254 P.3d at 239. Put differently, the\\nemployees in Sullivan worked in California approximately\\n1.8%, 6.7%, and 10% of the time, respectively. Id. What\\nmattered in Sullivan—and what matters here—is that when\\nthe employees worked in California, they did so for “entire\\ndays or weeks” at a time. Id. at 243.\\n\\n Second, practical considerations strongly support\\napplying California law to work performed in California, at\\nleast as a general rule; to hold otherwise “would lead to\\nbizarre and untenable results.” See Brief for Professors Peter\\nHay and Patrick J. Borchers, Dkt. No. 21, at 12–13\\n(hereinafter “Professors’ Amicus Brief”). If the law of the\\nstate in which work is performed is not the law that generally\\napplies, employers and employees alike would be subjected\\nto an unworkable scheme. Employers would be required to\\nproperly ascertain the residency status—itself not\\nnecessarily an easy task, as any student or seasonal worker\\ncould attest—of each of its employees. For every non-\\nresident employee, employers would then have to determine\\nwhether the wage laws of that employee’s state of residence\\napply extraterritorially, and then come up with different rules\\nfor each of its employees according to their state of residence\\nand any extraterritorial application of their home state’s\\nlaws. This would mean that at a single worksite, employees\\nworking side-by-side in the same position would not only be\\nowed vastly different minimum wages, but also that an\\nemployer would need to set different rules for meal and rest\\nbreaks for different employees, and so on and so forth. It\\ncannot be in any state’s legitimate “interest” to foist such an\\n\\n28 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nadministrative nightmare upon both employers and\\nemployees.\\n\\n Such a scenario would also result in an enormous\\ncompetitive advantage—or disadvantage—for prospective\\nemployees based solely on their state of residency.\\nEmployers would be incentivized to hire residents of states\\nwith low minimum wages and otherwise employer-friendly\\nwage laws, while residents of states with higher minimum\\nwages and more protective employment laws would\\nsuddenly be far less appealing. Amici Professors Hays and\\nBorchers persuasively point out that as defendants would\\nhave it, a college student still domiciled in Seattle while\\nattending a Nebraska university would have to be paid $15\\nper hour at a part-time job in Nebraska, “nearly double\\nNebraska’s minimum wage of $8 per hour.” Professors’\\nAmicus Brief at 13. This, of course, would put the student\\nat a crushing disadvantage; what rational employer would\\nhire her?\\n\\n Moreover, given the administrative cost involved in\\nattempting to comply with a patchwork of multiple states’\\nwage laws at a single workplace, some employers might\\ninstead choose to stick to hiring only resident employees, or\\nperhaps only non-resident employees from a particular state\\n(presumably one with a low minimum wage and minimally\\nprotective employment laws). 8\\n\\n\\n\\n 8\\n The California class consisted of those players who participated in\\nthe California League, which plays games exclusively within California\\nduring the championship season. The Arizona and Florida classes\\nconsisted of those who performed during spring training, extended\\nspring training, and the instructional leagues in those states. Thus, the\\ndissent’s fear that employers will be required to research applicable state\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 29\\n\\n We do not foreclose the possibility that there could be\\nsome circumstances in which a proper application of\\nCalifornia’s choice-of-law rules might lead to the\\napplication of another state’s wage and hour laws to work\\nperformed in California. Nor do we create a per se rule or\\nan unrebuttable presumption. We hold only that, given the\\nabove considerations, we are more than satisfied that the\\ndistrict court did not err in concluding that under Sullivan,\\nCalifornia law applies to the California class.\\n\\n B.\\n\\n We next address whether the district court erred in\\ndetermining that choice-of-law considerations defeated\\npredominance and adequacy for the proposed Arizona and\\nFlorida Rule 23(b)(3) classes, and conclude that the district\\ncourt’s determination must be reversed. Our conclusion is\\nanimated in part by several of the considerations outlined\\nabove, which apply with equal force to the Arizona and\\nFlorida classes. Moreover, the aforementioned enormous\\npractical implications of a contrary holding would be just as\\nproblematic and unworkable in Arizona and Florida as in\\nCalifornia.\\n\\n 1.\\n\\n With those considerations in mind, we apply California’s\\nthree-step governmental interest analysis, and conclude that\\nArizona law should apply to the work performed in Arizona,\\nand Florida law to the work performed in Florida. At the\\nfirst step, we agree with defendants that the differences in\\nstate law are “material,” meaning that “they make a\\n\\nlaws whenever an employee crosses state lines is overstated. Dissent at\\n84.\\n\\n30 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\ndifference in this litigation.” Mazza, 666 F.3d at 590. For\\nexample, some states have more expansive definitions of\\n“work,” others have differing available defenses, and we\\nhave previously held that the elements for a quantum meruit\\nclaim—alleged in both the Arizona and Florida classes—\\n“vary materially from state to state.” Id. at 591 (citing\\nCandace S. Kovacic, A Proposal to Simplify Quantum\\nMeruit Litigation, 35 Am. U.L. Rev. 547, 558–60 (1986)).\\n\\n 2.\\n\\n “Because the relevant laws differ,” we must “next\\nexamine each jurisdiction’s interest in the application of its\\nown law under the circumstances of the particular case to\\ndetermine whether a true conflict exists.” Sullivan, 254 P.3d\\nat 245 (alteration, internal quotation marks, and citation\\nomitted). We are not persuaded, as defendants contend, that\\na “true” conflict exists.\\n\\n First, under California’s choice-of-law principles, “a\\njurisdiction ordinarily has the predominant interest in\\nregulating conduct that occurs within its borders.” 9 Mazza,\\n666 F.3d at 592 (internal quotation marks omitted) (quoting\\nMcCann v. Foster Wheeler LLC, 225 P.3d 516, 534 (Cal.\\n2010)). The dissent contends that “California has long\\nrejected” this approach. Dissent at 71. In noting these\\nprinciples, we do not ignore the evolution of California’s\\nchoice of law doctrine. We recognize that the California\\nSupreme Court “renounced the prior rule, adhered to by\\ncourts for many years, that in tort actions the law of the place\\n\\n 9\\n Wage and hour laws are typically categorized as “conduct-\\nregulating,” as opposed to “loss-allocating.” See Professors’ Amicus\\nBrief at 15–16 (citing Hay, Borchers & Symeonides, Conflict of Laws\\n874–78 (5th ed. 2010)).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 31\\n\\nof the wrong was the applicable law in a California forum\\nregardless of the issues before the court” when it adopted the\\ngovernmental interest approach. Hurtado v. Superior Court,\\n522 P.2d 666 (Cal. 1974). Yet the California Supreme Court\\nhas acknowledged that while it “no longer follows the old\\nchoice-of-law rule that generally called for application of the\\nlaw of the jurisdiction in which a defendant’s allegedly\\ntortious conduct occurred without regard to the nature of the\\nissue that was before the court . . . California choice-of-law\\ncases nonetheless continue to recognize that a jurisdiction\\nordinarily has the predominant interest in regulating conduct\\nthat occurs within its borders.” McCann 225 P.3d at 534\\n(internal quotation marks and citation omitted) (emphasis in\\noriginal).\\n\\n Thus, when conducting the governmental interest\\nanalysis, we must also recognize that a state ordinarily has\\nthe predominant interest in regulating conduct within its\\nborders. We draw this conclusion not from California’s\\ninterest in regulating conduct within its own borders, but\\nfrom California’s choice-of-law principles. 10 Thus these\\n\\n 10\\n See e.g., McCann, 225 P.3d at 534, 537 (recognizing that although\\nCalifornia no longer uniformly applied the law of the jurisdiction in\\nwhich the allegedly tortious conduct occurred, Oklahoma’s interests\\n“would be more impaired if its law were not applied” as the plaintiff’s\\nexposure to asbestos occurred in Oklahoma); Reich v. Purcell, 67 432\\nP.2d 727, 730 (Cal. 1967) (“Missouri is concerned with conduct within\\nher borders and as to such conduct she has the predominant interest of\\nthe states involved.”); Castro v. Budget Rent-A-Car Sys., Inc., 65 Cal.\\nRptr. 3d 430, 442 (Cal. Ct. App. 2007) (“The accident and Castro's injury\\noccurred within Alabama's borders, thus giving Alabama a presumptive\\ninterest in controlling the conduct of those persons who use its roadways,\\nabsent some other compelling interest to be served by applying\\nCalifornia law.”); Hernandez v. Burger, 162 Cal. Rptr. 564, 568 (Cal. Ct.\\nApp. 1980) (“It is true that the place of the wrong is no longer treated as\\na controlling factor where application of the law of another jurisdiction\\n\\n32 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nprinciples are not limited to the California class but apply to\\nthe Florida and Arizona classes as well. See Mazza, 666 F.3d\\nat 593 (“The district court did not adequately recognize that\\neach foreign state [not just California] has an interest in\\napplying its law to transactions within its borders.”). The\\ndistrict court erred in ignoring these principles as a starting\\npoint, instead faulting plaintiffs for not addressing “in detail\\nthe interests of either Arizona or Florida in applying their\\nlaw” and focusing on the absence of Florida or Arizona cases\\nakin to Sullivan—despite the strong indications that Arizona\\nand Florida have the “predominant interest” in applying their\\nlaws to work performed within their state. See Mazza, 666\\nF.3d at 592.\\n\\n Second, Sullivan relied on several different\\nconsiderations to arrive at its conclusion that the existence\\nof a true conflict was “doubtful, at best”: (1) the states in\\nwhich the employees resided did not express an intent to\\napply their laws extraterritorially; (2) the employees’ states\\nof residence did not have a “legitimate interest” in shielding\\nan employer from California’s wage laws as to work\\nperformed in California; and (3) federalism and due process\\nmade extraterritorial reach doubtful under the\\ncircumstances. See 254 P.3d at 245–47. Although\\ndefendants vigorously argue that the first of those rationales\\nis inapplicable here—as discussed in greater detail below—\\nat a minimum, the second and third rationales do apply, and\\nweigh against the existence of a true conflict.\\n\\n\\nhaving a connection with the accident will serve a legitimate interest or\\npolicy of the other jurisdiction. However, the situs of the injury remains\\na relevant consideration.”); Cable v. Sahara Tahoe Corp., 155 Cal. Rptr.\\n770, 777 (Cal. Ct. App. 1979) (“The state with the ‘predominant’ interest\\nin controlling conduct normally is the state in which such conduct occurs\\nand is most likely to cause injury.”).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 33\\n\\n As to the first rationale, both defendants and the dissent\\ncontend that several states have expressed an interest in\\napplying their wage and hour laws to work performed\\noutside the state. In support of their position, they cite to a\\nhandful of cases where courts (largely district courts or\\nintermediate state courts, with the exceptions of West\\nVirginia and Washington) 11 have applied one state’s wage\\nlaws to work performed at least partially in another state.\\nFor several reasons, we are unpersuaded by defendants’\\narguments. For one, we read Sullivan as indicating that\\nunder California’s choice-of-law principles, a state has a\\nlegitimate interest in applying its wage laws extraterritorially\\nonly in two limited circumstances, neither of which apply\\nhere: one, when a state’s resident employee of that state’s\\nresident employer leaves the state “temporarily during the\\ncourse of the normal workday,” and two, “when the\\ntraveling, resident employee of a domestic employer would\\notherwise be left without the protection of another state’s\\n\\n\\n 11\\n In New v. Tac & C Energy, Inc., 355 S.E.2d 629 (W. Va. 1987),\\nthe West Virginia Supreme Court of Appeals applied its own conflict-\\nof-laws principles—relying on the Restatement (Second) of Conflicts\\n§ 196—to conclude that while there was a presumption that the law of\\nthe state where services were rendered applies, the presumption could be\\novercome by showing that another state had a “more significant\\nrelationship to the transaction and the parties.” Id. at 631. Where all\\nparties were residents of West Virginia, the employment contract was\\nmade and partially performed in West Virginia, and the plaintiffs were\\nonly in Kentucky for the duration of the work, the court concluded that\\nthe presumption was overcome and that West Virginia “had the more\\nsignificant connection to the employment relationship.” Id. California’s\\nchoice-of-law test, of course, does not utilize the “more significant\\nrelationship” test for choice-of-law questions in the wage and hour\\ncontext. See Sullivan, 254 P.3d at 244. New is therefore unpersuasive\\nhere. We discuss the Washington Supreme Court case below.\\n\\n34 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nlaw.” Id. at 242, 246 (citations, internal quotation marks,\\nand alterations omitted).\\n\\n Moreover, the cases on which defendants and the dissent\\nrely are, in large part, both factually and procedurally\\ninapposite to the circumstances of this case. 12 For example,\\ndefendants rely heavily on Bostain v. Food Exp., Inc., 153\\nP.3d 846, 851 (Wash. 2007) to argue that Washington has an\\ninterest in applying its wage laws extraterritorially. As the\\nCalifornia Supreme Court held in Sullivan, however, Bostain\\n“says nothing about a case such as this”—that is, a case\\nwhich (1) involves work performed entirely in one state, and\\n(2) presents an unavoidable conflict-of-laws issue. 254 P.3d\\nat 243. In Bostain, by contrast, either Washington law\\napplied to the work performed in both Washington and other\\nstates, or else no state’s law applied. Id. at 243, 246.\\nSignificantly, Bostain interpreted an overtime statute that\\nspecifically delineated the circumstances under which its\\nprovisions would apply to interstate truck drivers; as the\\nWashington Supreme Court noted, interstate truck drivers by\\ndefinition perform some of their work out of state. 153 P.3d\\nat 848–51. The statute at issue in Bostain did “not limit the\\nrequirement for overtime pay to hours worked” within the\\nstate’s borders. Id. at 851. Similarly, here, defendants point\\n\\n 12\\n Defendants’ repeated citation to Gonyou v. Tri-Wire Eng'g Sols.,\\nInc., 717 F. Supp. 2d 152 (D. Mass. 2010) is illustrative. In Gonyou, a\\nMassachusetts resident employee of a Massachusetts employer worked\\nlargely, although not entirely, in Connecticut. Id. at 153–54. The\\ndefendant filed a motion to dismiss on the ground that the Massachusetts\\novertime statute did not apply to work performed in Connecticut. Id. at\\n154–55. The court denied the motion but emphasized the limited nature\\nof its ruling: “As is eminently clear, this is a motion to dismiss and this\\nruling is strictly limited to the facts and circumstances of this case and\\nthis motion.” Id. at 155.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 35\\n\\nto no state statutes potentially applicable to the Arizona and\\nFlorida class members that limit their application to work\\nperformed within the state.\\n\\n 3.\\n\\n Although the existence of a “true” conflict is\\nquestionable, we need not decide whether a true conflict\\nexists, as the third step of California’s governmental interest\\ntest yields a clear answer: the laws of Arizona and Florida\\nshould apply to the work performed wholly within their\\nrespective boundaries. 13 See Sullivan, 254 F.3d at 247. As\\nthe California Supreme Court has explained the step three\\ninquiry:\\n\\n [T]he court does not “weigh” the conflicting\\n governmental interests in the sense of\\n determining which conflicting law\\n manifested the “better” or the “worthier”\\n\\n 13\\n Furthermore, in many of the cases cited by the dissent to\\ndemonstrate that some states have asserted an interest in applying their\\nwage and hour laws outside of their borders, courts have looked closely\\nat where the relevant work is performed. See e.g., Pierre v. Gts\\nHoldings, Inc., No. 15 CIV. 143 (PAC), 2015 WL 7736552, at *3–*4\\n(S.D.N.Y. Nov. 30, 2015) (concluding that New York labor laws apply\\nbecause, among other things, the majority of the plaintiff’s chauffeured\\nrides were conducted in New York); Baxi v. Ennis Knupp & Assocs.,\\nInc., No. 10-CV-6346, 2011 WL 3898034, at *14 (N.D. Ill. Sept. 2,\\n2011) (denying a motion to dismiss Illinois labor law claims because the\\nplaintiff, a foreign resident, performed some work in Illinois); Friedrich\\nv. U.S. Computer Sys., Inc., No. CIV. A. 90-1615, 1996 WL 32888, at\\n*8 (E.D. Pa. Jan. 22, 1996) (concluding that a Pennsylvania labor law\\napplies to the plaintiffs because the jury found the plaintiffs were “based\\nin Pennsylvania,” even if they were not residents of the state); Dow, 989\\nN.E.2d at 914 (concluding that Massachusetts law applied because,\\ngiven the nature of the plaintiff’s work, the work “sensibly may be\\nviewed as having ‘occurred’ in Massachusetts”).\\n\\n36 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n social policy on the specific issue. An\\n attempted balancing of conflicting state\\n policies in that sense is difficult to justify in\\n the context of a federal system in which,\\n within constitutional limits, states are\\n empowered to mold their policies as they\\n wish. Instead, the process can accurately be\\n described as a problem of allocating domains\\n of law-making power in multi-state\\n contexts—by determining the appropriate\\n limitations on the reach of state policies—as\\n distinguished from evaluating the wisdom of\\n those policies. Emphasis is placed on the\\n appropriate scope of conflicting state policies\\n rather than on the “quality” of those policies.\\n\\nMcCann, 225 P.3d at 533–34 (alterations and citations\\nomitted) (emphasis added).\\n\\n As discussed above, in Mazza, we faithfully applied the\\nprinciple under California’s choice-of-law jurisprudence\\nthat “a jurisdiction ordinarily has the predominant interest in\\nregulating conduct that occurs within its borders.” Mazza,\\n666 F.3d at 592 (internal quotation marks omitted) (quoting\\nMcCann, 225 P.3d at 534). We thus had no trouble\\nconcluding at step three that “each class member’s consumer\\nprotection claim should be governed by the consumer\\nprotection laws of the jurisdiction in which the transaction\\ntook place.” Id. at 594. Notably, we reached this conclusion\\nwithout specifically inquiring into the interests potentially\\nexpressed by any state’s statutory language or case law.\\nRather, our conclusion was dictated by the principle,\\ndiscussed above, that a jurisdiction ordinarily has the\\npredominant interest in regulating conduct within its own\\nborders. Id. at 591–92 (first citing State Farm Mut. Auto.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 37\\n\\nIns. Co. v. Campbell, 538 U.S. 408, 422 (2003); and then\\nciting McCann, 225 P.3d at 534).\\n\\n Moreover, in Sullivan, the court concluded that to\\nsubordinate California’s ability to apply its own wage laws\\nto work performed within the state would “unquestionably”\\ncause greater impairment to California than to the states that\\nmight seek to apply their wage laws to work performed by\\ntheir residents within California. 254 P.3d at 247. As\\ndescribed previously, while this holding was influenced by\\nthe absence of an expression of interest by Arizona or\\nColorado in applying their laws extraterritorially, it did not\\nrise or fall on that ground. See id. at 244–47. And although\\ndefendants point to a handful of cases that have entertained\\nthe potential application of one state’s wage laws to work\\nperformed in another state, they have not pointed to a single\\nstate with a potentially-applicable statute that expresses a\\nclear interest in applying to work performed wholly outside\\nthe state.\\n\\n But even if defendants were able to identify any states\\nthat had unambiguously expressed an interest in applying\\ntheir wage laws to work performed entirely in another state,\\nSullivan strongly militates against concluding that such an\\nexpression of interest would be adequate to overcome the\\nprinciple that the state in which the conduct at issue occurs\\nhas the “predominant interest” in applying their own law.\\nSee Mazza, 666 F.3d at 592–94; Sullivan, 254 P.3d at 245–\\n47. Forcing Arizona or Florida to allow the application of\\nother states’ wage laws in this case would be just as\\ndestructive to the balance Arizona and Florida have struck\\nbetween protecting workers and fostering a hospitable\\nbusiness environment within their states as allowing the\\napplication of Colorado or Arizona law in Sullivan would\\nhave been to the balance California struck between those\\n\\n38 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nsame interests. See Sullivan, 254 P.3d at 246–47. The\\ndistrict court fundamentally misunderstood the proper\\napplication of California’s choice-of-law principles—which,\\nwhen correctly applied, indicate that Arizona law should\\ngovern the Arizona class, and Florida law the Florida class.\\n\\n C.\\n\\n We next address whether the district court erred in\\nrefusing to certify a Rule 23(b)(2) class for unpaid work at\\ndefendants’ training facilities in Arizona and Florida on the\\nsole basis that choice-of-law issues undermined\\n“cohesiveness” and therefore made injunctive and\\ndeclaratory relief inappropriate. Because the district court’s\\nerrors in its choice-of-law analysis relating to the proposed\\nArizona and Florida Rule 23(b)(3) classes apply equally to\\nits refusal to certify the proposed Rule 23(b)(2) class, we\\nalso reverse the denial of the (b)(2) class.\\n\\n We further hold that the district court erred in imposing\\na “cohesiveness” requirement for the proposed Rule 23(b)(2)\\nclass. Although we have never explicitly addressed whether\\n“cohesiveness” is required under Rule 23(b)(2), courts that\\nhave imposed such a test treat it similarly to Rule 23(b)(3)’s\\npredominance inquiry14—something we have previously\\nrejected in no uncertain terms. See Walters v. Reno, 145 F.3d\\n1032, 1047 (9th Cir. 1998) (“[W]ith respect to 23(b)(2) in\\nparticular, the government’s dogged focus on the factual\\ndifferences among the class members appears to\\n\\n 14\\n The similarity between “cohesiveness” and predominance is\\nperhaps unsurprising, given that the Supreme Court described the\\npredominance inquiry under 23(b)(3) as testing whether a class is\\n“sufficiently cohesive to warrant adjudication by representation.”\\nAmchem Products, Inc. v. Windsor, 521 U.S. 591, 623 (1997) (emphasis\\nadded).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 39\\n\\ndemonstrate a fundamental misunderstanding of the rule.\\nAlthough common issues must predominate for class\\ncertification under Rule 23(b)(3), no such requirement exists\\nunder 23(b)(2).”); see also 2 Newberg on Class Actions\\n§ 4:34 (5th ed. 2012) (describing similarity between\\npredominance under Rule 23(b)(3) and “cohesiveness”\\nunder Rule 23(b)(2) in courts that have adopted it). We\\ntherefore remand for the district court to consider anew\\nwhether to certify the proposed Rule 23(b)(2) class. 15\\n\\n III.\\n\\n Having addressed the impact of choice-of-law questions,\\nwe turn to the issue next up at bat: whether the district court\\nerred in concluding that plaintiffs could meet the\\npredominance requirement for the proposed California,\\nFlorida, and Arizona (b)(3) classes through a combination of\\nrepresentative evidence and application of the “continuous\\nworkday” rule.\\n\\n Rule 23(b)(3)’s predominance requirement requires\\ncourts to ask “whether the common, aggregation-enabling\\nissues in the case are more prevalent or important than the\\nnon-common, aggregation-defeating, individual issues.”\\nTyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1045\\n(2016) (quoting 2 Newberg on Class Actions § 4:49 (5th ed.\\n2012)). A proposed (b)(3) class may be certified as long as\\n“one or more of the central issues in the action are common\\nto the class and can be said to predominate . . . even though\\nother important matters will have to be tried separately, such\\n 15\\n While the parties advanced numerous arguments regarding (b)(2)\\ncertification in the district court, and advance similar arguments—along\\nwith a few new ones—before us, we decline to pass on those other issues\\nin the first instance. See Stockwell, 749 F.3d at 1113, 1116–17; Davis v.\\nNordstrom, Inc., 755 F.3d 1089, 1094–95 (9th Cir. 2014).\\n\\n40 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nas damages or some affirmative defenses peculiar to some\\nindividual class members.” Id. (quoting 7AA C. Wright, A.\\nMiller, & M. Kane, Federal Practice and Procedure § 1778\\n(3d ed. 2005)).\\n\\n “[P]redominance in employment cases is rarely defeated\\non the grounds of differences among employees so long as\\nliability arises from a common practice or policy of an\\nemployer.” 7 Newberg on Class Actions § 23:33 (5th ed.\\n2012). Although the existence of blanket corporate policies\\nis not a guarantee that predominance will be satisfied, such\\npolicies “often bear heavily on questions of predominance\\nand superiority.” In re Wells Fargo Home Mortg. Overtime\\nPay Litig., 571 F.3d 953, 958 (9th Cir. 2009).\\n\\n Whether the district court was correct in concluding that\\nplaintiffs had satisfied the predominance requirement hinges\\non the application of two longstanding wage-and-hour\\ndoctrines to this case: first, the burden-shifting framework\\ninitially set forth in the Supreme Court’s seminal decision in\\nAnderson v. Mt. Clemens, 328 U.S. 680 (1946) and recently\\nexpanded upon in Tyson Foods v. Bouaphakeo, 136 S. Ct.\\n1036 (2016); and second, the so-called “continuous\\nworkday” rule. We address each of these doctrines and their\\napplication to this case in turn.\\n\\n A.\\n\\n In Mt. Clemens, the Supreme Court acknowledged the\\ndifficult bind that employees frequently confronted when\\nseeking to bring wage-and-hour claims against their\\nemployers: if their employers had failed to maintain proper\\ntimekeeping records, proving the hours of uncompensated\\nwork often posed “an impossible hurdle for the employee.”\\n328 U.S. at 687. Mt. Clemens held that such a catch-22 was\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 41\\n\\nnot in line with “the remedial nature of [the FLSA] 16 and the\\ngreat public policy which it embodies.” Id. After all, “[s]uch\\na result would place a premium on an employer’s failure to\\nkeep proper records in conformity with his statutory duty; it\\nwould allow the employer to keep the benefits of an\\nemployee’s labors without paying due compensation.” Id.\\n\\n To address this problem, Mt. Clemens established its\\nlandmark burden-shifting framework for actions in which\\nthe employer has kept inaccurate or inadequate records: if an\\nemployee “proves that he has in fact performed work for\\nwhich he was improperly compensated” and “produces\\nsufficient evidence to show the amount and extent of that\\nwork as a matter of just and reasonable inference,” then the\\nburden “shifts to the employer to come forward with\\nevidence of the precise amount of work performed or with\\nevidence to negative the reasonableness of the inference to\\nbe drawn from the employee’s evidence.” Id. at 687–88. If\\nthe employer does not rebut the employee’s evidence,\\ndamages may then be awarded to the employee, “even\\nthough the result be only approximate.” Id. at 688.\\n\\n Mt. Clemens explicitly rejected the notion that allowing\\napproximate damages in such situations would be unfair due\\nto its speculative and imprecise nature or because employers\\nsometimes make good-faith mistakes over what constitutes\\ncompensable “work”:\\n\\n\\n\\n 16\\n Although Mt. Clemens was decided under the FLSA, its holding\\nhas been consistently applied in the context of state wage-and-hour\\nclaims as well. See, e.g., Tyson, 136 S. Ct. at 1045–48; Torres v. Mercer\\nCanyons Inc., 835 F.3d 1125, 1140 (9th Cir. 2016); Hernandez v.\\nMendoza, 245 Cal. Rptr. 36, 39–40 (Cal. Ct. App. 1988) (applying Mt.\\nClemens to claims under California wage and hour law).\\n\\n42 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n The employer cannot be heard to complain\\n that the damages lack the exactness and\\n precision of measurement that would be\\n possible had he kept records in accordance\\n with the [statutory] requirements . . . And\\n even where the lack of accurate records\\n grows out of a bona fide mistake as to\\n whether certain activities or non-activities\\n constitute work, the employer, having\\n received the benefits of such work, cannot\\n object to the payment for the work on the\\n most accurate basis possible under the\\n circumstances . . . In such a case it would be\\n a perversion of fundamental principles of\\n justice to deny all relief to the injured person,\\n and thereby relieve the wrongdoer from\\n making any amend for his acts.\\n\\nId. (internal quotation marks and citation omitted).\\n\\n Seventy years after Mt. Clemens addressed the use of\\nrepresentative evidence at the trial stage to show damages,\\nTyson extended Mt. Clemens’ holding to answer two\\nimportant questions: whether representative evidence may\\nbe used at the class certification stage, and whether\\nrepresentative evidence may also be used to establish\\nliability in addition to damages. In Tyson, employees who\\nworked in more than 400 jobs across three departments at a\\nmeat processing plant sued under the FLSA and an Iowa\\nwage law, alleging that Tyson had not paid them overtime\\nfor time they spent donning and doffing protective gear; the\\nemployees also sought certification of a Rule 23 class and a\\nFLSA collective action. 136 S. Ct. at 1041–42.\\n\\n The district court certified the class and collective\\nactions, rejecting Tyson’s arguments that the claims were\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 43\\n\\ninappropriate for resolution on a classwide and collective\\nbasis due to the dissimilarity in the types of protective gear\\nworn and the variations in time spent donning and doffing\\nthat gear. Id. at 1042–43. Because Tyson had not kept\\nrecords of the donning and doffing time, plaintiffs relied on\\nrepresentative evidence to demonstrate both liability17 and\\ndamages: employee testimony, video recordings, and—most\\nsignificantly—an expert study that computed an estimated\\namount of time spent donning and doffing for each of the\\nthree departments based on hundreds of video observations.\\nId. at 1043. Although the expert estimated that the time\\nspent donning and doffing was 18 minutes per day for two\\nof the departments and 21.25 minutes for the other, id., the\\nsurvey data showed a great deal of variation in how long it\\ntook individual employees to don and doff. Id. at 1055\\n(Thomas, J., dissenting). Specifically, the time spent\\ndonning ranged from around thirty seconds to more than ten\\nminutes, and the time doffing varied from under two minutes\\nto over nine minutes. Id. After a jury verdict in the\\nemployees’ favor (albeit one that awarded less than half of\\nthe damages recommended by the employees’ expert based\\non the survey data), Tyson moved to decertify the class and\\nset aside the jury verdict, arguing that this variance made\\nclass and collective certification inappropriate. Id. at 1044–\\n\\n\\n\\n 17\\n Because the employees brought only overtime claims (as opposed\\nto minimum wage or other wage claims), “each employee had to show\\nhe or she worked more than 40 hours a week, inclusive of time spent\\ndonning and doffing, in order to recover.” Tyson, 136 S. Ct. at 1043.\\nThat the majority permitted the use of representative evidence to\\nestablish “an otherwise uncertain element of liability”—i.e., whether\\nclass members worked more than 40 hours per week—was one of the\\nkey bases for Justice Thomas’s vigorous dissent. See id. at 1057–59\\n(Thomas, J., dissenting).\\n\\n44 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n45. The district court denied the motion, and the Eighth\\nCircuit affirmed. Id.\\n\\n Tyson sought certiorari on the grounds that using\\nrepresentative evidence “manufactures predominance by\\nassuming away the very differences that make the case\\ninappropriate for classwide resolution,” “absolves each\\nemployee of the responsibility to prove personal injury,” and\\nstrips the employer of their ability to “litigate its defenses to\\nindividual claims.” Id. at 1046. Rejecting these arguments,\\nthe Supreme Court affirmed the class and collective\\ncertifications. Id. at 1046–47. Because of Tyson’s\\ndereliction of their recordkeeping duties, the employees\\nwere entitled to “introduce a representative sample to fill an\\nevidentiary gap created by the employer's failure to keep\\nadequate records.” Id. at 1047. The Court held that if the\\nrepresentative sample introduced were admissible and\\n“could have sustained a reasonable jury finding as to hours\\nworked in each employee's individual action, that sample is\\na permissible means of establishing the employees' hours\\nworked in a class action.” Id. at 1046–47.\\n\\n Stated another way, Tyson concluded that even where\\n“reasonable minds may differ” about whether representative\\nevidence is sufficiently probative of the requirements for\\nliability for a particular cause of action—in Tyson, whether\\nit was probative of the “time actually worked by each\\nemployee”—that question is to be resolved by the jury, not\\nat the class certification stage. Id. at 1049 (“The District\\nCourt could have denied class certification on this ground\\n[whether the representative evidence was “probative as to\\nthe time actually worked by each employee”] only if it\\nconcluded that no reasonable juror could have believed that\\nthe employees spent roughly equal time donning and\\ndoffing.”) (emphasis added). If the proffered representative\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 45\\n\\nevidence, however, were “statistically inadequate or based\\non implausible assumptions,” it “could not lead to a fair or\\naccurate estimate of the uncompensated hours an employee\\nhas worked.” Id. at 1048–49. But where the evidence is\\nadmissible—for expert evidence, using the Daubert\\nstandard—then the “no reasonable juror” standard at the\\nclass certification stage applies. See id. at 1049.\\n\\n B.\\n\\n Having established the parameters of when\\nrepresentative evidence may be used at the class certification\\nstage, we address the second significant wage-and-hour\\ndoctrine relevant to this case: the “continuous workday”\\nrule. The rule was first promulgated by the Department of\\nLabor (DOL) consistent with the Supreme Court’s decisions\\ninterpreting the FLSA prior to the enactment of the Portal-\\nto-Portal Act 18 in 1947. IBP, Inc. v. Alvarez, 546 U.S. 21,\\n27–28 (2005). It presumes that once the beginning of the\\n\\n 18\\n In response to what Congress perceived as excessively expansive\\njudicial interpretations of what constitutes compensable work under the\\nFLSA, IBP, 546 U.S. at 27–28, it passed the Portal-to-Portal Act to\\nexempt certain activities as compensable under FLSA:\\n\\n “(1) walking, riding, or traveling to and from the actual\\n place of performance of the principal activity or\\n activities which such employee is employed to\\n perform, and\\n\\n (2) activities which are preliminary to or postliminary\\n to said principal activity or activities,\\n\\nwhich occur either prior to the time on any particular workday at which\\nsuch employee commences, or subsequent to the time on any particular\\nworkday at which he ceases, such principal activity or activities.” 61\\nStat. 86–87 (codified at 29 U.S.C. § 254(a)).\\n\\n46 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nworkday is triggered, an employee performs compensable\\nwork throughout the rest of the day until the employee\\ncompletes their last principal activity (or the last activity\\nwhich is “integral and indispensable” to the employee’s\\nprincipal activities)—whether or not the employee actually\\nengages in work throughout that entire period. See id. at 28,\\n32–37; see also Alvarez v. IBP, Inc., 339 F.3d 894, 907 (9th\\nCir. 2003), aff’d, 546 U.S. 21 (2005) (holding that under the\\ncontinuous workday rule, “work time [is] continuous, not the\\nsum of discrete periods”).\\n\\n Of course, this rule raises inevitable questions: when\\ndoes the workday begin, and when does it end? The DOL\\ndefines the “workday” to generally mean “the period\\nbetween the commencement and completion on the same\\nworkday of an employee's principal activity or activities.”\\n29 C.F.R. § 790.6(b). The Supreme Court expanded upon\\nthis definition, interpreting “principal activity or activities”\\nto also include “all activities which are an integral and\\nindispensable part of the principal activities.” IBP, 546 U.S.\\nat 29–30 (internal quotation marks and citation omitted).\\nThus, any activity which is “integral and indispensable” to\\nprincipal activities, even if performed outside of a scheduled\\nshift, triggers the beginning of the “workday.” Id. at 31–37.\\n“Among the activities included as an integral part of a\\nprincipal activity are those closely related activities which\\nare indispensable to its performance,” 29 C.F.R. § 790.8(c),\\nsuch as knife-sharpening performed outside of a scheduled\\nshift by butchers at a meatpacking plant. Mitchell v. King\\nPacking Co., 350 U.S. 260, 261–63 (1956).\\n\\n C.\\n\\n With all of that in mind, we turn to how these two\\ndoctrines impact this case, and more specifically, whether\\nthe district court was correct in concluding that the\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 47\\n\\ncombination of Tyson and the continuous workday rule\\nenabled plaintiffs to show that they meet Rule 23(b)(3)’s\\npredominance requirement. Defendants contend that the\\ndistrict court erred in holding that plaintiffs had\\ndemonstrated predominance for two main reasons:\\n(1) because the Main Survey asked only about arrival and\\ndeparture times at the ballpark and not about what activities\\nthe players actually performed while at the ballpark,\\nplaintiffs cannot rely on the continuous workday theory\\nbecause there is no way to determine the beginning or end of\\nthe “workday,” and (2) the Main Survey revealed significant\\nvariations in players’ arrival and departure times, even\\namong players employed by the same MLB franchise.\\n\\n This task requires us to address the proposed Arizona\\nand Florida classes separately from the California class. As\\nan initial matter, however, we note that despite defendants’\\nrepeated suggestions to the contrary, the representative\\nevidence offered by plaintiffs was not limited to just the\\nMain Survey, nor are observational studies the only type of\\nevidence permitted to fill in evidentiary gaps under Tyson.\\nWe reject defendants’ erroneous view of the record and their\\ncramped reading of Tyson.\\n\\n 1.\\n\\n As to the Arizona and Florida classes, we easily affirm\\nthe district court’s determination. Recall that these two\\nclasses cover time spent participating in spring training,\\nextended spring training, and the instructional leagues—\\nperiods during which virtually all players are completely\\nunpaid for their participation. 19 Moreover, these classes do\\n\\n 19\\n Payroll data produced by defendants reveals that of the 21,211\\nplayers who participated in spring training between the 2009 and 2015\\n\\n48 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nnot bring overtime claims, but rather allege minimum wage\\nviolations. 20 Therefore—as the district court correctly\\nheld—liability can be established simply by showing that the\\nclass members performed any compensable work. 21 That is\\neasily resolved on a classwide basis by answering two\\nquestions: (1) are the players employees of defendants, and\\n(2) do the minor league team activities during these periods\\nconstitute compensable work under the laws of either\\nArizona or Florida? We hold that these two “common,\\naggregation-enabling issues in the case are more prevalent\\n[and] important than the non-common, aggregation-\\ndefeating, individual issues,” therefore making certification\\nappropriate. Tyson, 136 S. Ct. at 1045 (quoting 2 Newberg\\non Class Actions § 4:49 (5th ed. 2012)).\\n\\n Defendants do not seriously contest that their policy is to\\ndeny players compensation during spring training, extended\\nspring training, and the instructional leagues—nor could\\nthey credibly do so, given that the MLB’s own mandatory\\n\\nseasons, only 11 were paid a salary. Put differently, a mere .005% of\\nplayers received a salary during spring training, and those 11 players\\nmay be identified through payroll records and appropriately excluded\\nfrom the class. Likewise, a small number of MLB franchises pay players\\nduring extended spring training, but these players are identifiable\\nthrough payroll records and may either be excluded from the class or,\\npotentially, placed into a subclass.\\n 20\\n The Arizona and Florida classes also bring quantum meruit\\nclaims, and the Arizona class alleges recordkeeping violations, but the\\nparties do not dispute that these claims are irrelevant to this portion of\\nour predominance analysis.\\n\\n 21\\n We also note that the Arizona class’s claims are bolstered by the\\nfact that under Arizona law, failure to keep appropriate records of hours\\nworked “raise[s] a rebuttable presumption that the employer did not pay\\nthe required minimum wage rate.” Ariz. Rev. Stat. Ann. § 23-364.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 49\\n\\ncontract “obligates Player[s] to perform professional\\nservices on a calendar year basis, regardless of the fact that\\nsalary payments are to be made only during the actual\\nchampionship playing season.” And as we have long held,\\nsuch uniform corporate policies “carry great weight for\\ncertification purposes.” In re Wells Fargo Home Mortg.\\nOvertime Pay Litig., 571 F.3d at 958. This is not the “rare[]”\\ncase where predominance is defeated despite the existence\\nof an employer’s “common practice or policy.” 7 Newberg\\non Class Actions § 23:33 (5th ed. 2012).\\n\\n We also agree with the district court that as to these\\nclasses, many of defendants’ protests go to damages, not\\nliability. Damages may well vary, and may require\\nindividualized calculations. But “the rule is clear: the need\\nfor individual damages calculations does not, alone, defeat\\nclass certification.” Vaquero v. Ashley Furniture Indus.,\\nInc., 824 F.3d 1150, 1155 (9th Cir. 2016); see Tyson, 136\\nS. Ct. at 1045 (holding that where “one or more of the central\\nissues in the action are common to the class and can be said\\nto predominate,” certification may be appropriate “even\\nthough other important matters will have to be tried\\nseparately, such as damages.” (quoting 7AA C. Wright, A.\\nMiller, & M. Kane, Federal Practice and Procedure § 1778\\n(3d ed. 2005))).\\n\\n We do not, however, mean to minimize defendants’\\ncriticisms of the Main Survey. Indeed, we agree that there\\nare a number of legitimate questions about the\\npersuasiveness of the Main Survey, especially if it were the\\nonly representative evidence submitted in support of\\ncertification. But as we have mentioned, the Main Survey\\nwas but one piece of the plaintiffs’ representative\\nevidence—evidence that also included hundreds of internal\\n\\n50 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nteam schedules and public game schedules, payroll data, and\\nthe testimony of both players and league officials.\\n\\n At minimum, if the players are “employees” under either\\nArizona or Florida law and defendants are unable to prove\\nthat any affirmative defenses apply, the team schedules will\\nserve to conclusively demonstrate that the players spent time\\nworking for which they were uncompensated. See Ariz.\\nRev. Stat. Ann. § 23-362; Ariz. Admin. Code. § R20-5-\\n1202(19) (“‘Hours worked’ means all hours for which an\\nemployee covered under the Act is employed and required\\nto give to the employer, including all time during which an\\nemployee is on duty or at a prescribed work place and all\\ntime the employee is suffered or permitted to work.”);\\n29 C.F.R. § 778.223 (“As a general rule the term ‘hours\\nworked’ will include: (a) All time during which an employee\\nis required to be on duty or to be on the employer's premises\\nor at a prescribed workplace and (b) all time during which\\nan employee is suffered or permitted to work whether or not\\nhe is required to do so.”). 22 Moreover, if plaintiffs can\\npersuade a jury that their workday began at a particular\\ntime—either because they were required to report at that\\ntime, 23 or because they arrived of their own volition but\\n\\n 22\\n We rely on interpretations of the FLSA here because Florida’s\\nconstitution provides that “case law, administrative interpretations, and\\nother guiding standards developed under the federal FLSA shall guide\\nthe construction of [the constitutional amendment providing for a\\nminimum wage] and any implementing statutes or regulations.” Fla.\\nConst. art. X, § 24.\\n\\n 23\\n See 29 C.F.R. § 790.6 (“If an employee is required to report at the\\nactual place of performance of his principal activity at a certain specific\\ntime, his ‘workday’ commences at the time he reports there for work in\\naccordance with the employer's requirement.”).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 51\\n\\nengaged in work activities upon arriving (i.e., were\\n“permitted” to work)—the continuous workday doctrine\\neliminates the need for plaintiffs to prove which activities\\nthey engaged in throughout the day. 24 See IBP, 546 U.S. at\\n28, 32–37.\\n\\n Defendants should not “be heard to complain that the\\ndamages lack the exactness and precision of measurement\\nthat would be possible had [they] kept records in accordance\\nwith the [statutory] requirements,” even if their “lack of\\naccurate records grows out of a bona fide mistake as to\\nwhether certain activities or non-activities constitute work.”\\nMt. Clemens, 328 U.S. at 688. “Having received the benefits\\nof such work, [defendants] cannot object to the payment for\\nthe work on the most accurate basis possible under the\\ncircumstances.” Id.\\n\\n 2.\\n\\n We next address whether the district court was correct to\\nhold that predominance had been met for the California\\nclass. Given the differences in the types of claims brought\\nby the California class as compared to the Arizona and\\nFlorida classes, certification of the California class is more\\ncomplex and requires additional analysis. Unlike the\\nArizona and Florida classes, the California class brought\\nclaims relating to work performed during the championship\\nseason—a time when the players do get paid, albeit not\\nmuch. As a result, in order to prove liability on their\\novertime claims, the California class must show that its\\n 24\\n A jury may also decide that for baseball players, activities like\\nhitting practice with coaches and supervised weightlifting—much like\\nknife-sharpening by butchers at a meatpacking plant— are “integral and\\nindispensable” to the principal activity of playing baseball and therefore\\ntrigger the start of the “workday.” See Mitchell, 350 U.S. at 261–63.\\n\\n52 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nmembers worked more than 8 hours in a day, more than 40\\nhours in a week, and/or worked 7 days in a workweek. See\\nCal. Labor Code § 510; Mendoza v. Nordstrom, Inc., 393\\nP.3d 375, 381–82 (Cal. 2017). Likewise, to establish\\nliability on their minimum wage claims, the California class\\nmust demonstrate that they worked hours for which they\\nwere not paid at least minimum wage—but whereas the\\nArizona and Florida classes can demonstrate liability simply\\nby showing they worked any hours, the California class’s\\nburden is made more challenging by the fact that the players\\nreceive some pay. See Cal. Labor Code §§ 1182 et seq;\\nArmenta v. Osmose, Inc., 37 Cal. Rptr. 3d 460, 466–68 (Cal.\\nCt. App. 2005). Nonetheless, a number of considerations\\nlead us to affirm the district court’s determination.\\n\\n First, as with defendants’ uniform policy of not paying\\nplayers for participation outside of the championship season,\\ndefendants do not credibly dispute that their policy is to\\nnever pay overtime and to pay a fixed salary, regardless of\\nthe actual number of hours worked. We reiterate that\\ncommon corporate policies like this “carry great weight for\\ncertification purposes,” In re Wells Fargo Home Mortg.\\nOvertime Pay Litig., 571 F.3d at 958, and that predominance\\nis “rarely” defeated in cases where such uniform policies\\nexist. See 7 Newberg on Class Actions § 23:33 (5th ed.\\n2012).\\n\\n Second, the team schedules alone—independent of the\\nMain Survey or any other evidence—may suffice to show\\novertime liability. As the district court noted, plaintiffs’\\nexpert testified that approximately 65–85% of California\\nLeague players had at least one workweek with games on all\\nseven days, and that nearly half of all workweeks included\\ngames on all seven days. For those workweeks, the players\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 53\\n\\nwould be entitled to overtime pay for their work on the\\nseventh day of the workweek. See Cal. Labor Code § 510.\\n\\n Third, and most significantly, we are persuaded that\\nunder Tyson, the representative evidence plaintiffs offered\\nwas adequate to meet their burden at this stage. As we\\nobserved in the preceding section, defendants do identify\\nmultiple legitimate criticisms of the Main Survey, and it is\\ncertainly possible that a jury may not find the Main Survey—\\neven in combination with all of plaintiffs’ other evidence—\\nadequate proof of liability (or at least not to the extent\\nplaintiffs allege). In particular, a jury may be persuaded by\\ndefendants’ arguments that players did not begin\\ncompensable work upon arriving at the ballpark or that\\nplayers stopped engaging in compensable work long before\\nthey left the ballpark, such that the Main Survey’s estimated\\narrival and departure times are insufficient to clear the\\npreponderance hurdle. As we explain below, however,\\nTyson counsels that such criticisms do not doom certification\\nhere unless no reasonable jury could conclude that the\\ncombination of the Main Survey and plaintiffs’ other\\nrepresentative evidence was probative of the amount of time\\nplayers actually spent performing compensable work. See\\nTyson, 136 S. Ct. at 1046–49. And while defendants\\ncorrectly point out that the Main Survey revealed meaningful\\nvariations in players’ arrival and departure times, the same\\nwas true of the employees’ donning and doffing times in\\nTyson—yet such variation did not preclude certification\\nthere. See id. at 1043; id. at 1055 (Thomas, J., dissenting).\\n\\n Because defendants do not challenge the district court’s\\nruling on admissibility under Daubert, the defects they have\\nidentified with the Main Survey could only have defeated\\ncertification upon a conclusion that all of the representative\\nevidence offered—the Main Survey, schedules, testimony,\\n\\n54 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nand the like—could not have “sustained a reasonable jury\\nfinding as to hours worked in each employee’s individual\\naction.” See Tyson, 136 S. Ct. at 1046–47. As in Tyson, the\\ndistrict court “made no such finding,” id. at 1049, and indeed\\nfound the opposite:\\n\\n Plaintiffs will be able to use the survey data\\n in combination with other evidence that may\\n be sufficient to allow a jury to draw\\n conclusions based on reasonable inference as\\n to when players were required to be at the\\n ballpark and how long after games they were\\n required to remain at the ballpark. . . . Thus,\\n as in Tyson Foods, it appears that\\n representative evidence can be combined\\n with actual records of time spent engaged in\\n the various activities to derive a reasonable\\n estimate of the amount of time worked by\\n class members.\\n\\nWe are then left to ask whether “the record here provides [a]\\nbasis for [us] to second-guess that conclusion.” Id.\\n\\n After reviewing the record, we are satisfied that we\\nshould not disturb the district court’s determination, in part\\ndue to California’s expansive definition of “employ” and\\n“hours worked.” 25 Under California law, to “employ”\\n\\n 25\\n Unlike Arizona and Florida law—the former of which is silent on\\nthe incorporation of FLSA doctrines, and the latter of which expressly\\nincorporates them—we are not persuaded that the continuous workday\\nrule should apply to the California class. We view California’s definition\\nof “hours worked” as more expansive and more employee-friendly than\\nunder the FLSA, even with the incorporation of the continuous workday\\nrule. The California Supreme Court has “cautioned against confounding\\nfederal and state labor law,” and has consistently held that “absent\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 55\\n\\nmeans “to engage, suffer, or permit to work.” Cal. Code\\nRegs. tit. 8, §§ 11040(2)(E), 11100(2)(E) (emphasis\\nadded). 26 “Hours worked” means “the time during which an\\nemployee is subject to the control of an employer, and\\nincludes all the time the employee is suffered or permitted to\\nwork, whether or not required to do so.” Cal. Code Regs. tit.\\n8, §§ 11040(2)(K), 11100(2)(H). Inexplicably, however,\\ndefendants claim that under California law, “time spent\\nengaging in activities that are not required by, or under the\\ncontrol of, an employer is not compensable and does not\\nbegin or extend a workday.” This is a tortured and wholly\\nunsupported reading of the law, and is manifestly contrary\\nto one of the cases defendants themselves cite in support of\\ntheir argument. See Morillion v. Royal Packing Co., 995\\nP.2d 139, 143 (Cal. 2000) (“[T]he two phrases—‘time\\nduring which an employee is subject to the control of an\\nemployer’ and ‘time the employee is suffered or permitted\\nto work, whether or not required to do so’—can also be\\ninterpreted as independent factors, each of which defines\\nwhether certain time spent is compensable as ‘hours\\nworked.’”) (citation omitted).\\n\\n\\n\\n\\nconvincing evidence of the [California agency’s] intent to adopt the\\nfederal standard for determining whether time is compensable under\\nstate law, we decline to import any federal standard, which expressly\\neliminates substantial protections to employees, by implication.”\\nTroester v. Starbucks Corp., 421 P.3d 1114, 1119 (Cal. 2018)\\n(alterations, citations, and internal quotation marks omitted).\\n\\n 26\\n The California Industrial Welfare Commission’s wage orders\\n“have the force of law.” Alvarado v. Dart Container Corp. of Cal., 411\\nP.3d 528, 532 (Cal. 2018). We need not decide today which wage order\\napplies to minor league players, as all of the most relevant orders define\\n“employ” and “hours worked” the same way.\\n\\n56 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n Indeed, Morillion counsels that “hours worked” includes\\nall time the employer “permit[s]” an employee to work, even\\nif the work is not required and the employee is not under the\\nemployer’s control. See id. Thus, a player who arrives early\\nor stays late at the ballpark of their own volition and\\nperforms “work” activities during that time is still owed\\ncompensation because the player was “permitted” to work,\\ndespite the work not being required.\\n\\n Likewise, under Morillion, if players were expected to\\narrive or depart at a particular time—whether that\\nrequirement was de facto or official—it is immaterial what\\nactivities the players actually engaged in while at the\\nballpark. Even if the players spent their time at the ballpark\\ndoing things like eating or showering, they were still under\\ntheir employer’s control and unable “to use the time\\neffectively for their own purposes,” and thus were owed\\ncompensation. See id. at 146. Indeed, Morillion explicitly\\nrejected an analogous argument by the employer in that case:\\n\\n We reject Royal’s contention that plaintiffs\\n were not under its control during the required\\n bus ride because they could read on the bus,\\n or perform other personal activities.\\n Permitting plaintiffs to engage in limited\\n activities such as reading or sleeping on the\\n bus does not allow them to use the time\\n effectively for their own purposes . . .\\n Plaintiffs were foreclosed from numerous\\n activities in which they might otherwise\\n engage if they were permitted to travel to the\\n fields by their own transportation. Allowing\\n plaintiffs the circumscribed activities of\\n reading or sleeping does not affect, much less\\n eliminate, the control Royal exercises by\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 57\\n\\n requiring them to travel on its buses and by\\n prohibiting them from effectively using their\\n travel time for their own purposes. Similarly\\n . . . listening to music and drinking coffee\\n while working in an office setting can also be\\n characterized as personal activities, which\\n would not otherwise render the time working\\n noncompensable.\\n\\nId. (alterations, internal quotation marks, and citations\\nomitted). Thus, if plaintiffs use their representative\\nevidence—especially the Main Survey and the testimony of\\nplayers and league officials—to persuade a jury that they\\nwere required to be at the ballpark at particular times, they\\nneed not show how the players spent that time.\\n\\n The fourth and final consideration weighing in favor of\\naffirming the district court’s determination is our standard of\\nreview. Abuse of discretion is always a relatively deferential\\nstandard, but when we review a grant of class certification,\\n“we accord the district court noticeably more deference than\\nwhen we review a denial.” Abdullah, 731 F.3d at 956\\n(citation omitted). Were we to review de novo, this would\\nlikely be a closer call. But as they say, tie goes to the\\nrunner—and, under our deferential standard, to the district\\ncourt.\\n\\n D.\\n\\n Finally, defendants, citing to Wal-Mart Stores, Inc. v.\\nDukes, 564 U.S. 338, 352 (2011), contend that the district\\ncourt was required to “rigorously analyze” the Main Survey,\\nrather than evaluating its admissibility under Daubert and its\\nappropriateness for meeting class certification requirements\\nunder Tyson. Tyson requires that we reject this argument.\\nThere, the Court explicitly distinguished the use of\\n\\n58 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nrepresentative evidence to establish hours worked in wage\\nand hour claims from the use of representative evidence in\\ncases like Wal-Mart. Tyson, 136 S. Ct. at 1048.\\nSpecifically—as we have explained—for wage and hour\\ncases where the employer has failed to keep proper records,\\nTyson holds that once a district court has found expert\\nevidence to be admissible, it may only deny its use to meet\\nthe requirements of Rule 23 certification if “no reasonable\\njuror” could find it probative of whether an element of\\nliability was met. Id. at 1049. Given the similarities between\\nthis case and Tyson, the rule set forward in Tyson controls,\\nand “[defendants’] reliance on Wal-Mart is misplaced.” 27\\nId. (citation omitted).\\n\\n IV.\\n\\n We next address whether the district court properly\\ncertified the FLSA collective action.\\n\\n FLSA permits employees to bring lawsuits on behalf of\\n“themselves and other employees similarly situated.” 29\\nU.S.C. § 216(b). We recently delineated the appropriate\\nstandard for FLSA collective certification in Campbell v.\\nCity of Los Angeles, 903 F.3d 1090 (9th Cir. 2018). As we\\nexplained in Campbell, “there is no established definition of\\nthe FLSA’s ‘similarly situated’ requirement, nor is there an\\nestablished test for enforcing it.” Id. at 1111 (citing Thiessen\\nv. Gen. Elec. Capital Corp., 267 F.3d 1095, 1102 (10th Cir.\\n2001)). In Campbell, we rejected both the minority\\napproach to FLSA collective certification—which treats a\\nFLSA collective as analogous to a Rule 23(b)(3) class—and\\n\\n 27\\n Tyson expressly cautioned that this rule should be read narrowly\\nand not assumed to apply outside of the wage and hour context. 136\\nS. Ct. at 1049.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 59\\n\\nthe majority “ad hoc” approach. Id. at 1111–1117. The\\nformer approach, we observed, is inconsistent with the\\nstatute itself, as well as the choice of Congress and the\\nAdvisory Committee on Rules to distinguish FLSA\\ncollectives from Rule 23 class actions. Id. at 1111–1113.\\nAnd while the latter approach—the so-called ad hoc\\napproach—is a “significant improvement” over the minority\\napproach, it has two major flaws that led us to decline to\\nadopt it. Id. at 1113–1116. First, this approach\\ninappropriately “focus[es] on differences rather than\\nsimilarities among the party plaintiffs,” leading district\\ncourts to “treat[] difference as disqualifying,” rather than\\n“treat[ing] the requisite kind of similarity as the basis for\\nallowing partially distinct cases to proceed together.” Id. at\\n1117. Second, because the ad hoc approach allows district\\ncourts to weigh “fairness and procedural considerations,” it\\n“invites courts to import, through a back door, requirements\\nwith no application to the FLSA,” such as Rule 23’s\\npredominance, adequacy, and superiority requirements. Id.\\nat 1115.\\n\\n Because of the flaws in the two predominant approaches\\nto FLSA collective certification, we instead developed our\\nown standard: “[p]arty plaintiffs are similarly situated, and\\nmay proceed in a collective, to the extent they share a similar\\nissue of law or fact material to the disposition of their FLSA\\nclaims.” Id. at 1117. Significantly, as long as the proposed\\ncollective’s “factual or legal similarities are material to the\\nresolution of their case, dissimilarities in other respects\\nshould not defeat collective treatment.” Id. at 1114\\n(emphasis omitted).\\n\\n The district court here did not have the benefit of our\\nopinion in Campbell, and instead followed the vast majority\\nof district courts in this circuit by applying the ad hoc\\n\\n60 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\napproach. See, e.g., Lewis v. Wells Fargo & Co., 669\\nF. Supp. 2d 1124, 1127 (N.D. Cal. 2009) (“Although various\\napproaches have been taken to determine whether plaintiffs\\nare ‘similarly situated,’ district courts in this circuit have\\nused the ad hoc, two-tiered approach.”). While legally\\nincorrect, we conclude that the district court’s erroneous use\\nof the ad hoc approach was harmless under the\\ncircumstances, 28 and we affirm the collective’s certification.\\n\\n The district court found that plaintiffs met their burden\\nof demonstrating they were “similarly situated,” reasoning:\\n\\n First, by eliminating the winter conditioning\\n claims and pursuing on a classwide basis only\\n claims that are based on the continuous\\n workday doctrine, Plaintiffs have\\n significantly reduced the need to engage in\\n individualized inquiries relating to the type of\\n work performed. Second, the Court is now\\n persuaded that the payroll records maintained\\n by Defendants will allow any variations in\\n compensation to be analyzed without\\n burdensome individualized inquiries. This is\\n especially true as to the spring training,\\n extended spring training and instructional\\n league claims because players generally were\\n not compensated for their participation in\\n these activities and the small fraction of\\n players who did receive compensation for\\n 28\\n As we explained in Campbell, the ad hoc approach imposes a\\nhigher bar for certification than the FLSA requires. See Campbell, 903\\nF.3d at 1114–1116. Thus, if the collective was appropriately certified\\nunder the more stringent ad hoc approach, a fortiori the collective would\\nbe appropriately certified under Campbell’s more lenient approach to\\n“similarly situated.” See id.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 61\\n\\n these activities can be identified using payroll\\n records maintained by Defendants. Third, as\\n discussed above, the Court finds that the\\n defenses asserted by Defendants to the FLSA\\n present common questions that are not likely\\n to be overwhelmed by the need to conduct\\n individualized inquiries. Finally, the\\n possibility that the Court will be required to\\n apply the laws of numerous states (or at a\\n minimum, conduct numerous choice of law\\n inquiries) is not present as to the FLSA class,\\n which will require the Court to apply only\\n federal wage and hour law.\\n\\n Defendants’ arguments in support of reversal echo those\\nthey make in relation to the Rule 23(b)(3) classes, and we\\nreject them for largely the same reasons. Cf. Tyson, 136\\nS. Ct. at 1036 (“For purposes of this case . . . if certification\\nof respondents’ class action under [Rule 23] was proper,\\ncertification of the collective action was proper as well.”).\\nWe therefore expand on our earlier reasoning only briefly.\\n\\n Because the FLSA collective covers work performed\\nduring spring training, extended spring training, and the\\ninstructional leagues—that is, work for which the players\\nreceived no pay—we affirm the certification of the collective\\nfor that work. Specifically, for these time periods, two\\ncommon legal questions drive the litigation: are the players\\nemployees, and do the activities they perform during those\\ntimes constitute compensable work? As nearly all players\\nare unpaid during these time periods, if the answers to those\\ntwo questions are resolved in plaintiffs’ favor, liability may\\nbe established by showing that the players performed any\\nwork.\\n\\n62 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n We also affirm the district court’s certification of the\\nFLSA collective as to plaintiffs’ overtime claims, although\\nthis holding requires additional explanation. Critical to our\\ndecision is that plaintiffs allege a single, FLSA-violating\\npolicy—the failure to pay overtime under any\\ncircumstances—and argue a common theory of defendants’\\nstatutory violations: that defendants “suffer or permit”\\nplaintiffs to perform compensable work before and after\\nscheduled practice and game times. These are “similar\\nissue[s] of law or fact material to the disposition of their\\nFLSA claims,” thus making plaintiffs “similarly situated.”\\nCampbell, 903 F.3d at 1117. And as previously discussed,\\nwe believe a reasonable jury could find that all of plaintiffs’\\nevidence—not just the Main Survey, but also the schedules,\\ntestimony, and payroll data—sustains a “just and reasonable\\ninference” as to the hours players actually worked. See\\nTyson, 136 S. Ct. at 1046–47.\\n\\n Specifically, there are several overlapping ways that\\nplaintiffs may be able to rely on their representative evidence\\nto persuade a jury that they have worked overtime hours for\\nwhich they were not compensated. Under any of these\\nscenarios, the continuous workday rule lends significant\\nassistance to plaintiffs by eliminating the need for plaintiffs\\nto prove exactly which activities they engaged in throughout\\nthe day. See IBP, 546 U.S. at 28, 32–37.\\n\\n First, plaintiffs could potentially use their evidence—\\nparticularly the Main Survey, but also the testimony of\\nplayers and league officials—to establish approximate times\\nthat they were required to arrive at and depart from the\\nballpark. This would obviate the need for plaintiffs to\\ndemonstrate which activities they engaged in upon arrival or\\nprior to departure. See 29 C.F.R. § 790.6 (“If an employee\\nis required to report at the actual place of performance of his\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 63\\n\\nprincipal activity at a certain specific time, his ‘workday’\\ncommences at the time he reports there for work in\\naccordance with the employer's requirement.”); 29 C.F.R.\\n§ 778.223 (“As a general rule the term ‘hours worked’ will\\ninclude . . . [a]ll time during which an employee is required\\nto be on duty or to be on the employer’s premises or at a\\nprescribed workplace.”) (emphasis added).\\n\\n Second, plaintiffs could rely on their representative\\nevidence to demonstrate that before and after the times they\\nwere required to be at the ballpark, they still performed\\nactivities at the ballpark that were “an integral and\\nindispensable part of [their] principal activities” and were\\ntherefore compensable. See IBP, 546 U.S. at 29–30. As\\nmentioned previously, a jury may well determine that\\nactivities like batting practice or supervised weightlifting are\\nto baseball players what knife-sharpening is to butchers at a\\nmeatpacking plant—that is, activities that are “integral and\\nindispensable” to the principal activity of playing baseball.\\nSee Mitchell, 350 U.S. at 261–63. If so, such activities\\nwould trigger the start of the “workday” within the meaning\\nof the FLSA. Plaintiffs may have somewhat of an uphill\\nbattle proceeding under this second theory on a collective-\\nwide basis, but we are certainly not prepared to say that no\\nreasonable jury could find defendants liable for overtime\\nviolations under this theory. See Tyson, 136 S. Ct. at 1048–\\n49; cf. Campbell, 903 F.3d at 1117–1119 (explaining that\\npost-discovery decertification motions should be evaluated\\nunder the summary judgment standard where “overlap exists\\nbetween the availability of the collective action mechanism\\nand the merits of the underlying claim”).\\n\\n Finally, if internal team schedules establish that\\nplaintiffs had required team-related activities for forty hours\\n\\n64 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\na week, 29 then plaintiffs can establish liability simply by\\nshowing that they performed any additional work beyond\\nthose officially-scheduled times. Cf. Tyson, 136 S. Ct. at\\n1036 (Thomas, J., dissenting) (explaining that in Mt.\\nClemens, the employer was “presumptively liable to all\\nemployees because they all claimed to work 40 hours per\\nweek. All additional uncompensated work was necessarily\\nunpaid overtime.”) (citation omitted).\\n\\n Under any of these theories, damages will inevitably be\\nindividualized, at least to some extent. But just as the need\\nfor individualized damage calculations is insufficient to\\ndefeat Rule 23 certification, “[i]ndividual damages amounts\\ncannot defeat collective treatment under the more forgiving\\nstandard” for FLSA collective certification. See Campbell,\\n903 F.3d at 1117 (citing Leyva v. Medline Indus. Inc., 716\\nF.3d 510, 514 (9th Cir. 2013)). District courts are well-\\nequipped to deal with issues of individualized calculations\\nin the wage-and-hour context, and may use “any of the\\npractices developed to deal with Rule 23 classes facing\\nsimilar issues.” Id. at 18 (citing Jimenez v. Allstate Ins. Co.,\\n765 F.3d 1161, 1167 (9th Cir. 2014)).\\n\\n\\n\\n 29\\n Given the internal team schedules in the record, this may be an\\neasy task, particularly for spring training and extended spring training.\\nFor example, a spring training schedule for one of the San Francisco\\nGiants’ affiliates involved a workday beginning at 6:30 AM on the day\\nof a 1:00 PM away game, with a 50 minute window provided for transit\\nbetween the training facility and the ballpark. Assuming for the sake of\\nargument that the 1:00 PM game lasted 2.5 hours and that the return trip\\nto the training facility took the same amount of time—50 minutes—as\\nthe outgoing trip, that day alone entailed approximately 10 hours of work\\nif the players left the training facility immediately upon their return (and\\nbased on the testimony in the record, that assumption seems\\nimplausible).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 65\\n\\n As is true in all FLSA cases, underlying our decision\\ntoday is the background principle that “because the FLSA is\\na remedial statute, it must be interpreted broadly.” Lambert\\nv. Ackerley, 180 F.3d 997, 1003 (9th Cir. 1999) (en banc)\\n(citing Tennessee Coal, Iron & R. Co. v. Muscoda Local No.\\n123, 321 U.S. 590, 597 (1944)). After all, the FLSA does\\nnot deal “with mere chattels or articles of trade but with the\\nrights of those who toil.” Tennessee Coal, 321 U.S. at 597.\\nWe are satisfied that certification of the collective is not only\\nappropriate under our interpretation of “similarly situated,”\\nbut also that it is consistent with “the great public policy”\\nembodied by the FLSA. Mt. Clemens, 328 U.S. at 687.\\n\\n V.\\n\\n For the reasons explained above, we AFFIRM in part,\\nREVERSE in part, and REMAND for further proceedings\\nconsistent with this opinion.\\n\\n Plaintiffs-Appellants/Cross-Appellees shall recover\\ntheir costs on appeal.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426349/\", \"author_raw\": \"PAEZ, Circuit Judge:\"}, {\"author\": \"IKUTA, Circuit Judge, dissenting\", \"type\": \"dissent\", \"text\": \"IKUTA, Circuit Judge, dissenting:\\n\\n The proposed classes here comprise employees who\\nreside in at least 19 states, who are suing employers who are\\nheadquartered in at least 22 states, relating to work that took\\nplace in three different states. Determining whether to\\ncertify a class in these cases would (among other things)\\nrequire identifying the relevant laws of each of the\\npotentially affected jurisdictions, examining each\\njurisdiction’s interest in the application of its own law to\\ndetermine whether a true conflict exists, and then deciding\\nwhich jurisdiction’s interest would be most impaired if its\\n\\n66 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nlaw were not applied. Sullivan v. Oracle Corp., 51 Cal. 4th\\n1191, 1202–03 (2011). No wonder the district court\\nconcluded that consideration of the plaintiffs’ claims on a\\nclasswide basis would be overwhelmed by individualized\\nchoice-of-law inquiries.\\n\\n Yet the majority feels empowered to cut through all these\\ncomplexities by applying a simple rule of its devise: just\\napply the law of the jurisdiction where the work took place.\\nUnder this simple formula, each class can readily be certified\\nwithout any fuss. One may admire the simplicity of this\\nrule—but unfortunately, it is contrary to our framework for\\nanalyzing the intersection of class action and choice-of-law\\nissues, overlooks the complexity of California’s choice-of-\\nlaw rules, and creates significant practical and logistical\\nproblems. I therefore dissent.\\n\\n I\\n\\n The plaintiffs in this case are current or former Minor\\nLeague Baseball players who played during the period from\\n2009 to 2015. They sued Major League Baseball (MLB)\\n(which they argue is a joint employer of all minor league\\nplayers) and the MLB Clubs for which they worked for\\nviolations of federal and state labor laws, including the\\nfederal Fair Labor Standards Act, state minimum wage laws,\\nand state overtime laws. The plaintiffs argue that they were\\nentitled to the minimum wage and overtime rates established\\nby California, Arizona, or Florida for work they performed\\nin those states.\\n\\n MLB is an unincorporated association headquartered in\\nNew York. The MLB Clubs, which are corporate entities\\nthat own MLB teams, are members of the MLB. All told,\\nthere are 30 MLB Clubs, based in 17 states throughout the\\nUnited States (with one Club located in Canada). The MLB\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 67\\n\\nClubs employ around 6,000 minor league players. Each of\\nthese players signs a Uniform Player Contract, which\\ngoverns the employment relationship between the player and\\nan MLB Club. The Uniform Player Contract contains a New\\nYork choice-of-law provision.\\n\\n Each MLB Club is associated with at least six minor\\nleague affiliate teams; most Clubs have seven or eight.\\nMinor league affiliate teams are loose associations or\\ngroups, rather than corporate entities; they do not function as\\nemployers. The minor league teams are located in one of 44\\ndifferent states.\\n\\n Each spring, each Major League Club sends its minor\\nleague players to spring training in either Arizona or Florida.\\nFollowing spring training, the Club assigns selected\\nemployee-players to play on one or more of its minor league\\naffiliate teams. Employees who are not selected to play on\\nan affiliate team remain at the Arizona or Florida facilities\\nfor extended spring training. The Clubs reassign their\\nemployee-players to different minor league affiliate teams\\nthroughout the five-month championship season, sometimes\\nplaying on a minor league team for only a single game.\\n\\n During each championship season, the affiliate minor\\nleague teams play against other teams in one of several\\nminor leagues. One of these minor leagues, the California\\nLeague, is comprised of eight to ten minor league affiliate\\nteams. During the 2010 through the 2015 championship\\nseasons, a total of 2,113 minor league players were assigned\\nto play for affiliate teams in the California League. While\\nthe California League plays its championship season games\\nonly in California, the players participating in the California\\nLeague are employees of MLB Clubs located in one of six\\ndifferent states: California, Arizona, Ohio, Colorado,\\nWashington, or Texas. Several of the plaintiffs in this appeal\\n\\n68 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nwho played in the California League during the\\nchampionship season worked for MLB Clubs located\\noutside of California. For example, Ryan Kiel, who played\\nin the California League on the Bakersfield Braves during\\npart of the 2012 championship season, is a resident of\\nFlorida and an employee of the Cincinnati Reds, a Club\\nheadquartered in Cincinnati, Ohio. Brad McAtee, a New\\nYork resident and another representative of the California\\nclass, worked for the Colorado Rockies, a club\\nheadquartered in Denver, Colorado; he trained or played in\\nWashington, Arizona, California, and New York. And\\nanother California class representative, Mitch Hilligoss,\\nresides in Illinois and was employed by both the New York\\nYankees and the Texas Rangers. He played not only in\\nCalifornia, but also in Arizona, Texas, and South Carolina\\nduring the 2010 and 2011 seasons. In short, the potentially\\naffected jurisdictions include: (1) Arizona and Florida,\\nwhere the employees trained for varying lengths of time; (2)\\nthe states in which the players reside, which includes at least\\n19 states (only accounting for the 61 class representatives);\\nand (3) the states in which the players’ employers (the 22\\nMLB Clubs) are located. Because the employees argue that\\nMLB (headquartered in New York) is also an employer, and\\nbecause the Uniform Player Contract provides that the laws\\nof New York apply to any dispute under the contract, New\\nYork minimum wage and overtime law is likewise\\napplicable.\\n\\n Plaintiffs initially sought certification of eight classes\\nunder Federal Rule of Procedure 23(b)(3): a California\\nclass, a Florida class, an Arizona class, a North Carolina\\nclass, a New York class, a Pennsylvania class, a Maryland\\nclass, and an Oregon class. The district court declined to\\ncertify the plaintiffs’ proposed classes, in part because they\\npresented significant choice-of-law problems that could not\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 69\\n\\nbe handled on a classwide basis. The plaintiffs then moved\\nfor reconsideration, narrowing the proposed classes to the\\nFlorida and Arizona classes, 1 and the California class. 2 The\\nproposed Arizona class consists of players who are\\nemployees of Major League Baseball Clubs located in 14\\nstates, who are residents of at least 13 states (only accounting\\nfor the 25 class representatives), and who were assigned to\\nspring training in Arizona for four weeks or more. The\\nproposed Florida class consists of players who are\\nemployees of Major League Baseball Clubs located in 17\\nstates, who are residents of at least 13 states (only accounting\\nfor the 29 class representatives), and who were assigned to\\nspring training in Florida for four weeks or more. The\\nproposed California class consists of 2,113 players who are\\nemployees of the 11 Major League Baseball Clubs that had\\naffiliate teams in the California League during the 2010\\nthrough 2015 championship seasons, who are residents of at\\nleast 11 states (only accounting for the named class\\nrepresentatives), and who played on an affiliate team in the\\nCalifornia League during the 2010 through 2015\\nchampionship seasons.\\n\\n\\n\\n 1\\n The Florida and Arizona classes were defined (respectively) as\\nincluding “[a]ny person who, while signed to a Minor League Uniform\\nPlayer Contract, participated in spring training, instructional leagues, or\\nextended spring training in [Florida or Arizona] on or after Feb 7, 2009,\\nand had not signed a Major League Uniform Player Contract before\\nthen.”\\n\\n 2\\n The California class was defined as “[a]ny person who, while\\nsigned to a Minor League Uniform Player Contract, participated in the\\nCalifornia League on or after February 7, 2010, and had not signed a\\nMajor League Uniform Player Contract before then.”\\n\\n70 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n The district court declined to certify a Florida class and\\nan Arizona class of plaintiffs under Rule 23(b)(3) of the\\nFederal Rules of Civil Procedure. 3 It held that under\\nCalifornia choice-of-law principles, the problems that would\\nhave to be navigated in order to adjudicate the claims of the\\nFlorida and Arizona classes presented significant\\nindividualized issues that could not be handled on a\\nclasswide basis. We review this determination for abuse of\\n\\n\\n 3\\n Rule 23(b)(3) provides that:\\n\\n A class action may be maintained if Rule 23(a) is\\n satisfied and if: . . .\\n\\n (3) the court finds that the questions of law or fact\\n common to class members predominate over any\\n questions affecting only individual members, and\\n that a class action is superior to other available\\n methods for fairly and efficiently adjudicating the\\n controversy. The matters pertinent to these\\n findings include:\\n\\n (A) the class members’ interests in individually\\n controlling the prosecution or defense of separate\\n actions;\\n\\n (B) the extent and nature of any litigation\\n concerning the controversy already begun by or\\n against class members;\\n\\n (C) the desirability or undesirability of\\n concentrating the litigation of the claims in the\\n particular forum; and\\n\\n (D) the likely difficulties in managing a class\\n action.\\n\\nFed. R. Civ. Proc. 23(b)(3).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 71\\n\\ndiscretion. Abdullah v. U.S. Sec. Assocs., 731 F.3d 952, 956\\n(9th Cir. 2013).\\n\\n II\\n\\n A brief summary of the legal framework for deciding\\nwhether choice-of-law issues preclude certifying a class\\nunder Rule 23(b)(3) is helpful here. In short, before\\ncertifying a class under this provision, the court must find\\n“that the questions of law or fact common to class members\\npredominate over any questions affecting only individual\\nmembers.” Fed. R. Civ. P. 23(b)(3). When the plaintiffs\\nbring a class action involving multiple jurisdictions, a court\\nmust consider the impact of potentially varying state laws.\\nSee Zinser v. Accufix Research Inst., Inc., 253 F.3d 1180,\\n1188–89 (9th Cir. 2001). If the forum state’s substantive law\\nmay be constitutionally applied to parties in other states, the\\ndistrict court must apply the forum state’s choice-of-law\\nrules to determine which laws apply. See Mazza v. Am.\\nHonda Motor Co., 666 F.3d 581, 589–90 (9th Cir. 2012).\\nAfter applying the forum state’s choice-of-law rules, if the\\ndistrict court determines that the laws of only one state\\napply, then variations in state law do not raise a barrier to\\nclass certification. See id. at 590–91. But if the plaintiffs’\\nclaims must be adjudicated under the laws of multiple\\njurisdictions, the district court will have to determine\\nwhether the complexities and managerial problems defeat\\npredominance. See Zinser, 253 F.3d at 1188–89.\\n\\n The forum state here is California, and thus California’s\\nchoice-of-law rules apply. A brief dive into the history of\\nCalifornia’s choice-of-law jurisprudence indicates that\\nCalifornia has long rejected the approach that the majority\\nnow adopts.\\n\\n72 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n In the first half of the twentieth century, California courts\\nagreed that it was “the settled law in the United States that\\nan action in tort is governed by the law of the jurisdiction\\nwhere the tort was committed.” Loranger v. Nadeau, 215\\nCal. 362, 364–66 (1932), overruled in part by Reich v.\\nPurcell, 67 Cal. 2d 551 (1967). California courts would\\ntherefore generally “determine the substantive matters\\ninherent in the cause of action by adopting as their own the\\nlaw of the place where the tortious acts occurred, unless it\\n[was] contrary to the public policy of” California. Grant v.\\nMcAuliffe, 41 Cal. 2d 859, 862 (1953). This typical\\napproach was reflected in the Restatement (First) of the\\nConflict of Laws. See Restatement (First) of Conflict of\\nLaws § 377 (1934) (applying the law of “[t]he place of the\\nwrong”). California courts “assumed that the law of the\\nplace of the wrong created the cause of action and\\nnecessarily determined the extent of the liability.” Reich, 67\\nCal. 2d at 553. Therefore, when the injury at issue occurred\\nin California, courts would generally apply California law.\\nSee Loranger, 215 Cal. at 364–66.\\n\\n But this approach came under fire for being an inflexible\\nand mechanical rule. See Travelers Ins. Co. v. Workmen’s\\nComp. Appeals Bd., 68 Cal. 2d 7, 14 n.6 (1967). Moreover,\\n“[i]n a complex situation involving multi-state contacts,”\\nCalifornia courts realized that “no single state alone can be\\ndeemed to create exclusively governing rights.” Reich, 67\\nCal. 2d at 553. In response, California courts began adopting\\na more flexible approach. See, e.g., id.; Hurtado v. Super.\\nCt. of Sacramento Cty., 11 Cal. 3d 574, 581–82 (1974). In a\\n“landmark opinion . . . for a unanimous court in Reich v.\\nPurcell,” the California Supreme Court “renounced the prior\\nrule, adhered to by courts for many years, that in tort actions\\nthe law of the place of the wrong was the applicable law in a\\nCalifornia forum regardless of the issues before the court.”\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 73\\n\\nHurtado, 11 Cal. 3d at 579. Instead, California concluded\\nthat each state’s interest in applying its own law must be\\nevaluated. See id. In 1971, the Restatement (Second) of\\nConflict of Laws reflected the general movement away from\\nthe law-of-the-situs approach espoused by the First\\nRestatement by replacing it with a more flexible approach\\nthat considered each state’s interest in applying its own laws.\\nSee Restatement (Second) of Conflict of Laws § 6 (1971);\\nsee also id. introduction (describing the revised approach as\\nan “enormous change” from the “rigid rules” laid out in the\\nFirst Restatement). California courts described the new\\napproach to choice-of-law principles, which reflected the\\napproach of the Second Restatement, as a “governmental\\ninterest approach” that required consideration of the interests\\nof all the involved states. See, e.g., Dixon Mobile Homes,\\nInc. v. Walters, 48 Cal. App. 3d 964, 972 (1975). In Offshore\\nRental Co. v. Continental Oil Co., the California Supreme\\nCourt definitively announced that “[q]uestions of choice of\\nlaw are determined in California . . . by the ‘governmental\\ninterest analysis,’” which requires the court to “search to find\\nthe proper law to apply based upon the interests of the\\nlitigants and the involved states.” 22 Cal. 3d 157, 161\\n(1978).\\n\\n Today, California courts no longer apply “the old choice-\\nof-law rule that generally called for application of the law of\\nthe jurisdiction in which a defendant’s allegedly tortious\\nconduct occurred without regard to the nature of the issue\\nthat was before the court.” McCann v. Foster Wheeler LLC,\\n48 Cal. 4th 68, 97 (2010) (emphasis in original). Instead,\\nCalifornia courts apply the three-step governmental interest\\ntest. Hairu Chen v. Los Angeles Truck Ctrs., LLC, No.\\nS240245, 2019 WL 3281346, at *3 (Cal. July 22, 2019).\\n“First, the court determines whether the relevant law of each\\nof the potentially affected jurisdictions with regard to the\\n\\n74 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nparticular issue in question is the same or different.” Id.\\n(internal quotation marks omitted). If there is a difference,\\n“the court examines each jurisdiction’s interest in the\\napplication of its own law under the circumstances of the\\nparticular case to determine whether a true conflict exists.”\\nId. (internal quotation marks omitted). As the final step, “if\\nthe court finds that there is a true conflict, it carefully\\nevaluates and compares the nature and strength of the\\ninterest of each jurisdiction in the application of its own law\\nto determine which state’s interest would be more impaired\\nif its policy were subordinated to the policy of the other state,\\nand then ultimately applies the law of the state whose interest\\nwould be the more impaired if its law were not applied.” Id.\\n(cleaned up).\\n\\n Although California choice-of-law cases “continue to\\nrecognize that a jurisdiction ordinarily has the predominant\\ninterest in regulating conduct that occurs within its borders,”\\nsee McCann, 48 Cal. 4th at 97–98 (internal quotation marks\\nomitted), California courts have not relied on this general\\nprinciple to shortcut the required three-part analysis, see,\\ne.g., Sullivan, 51 Cal. 4th at 1202. Indeed, in McCann, a\\ncase on which the majority relies for its rule, Maj. Op. at 30–\\n31, the California Supreme Court walked through each of the\\nsteps of the governmental interest analysis to determine\\nwhether to apply the law of Oklahoma (where the tort\\noccurred) or California (where the plaintiff resided). 48 Cal.\\n4th at 96–98. Only after determining at the second step that\\n“each state has an interest in having its law applied under the\\ncircumstances of the present case,” id. at 96, did the court\\nproceed to the third step and determine that Oklahoma law\\napplied, in part because “a failure to apply California law on\\nthe facts of the present case will effect a far less significant\\nimpairment of California’s interest,” id. at 99 (emphasis\\nadded). In short, as the California Supreme Court recently\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 75\\n\\nexplained, “the governmental interest test is far from a\\nmechanical or rote application of various factors,” Hairu\\nChen, 2019 WL 3281346, at *5, and California courts must\\nscrupulously apply each step of the three-step test. 4\\n\\n California courts also apply the governmental interest\\nanalysis in cases where plaintiffs and defendants raise\\nchoice-of-law issues, even outside the tort context. In\\nSullivan, the California Supreme Court applied the\\ngovernmental interest analysis to a wage-and-hour dispute,\\nin a case where plaintiffs contended California’s overtime\\nlaw governed their work in California, and the defendant\\ncontended the laws of plaintiffs’ home states governed. 51\\nCal. 4th at 1202. Sullivan did not merely apply California’s\\novertime law, although California was the site where the\\nwork occurred. See id. As explained below, Sullivan made\\na detailed analysis of each of the three steps of the\\ngovernmental interest test. See id.\\n\\n At the same time as California courts were migrating\\ntowards the multifaceted governmental interest test\\nespoused by the Second Restatement, California courts also\\nadopted the Second Restatement’s approach to contractual\\nchoice of law provisions. See Gamer v. duPont Glore\\nForgan, Inc., 65 Cal. App. 3d 280, 287–88 (1976). Under\\nthis test, courts would generally defer to the law of the state\\nchosen by the parties unless either “the chosen state has no\\nsubstantial relationship to the parties or the transaction and\\n\\n 4\\n Indeed, in the California class action context, the California\\nSupreme Court has made clear there are no presumptive choice-of-law\\nrules. Rather, a “trial court cannot reach an informed decision on\\npredominance and manageability without first determining whether class\\nclaims will require adjudication under the laws of other jurisdictions and\\nthen evaluating the resulting complexity where those laws must be\\napplied.” Hairu Chen, 2019 WL 3281346, at *5.\\n\\n76 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nthere is no other reasonable basis for the parties choice, or\\n. . . application of the law of the chosen state would be\\ncontrary to a fundamental policy of a state which has a\\nmaterially greater interest than the chosen state in the\\ndetermination of the particular issue and which . . . would be\\nthe state of the applicable law in the absence of an effective\\nchoice of law by the parties.” Nedlloyd Lines B.V. v. Super.\\nCt. of San Mateo Cty., 3 Cal. 4th 459, 465 (1992).\\n\\n In undertaking the predominance analysis under Rule\\n23(b), the court is required to consider the full scope of\\nCalifornia’s choice-of-law framework, including each\\nstate’s interest in applying its own law, as well as the\\ncontractual choice-of-law provision. See Mazza, 666 F.3d at\\n590–91. If individualized choice-of-law inquiries swamp\\npredominance, then the class cannot be certified. See id.\\n\\n III\\n\\n In addressing the choice-of-law framework in the\\ncontext of a Rule 23(b) inquiry, the majority concedes that\\nthe differences in state law involved in this case are material.\\nMaj. Op. at 29–30. But instead of undertaking California’s\\nchoice-of-law analysis by identifying the relevant laws of\\neach potentially affected jurisdiction and examining each\\njurisdiction’s interest in the application of its own law, the\\nmajority sidesteps this analysis entirely by relying solely on\\nits general rule that the jurisdiction where an employee’s\\nwork occurs has the predominant interest in regulating\\nconduct that occurs within its borders. Maj. Op. at 30–35.\\nNot only is this approach contrary to substantive California\\nlaw, but the majority’s justification of this approach on\\npractical grounds is entirely misguided.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 77\\n\\n A\\n\\n First, as the above description of California law makes\\nclear, the majority misreads and misapplies substantive\\nCalifornia law. In considering whether the district court\\nerred in declining to certify the Arizona and Florida classes,\\nthe majority interprets California’s choice-of-law rules as\\nestablishing the general principle that California has the\\npredominant interest in regulating conduct occurring within\\nits borders. Maj. Op. at 31. In this vein, the majority asserts\\nthat Sullivan “strongly militates” against concluding that any\\nother state has an interest in wage and hour laws that “would\\nbe adequate to overcome the presumption that the state in\\nwhich the conduct at issue occurs has the ‘predominant\\ninterest’ in applying their own law.” Maj. Op. at 37. These\\nconclusions are wrong in two different ways.\\n\\n Most important, the majority misreads California’s\\nchoice-of-law rules to conclude that the law of the situs\\nwhere the work took place controls. This is clearly contrary\\nto California law: as shown above, California courts have\\nexpressly rejected the blanket rule that the law of the situs\\napplies, Travelers, 68 Cal. 2d at 11, and “when application\\nof the law of the place of the wrong would defeat the\\ninterests of the litigants and of the states concerned,” they do\\nnot apply that law. Reich, 67 Cal. 2d at 554; see also\\nBerhard v. Harrah’s Club, 16 Cal. 3d 313, 316, 323 (1976)\\n(applying California law where the tort occurred in Nevada\\nbut the harm was felt in California). 5 Even where, as here, a\\n\\n 5\\n The majority also errs in applying substantive California law to\\ndetermine Arizona’s and Florida’s interests in the application of their\\nown laws, the second step of California’s governmental interest test.\\nMaj. Op. at 30–32. In other words, because the California Supreme\\nCourt has expressed a strong interest in regulating wage and hour claims\\n\\n78 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\ncontractual choice-of-law provision is involved, California\\napplies the law of the parties’ choosing only after\\nconsidering the relevant state interests. See Nedlloyd, 3 Cal.\\n4th at 465. For example, in Washington Mutual Bank, FA v.\\nSuperior Court, the California Supreme Court analyzed a\\nstate class action that involved both a contractual choice-of-\\nlaw provision and the applicability of the governmental\\ninterest test. 24 Cal. 4th 906, 915 (2001). The court\\ndetermined that the test from the Restatement (Second) of\\nConflict of Laws under Nedlloyd applied to the class action,\\nid. at 918, and that if the choice-of-law provision did not\\napply under Nedlloyd, the court must undertake the\\ngovernmental interest analysis, id. at 919–21.\\n\\n Second, in the context of wage-and-hour disputes, the\\nmajority wildly overreads Sullivan. In Sullivan, the\\nCalifornia Supreme Court expressly limited its ruling to the\\nsituation before it: the state’s interest in applying California\\nlabor law to nonresident employees working for a California\\nemployer. Sullivan, 51 Cal. 4th at 1194–95. The court was\\ncareful not to address any other scenario. See id. Therefore,\\nthe majority’s extension of Sullivan to establish a general\\nrule that California has a superior interest in applying its law\\nto wage-and-hour claims that arise within its borders, Maj.\\nOp. at 37–38, (let alone generalizing the majority’s\\n\\nwithin its borders, the majority assumes that Arizona and Florida have\\nthe exact same interest. To support this assumption, the majority cites\\nCalifornia cases which determined—after the application of the\\ngovernmental interest test—that a particular foreign state had a superior\\ninterest in having its law applied. The majority fails to identify any\\nArizona or Florida opinion expressing such an interest, however. This\\nis clearly wrong. Although the district court is bound to apply the choice-\\nof-law provisions of California (the forum state), the district court may\\nnot impute California’s interest in regulating conduct within its borders\\nto Arizona and Florida.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 79\\n\\nextrapolation of California’s rule to all other states) is not\\nsupported by Sullivan.\\n\\n A brief description of Sullivan reveals the majority’s\\nerror. In Sullivan, the California Supreme Court responded\\nto a certified question regarding whether California labor\\nlaw applied to nonresident employees who worked both in\\nCalifornia and in other states for a California-based\\nemployer. 51 Cal. 4th at 1194. The employees at issue\\nworked as instructors for Oracle Corporation, a large\\nCalifornia-based company. Id. at 1194–95. Two of the\\nemployees were residents of Colorado; while they worked\\nprimarily in Colorado, they were required to travel and work\\nin other states, including California. Id. at 1195. A third\\nemployee was an Arizona resident, but worked 20 days in\\nCalifornia. Id. Oracle did not pay these employees overtime\\non the ground that they were exempt under California and\\nfederal overtime laws as instructors. Id. The employees\\nsued Oracle, seeking unpaid overtime compensation. Id.\\nThe question certified to the California Supreme Court was\\nwhether California overtime law applied to the employees’\\nwork in California. Id. at 1196.\\n\\n In its response to the certified question, the California\\nSupreme Court addressed two distinct inquiries: first,\\nwhether, as a matter of statutory construction, the California\\nLabor Code’s overtime provisions applied to work\\nperformed in California by nonresidents, id. at 1196–97, and\\nsecond, whether California’s choice-of-law principles\\ndirected the court to apply the California Labor Code to the\\nplaintiffs, id. at 1202–06. Sullivan focused on the question\\nwhether a California employer had to pay its employees\\nunder California’s overtime law or under the overtime law\\nof the state where the employees resided during the period\\nwhen the employees worked in California. See id. at 1196.\\n\\n80 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nBecause the employer in that case was Oracle, a resident of\\nCalifornia, the court did not have to consider whether the\\novertime law of the state of a nonresident employer (the\\nissue in our case) might apply.\\n\\n Sullivan first made a point of carefully examining\\nCalifornia’s overtime statute to ensure it applied to\\nnonresident employees of a California employer. Id. at\\n1197. The court noted that the plain text of the applicable\\novertime statute stated that the statute applied to “all\\nindividuals,” which would include residents and\\nnonresidents alike. Id. It also noted that the legislature knew\\nhow to exclude nonresidents when it wanted to do so,\\nbecause it had expressly exempted some out-of-state\\nemployers from complying with workers’ compensation\\nprovisions. Id. Therefore, Sullivan held the overtime statute\\nwould apply to the plaintiffs in the case before it.\\n\\n Because the statute was potentially applicable to\\nnonresidents by its terms, the California Supreme Court then\\napplied California’s three-step governmental interest test to\\ndetermine which state’s law applied. Id. at 1202–03.\\nSullivan first asked whether the overtime law of California\\nwas the same or different than the overtime laws of Colorado\\nand Arizona, where the employees resided. Id. at 1203. The\\ncourt determined that the laws were different. Id. Federal\\novertime law applied in Arizona, and federal law required\\nless overtime compensation than California. Id. Colorado\\novertime law applied in Colorado, but it too required less\\ncompensation than California. Id.\\n\\n Sullivan next examined “each jurisdiction’s interest in\\nthe application of its own law under the circumstances of the\\nparticular case to determine whether a true conflict exists.”\\nId. at 1203. Relying on the California statute and case law,\\nSullivan first noted that “California has, and has\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 81\\n\\nunambiguously asserted, a strong interest in applying its\\novertime law to all nonexempt workers, and all work\\nperformed, within its borders.” Id. Arizona had no overtime\\nlaw, and Colorado’s statute expressly did not apply out of\\nstate, so the court found that neither Arizona nor Colorado\\nhad “asserted an interest in regulating overtime work\\nperformed in other states.” Id. at 1204. Therefore, there was\\nno true conflict. See id. The court acknowledged, however,\\nthat states could have an interest in the extraterritorial\\napplication of their employment laws under certain limited\\ncircumstances. See id. at 1199.\\n\\n The final step in the governmental interest analysis was\\nto determine which state’s interest would be more impaired\\nif its policy were subordinated to the policy of the other state.\\nSee id. at 1205–06. The court concluded that California’s\\ninterests would be more impaired if nonresidents employed\\nin California were covered only by the law of the\\nnonresident’s state. Id. Among other considerations,\\nSullivan reasoned that adopting a different rule might\\nencourage California employers to hire nonresidents of\\nCalifornia to work in California. Id. at 1206. By contrast,\\nColorado and Arizona had no interest in applying their\\novertime laws to their residents working in California. See\\nid.\\n\\n Sullivan therefore concluded that California’s overtime\\nlaw “does apply to overtime work performed in California\\nfor a California-based employer by out-of-state plaintiffs in\\nthe circumstances of this case.” Id. The court did not\\naddress whether the same rule would apply for a nonresident\\nemployer.\\n\\n Contrary to the majority’s conclusion, Sullivan did not\\nestablish a rule that every California wage-and-hour law\\napplies to all persons working in California regardless of\\n\\n82 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\ntheir state of residence or their employer’s state of residence.\\nTo the contrary, rather than enunciate such a rule, Sullivan\\ncarefully analyzed the law and policy of each relevant\\njurisdiction, consistent with California’s governmental\\ninterest test. See id. at 1202–06. Sullivan expressly limited\\nits analysis to the particular facts of the case before it: a case\\ninvolving California overtime law, a California employer,\\nand employees residing in Arizona and Colorado. See id.\\nSullivan specified that it was not applying its rule to out-of-\\nstate employers, as is the case here. Id. at 1201 (noting that\\nthe court did not need to address “the asserted burdens on\\nout-of-state businesses to which Oracle refers,” in part\\nbecause “no out-of-state employer is a party to this\\nlitigation[, and] Oracle itself is based in California”).\\nFurther, Sullivan clarified that its holding did not apply to\\nany California labor law other than the overtime law,\\nexplaining, “[w]hile we conclude the applicable conflict-of-\\nlaws analysis does require us to apply California’s overtime\\nlaw to full days and weeks of work performed here by\\nnonresidents one cannot necessarily assume the same result\\nwould obtain for any other aspect of wage law.” Id. at 1201\\n(citation omitted). Indeed, “California’s interest in the\\ncontent of an out-of-state business’s pay stubs, or the\\ntreatment of its employees’ vacation time, for example, may\\nor may not be sufficient to justify choosing California law\\nover the conflicting law of the employer’s home state.” Id.\\n\\n Moreover, Sullivan acknowledged that different\\noutcomes could result under different circumstances. By\\nbeginning its analysis with the statutory language, Sullivan\\nindicated that the state legislature could decide not to apply\\nits employment laws to some employees who work in-state,\\nid. at 1197 (conducting statutory analysis to confirm that the\\nCalifornia overtime legislation applied to “any individual”),\\nor could exempt out-of-state employers who send employees\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 83\\n\\ninto California from complying with California law, as it did\\nin the case of workers’ compensation law, id., or could\\nchoose not to apply overtime law to employees who reside\\nout of state, id. at 1198. Similarly, Sullivan acknowledged\\nthat a truck driver employee based at a Washington facility\\nof a California employer could be entitled to overtime\\ncompensation under Washington law for the time he spent\\ndriving outside the state. See id. at 1200, 1204.\\n\\n In fact, Sullivan expressly rejected the arguments that it\\nwas adopting a general rule that California’s employment\\nlaws applied in all contexts, holding instead that disputes in\\neach different context would be “resolved under the\\napplicable conflict of laws analysis.” Id. at 1200. “In any\\nevent,” the court explained, “to the extent other states have\\nlegitimate interests in applying their own wage laws to their\\nown residents for work performed in California, the\\napplicable conflict-of-laws analysis takes those interests into\\naccount.” Id. at 1202. In other words, Sullivan rejected the\\nvery approach that the majority now adopts, and instead,\\nSullivan stands for the proposition that the determination of\\nwhich state’s law applies requires a careful analysis of each\\nrelevant state’s law and policies.\\n\\n B\\n\\n Second, the majority’s argument that practical\\nconsiderations compel the adoption of a general rule has the\\nsituation entirely backwards.\\n\\n The only practical consideration flagged by the majority\\nis that, absent a rule that the hours and wage laws of the situs\\nalways apply to workers within its borders, Maj. Op. at 35–\\n36, employers would be required to properly ascertain the\\nresidency status of each of its employees, to track applicable\\nstate laws, and to determine which law applies, Maj. Op. at\\n\\n84 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n27–28. Such a concern does not arise if the state law at issue\\nmerely requires a resident employer to pay each of its\\nemployees according to the resident state’s laws, even when\\nthe employee is working temporarily in another state. In\\nother words, if an MLB Club in Ohio paid each of its player–\\nemployees pursuant to Ohio overtime law, the MLB Club\\nwould have no extra burden at all. Unlike Sullivan, the\\nmajority fails to recognize that states may enact many\\ndifferent types of laws, and that conflicts between state laws\\ncan be resolved through the application of choice-of-law\\nrules. Cf. Sullivan, 51 Cal. 4th at 1201–02.\\n\\n On the other hand, the rule the majority establishes today\\ncould have dire consequences for employers and employees.\\nFor example, a rule requiring that the law of the situs always\\napplies would require employers to research and comply\\nwith various states’ laws whenever their employees traveled\\nfor short conferences or business meetings. An employer\\nwould have to research applicable state law whenever an\\nemployee traveled across state lines, including when an\\nemployee was in transit. Presumably, when an employee\\ntraveled across state lines by car or airplane, the employer\\nwould need to track the amount of time the employee spent\\nin each state during travel in order to comply with this rule.\\nSuch a rule would make it difficult for employers to\\ncompensate interstate truck drivers or traveling salespersons.\\nMoreover, the majority’s rule would also burden employees\\nwho would no longer be protected by the laws of their\\nresident state or employer’s state while traveling for work,\\nforcing the employees to earn less money for work travel.\\nRather than adopting a rule that the law of the situs applies,\\nthe better solution is faithfully adhering to long-established\\nchoice-of-law principles, which resolve the issue in a\\nreasonable and time-tested way.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 85\\n\\n IV\\n\\n Because it is not possible to derive a general rule from\\nSullivan, and California’s choice-of-law rules weigh against\\nany such rule, the majority should have considered the\\napplicability of California’s choice-of-law rules to the\\nplaintiffs’ claims.\\n\\n Given that a minimum of 22 states potentially have an\\ninterest in applying their wage and hour laws, and that (as\\nthe majority concedes) there are material differences\\nbetween the states, applying California’s three-step\\ngovernmental interest test would be a significant task.\\n\\n First, as a threshold matter, the court must analyze the\\ncontractual choice-of-law provision (i.e., New York) in the\\ngovernmental law analysis under Nedlloyd, 3 Cal. 4th at 466,\\nand the Restatement (Second) of Conflict of Laws. This\\nwould require the court to analyze whether New York law\\nhas a substantial relationship to the parties or transactions\\nhere and whether application of New York law would be\\ncontrary to Arizona’s or Florida’s interests. See id. at 465.\\n\\n Second, if the contractual choice-of-law provision does\\nnot govern, a court applying Sullivan would first have to\\ndetermine whether the minimum wage laws and overtime\\nlaws of Arizona and Florida apply by their terms to\\nnonresident employees who work for nonresident\\nemployers, Sullivan, 51 Cal. 4th at 1202–03. Assuming the\\nlaws did apply, the court would then have to identify the\\nrelevant laws of each of the potentially affected jurisdictions.\\nSee id. at 1203. It would then have to determine whether\\nthere is a conflict between the laws of Arizona and Florida,\\non the one hand, and the laws of the different states in which\\nthe employees and employers reside. See id.\\n\\n86 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n If there is a true conflict, then the court would have to\\ncompare the nature and strength of each jurisdiction’s\\ninterest in the application of its own law to determine\\nwhether a true conflict exists under the circumstances of the\\nparticular case. See id. at 1203–05. Contrary to the majority,\\nMaj. Op. at 34–35, other states have an interest in applying\\ntheir wage and hour laws outside their borders. For example,\\nthe Boston Red Sox is an MLB Club headquartered in\\nBoston, Massachusetts, and a franchise defendant in this\\nlawsuit. Massachusetts has previously applied its wage-and-\\nhour laws extraterritorially. See Dow v. Casale, 989 N.E. 2d\\n909 (Mass. App. Ct. 2013). Moreover, MLB Clubs in\\nIllinois, Pennsylvania, New York, and Washington are also\\ndefendants in this proposed class action, and courts have\\napplied wage-and-hour laws in those states extraterritorially.\\nSee Baxi v. Ennis Knupp & Assocs., Inc., No. 10-cv-6346,\\n2011 WL 3898034, at *14–15 (N.D. Ill. Sept. 2, 2011);\\nTruman v. DeWolff, Boberg & Assocs., No. 07-cv-1702,\\n2009 WL 2015126, at *2 (W.D. Pa. July 7, 2009); Friedrich\\nv. U.S. Comput. Sys., Inc., No. 90-cv-1615, 1996 WL 32888,\\nat *8 (E.D. Pa. Jan. 22, 1996); Pierre v. Gts Holdings, Inc.,\\nNo. 15-cv-143, 2015 WL 7736552, at *1, *5 (S.D.N.Y. Nov.\\n30, 2015); Bostain v. Food Express, Inc., 159 Wash.2d 700,\\n709–711 (Wash. 2007) (en banc). 6\\n\\n It is not surprising that the district court determined that\\nthis type of analysis would defeat the predominance that\\nRule 23(b)(3) requires. No two player-employees’\\n\\n 6\\n The majority notes that, in many cases, state “courts have looked\\nclosely at where the relevant work is performed” to determine whether\\nto apply the state’s laws extraterritorially. Maj. Op. at 35 n.13.\\nCertainly, state courts look to where the work is performed as one factor\\nto determine which state’s law applies. The majority errs by concluding\\nthat where the work is performed is effectively the only relevant factor\\nin the choice-of-law analysis.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 87\\n\\ncircumstances are alike; the players hail from at least 19\\nresident states, worked for one or more MLB Clubs based in\\none of 22 states for varying lengths of time, and played on\\none or more minor league affiliate teams in an assortment of\\nstates for as little as one day or as long as an entire season.\\nSullivan and California’s choice-of-law analysis require the\\ncourt to consider all of the relevant states’ laws and weigh\\nthe commensurate state interests in applying those laws. The\\nhighly individualized nature of the choice-of-law inquiry\\nwith respect to each player could swamp the predominance\\nrequired for certification under Rule 23(b)(3). See Hanlon\\nv. Chrysler Corp., 150 F.3d 1011, 1022 (9th Cir. 1998);\\nWash. Mut. Bank, FA, 24 Cal. 4th at 922. In any event, the\\ndistrict court did not abuse its discretion by refusing to\\ncertify the Florida and Arizona classes.\\n\\n For the same reason, the district court erred in certifying\\nthe California class without completing its choice-of-law\\nanalysis. Sullivan’s conclusion does not control where the\\nrelevant employer is not a California-based employer. 51\\nCal. 4th at 1197–98. While Sullivan held that California’s\\novertime laws apply to employees of a California employer\\nwho are residents of Arizona and Colorado but work\\noccasionally in California, Sullivan did not address the\\napplication of both overtime and minimum wage laws to\\nemployees of out-of-state employers who work occasionally\\nin California. Id. at 1197–98. Instead, Sullivan requires a\\ncourt to apply the three-part governmental interest analysis,\\nincluding weighing the interests of the employees’ and\\nemployers’ resident states in applying their own laws. Id. at\\n1202–03.\\n\\n Here, more than half of the MLB Clubs with minor\\nleague affiliates that play in the California League are out-\\nof-state employers. Moreover, the plaintiffs argue that the\\n\\n88 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nMLB, a New York-based entity, is also an employer. The\\nplayers themselves hail from at least 11 states, even if only\\nthe 26 class representatives named in this lawsuit were\\nincluded in the class. In addition, 68.7% to 74.7% of the\\nplayers who were assigned to a minor league affiliate in the\\nCalifornia League also played as a member of a minor league\\naffiliate in a different state during the 2010 to 2015\\nchampionship seasons. Approximately 11% of the proposed\\nclass members from the 2010 championship season were\\nassigned to an affiliate in the California League for one week\\nor less. Sullivan requires that the court weigh each relevant\\njurisdiction’s interest in applying its laws, including all of\\nthe relevant variables: whether the players are employed by\\nan out-of-state MLB Club; whether the players are\\nnonresidents of California; whether the players spent only a\\nshort time in California; whether any other state’s law might\\napply; and whether that state’s interest in applying its own\\nlaw outweighs California’s interest. See 51 Cal. 4th at 1202–\\n03. Because the choice-of-law inquiries cannot be neatly\\nsolved with a law-of-the-situs rule as the majority suggests,\\nindividual choice-of-law issues also appear to defeat\\npredominance for the California class.\\n\\n V\\n\\n No doubt the analysis of the intersection between Rule\\n23(b)(3)’s predominance inquiry and California’s choice-of-\\nlaw inquiry is multilayered and complex, particularly in a\\ncase like this one, involving different types of wage and hour\\nclaims, employers residing in multiple states, employees\\nresiding in multiple states, and three states where work was\\nperformed. But the majority errs in attempting to sidestep\\nthe analysis entirely in one fell swoop by the simple\\nexpedient of declaring that each jurisdiction generally has a\\npredominant interest in regulating conduct that occurs within\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 89\\n\\nits borders, a conclusion that is contrary to the requirement\\nthat California courts undertake the governmental interest\\nanalysis in every case. Although the majority gives lip\\nservice to the possibility of exceptions to this rule, its failure\\nto consider all the variables in this case to determine whether\\nany exception was applicable here gives the lie to such\\nclaimed flexibility. Because the majority’s conclusion that\\ncourts can sidestep a choice-of-law analysis by relying on a\\ngeneral rule is contrary to our precedents, and because it will\\nimpose burdens on employers and disadvantage employees\\nin many circumstances, I dissent.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426349/\", \"author_raw\": \"IKUTA, Circuit Judge, dissenting\"}]}"},"judge1":{"kind":"string","value":"MICHAEL R MURPHY"},"judge2":{"kind":"string","value":"RICHARD A PAEZ"},"judge3":{"kind":"string","value":"SANDRA S IKUTA"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649096/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Aaron Senne v. Kansas City Royals Baseball\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-16\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nAARON SENNE; MICHAEL LIBERTO; Nos. 17-16245\nOLIVER ODLE; BRAD MCATEE; CRAIG 17-16267\nBENNIGSON; MATT LAWSON; KYLE 17-16276\nWOODRUFF; RYAN KIEL; KYLE\nNICHOLSON; BRAD STONE; MATT D.C. No.\nDALY; AARON MEADE; JUSTIN 3:14-cv-00608-\nMURRAY; JAKE KAHAULELIO; RYAN JCS\nKHOURY; DUSTIN PEASE; JEFF\nNADEAU; JON GASTON; BRANDON\nHENDERSON; TIM PAHUTA; LEE OPINION\nSMITH; JOSEPH NEWBY; RYAN\nHUTSON; MATT FREVERT; ROBERTO\nORTIZ; WITER JIMENEZ; KRIS WATTS;\nMITCH HILLIGOSS; DANIEL BRITT;\nYADEL MARTI; HELDER VELAQUEZ;\nJORGE JIMENEZ; JORGE MINYETY;\nEDWIN MAYSONET; JOSE DIAZ; NICK\nGIARRAPUTO; LAUREN GAGNIER;\nLEONARD DAVIS; GASPAR SANTIAGO;\nGRANT DUFF; OMAR AGUILAR; MARK\nWAGNER; DAVID QUINOWSKI;\nBRANDON PINCKNEY, Individually\nand on Behalf of All Those Similarly\nSituated; JAKE OPITZ; BRETT\nNEWSOME,\n Plaintiffs-Appellants,\n\n v.\n\n2 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n\nKANSAS CITY ROYALS BASEBALL\nCORP.; MARLINS TEAMCO LLC; SAN\nFRANCISCO BASEBALL ASSOCIATES,\nLLC; OFFICE OF THE COMMISSIONER\nOF BASEBALL, DBA Major League\nBaseball, an unincorporated\nassociation; ALLAN HUBER SELIG,\n“BUD”; ANGELS BASEBALL LP; ST.\nLOUIS CARDINALS, LLC; COLORADO\nROCKIES BASEBALL CLUB, LTD.;\nCINCINNATI REDS, LLC; HOUSTON\nBASEBALL PARTNERS LLC;\nATHLETICS INVESTMENT GROUP,\nLLC; ROGERS BLUE JAYS BASEBALL\nPARTNERSHIP; PADRES L.P.; SAN\nDIEGO PADRES BASEBALL CLUB, L.P.;\nMINNESOTA TWINS, LLC; DETROIT\nTIGERS, INC.; LOS ANGELES\nDODGERS LLC; STERLING METS L.P.;\nAZPB L.P.; NEW YORK YANKEES\nP’SHIP; RANGERS BASEBALL EXPRESS,\nLLC; MILWAUKEE BREWERS\nBASEBALL CLUB, INC.; CHICAGO\nCUBS BASEBALL CLUB, LLC;\nPITTSBURGH ASSOCIATES, LP;\nBASEBALL CLUB OF SEATTLE, LLP;\nLOS ANGELES DODGERS HOLDING\nCOMPANY LLC; RANGERS BASEBALL,\nLLC,\n Defendants-Appellees.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 3\n\n Appeals from the United States District Court\n for the Northern District of California\n Joseph C. Spero, Magistrate Judge, Presiding\n\n Argued and Submitted June 13, 2018\n San Francisco, California\n\n Filed August 16, 2019\n\n Before: Michael R. Murphy, * Richard A. Paez,\n and Sandra S. Ikuta, Circuit Judges.\n\n Opinion by Judge Paez;\n Dissent by Judge Ikuta\n\n\n\n\n *\n The Honorable Michael R. Murphy, United States Circuit Judge\nfor the U.S. Court of Appeals for the Tenth Circuit, sitting by\ndesignation.\n\n4 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n SUMMARY **\n\n\n Labor Law / Class and Collective Certification\n\n The panel affirmed in part and reversed in part the\ndistrict court’s orders certifying a class and a collective\naction for wage-and-hour claims brought by minor league\nbaseball players under the Fair Labor Standards Act and\nstate law.\n\n The district court certified a California class under\nFederal Rule of Civil Procedure 23(b)(3) but denied\ncertification for Arizona and Florida classes and for a Rule\n23(b)(2) class. The district court also certified an FLSA\ncollective.\n\n The panel held that, as to the state law claims, California\nchoice-of-law rules applied. The panel held that under\nSullivan v. Oracle Corp., 254 P.3d 237 (Cal. 2011),\nCalifornia law applied to the Rule 23(b)(3) California class.\nThe panel reversed the district court’s determination that\nchoice-of-law considerations defeated the predominance and\nadequacy requirements for the proposed Arizona and Florida\nRule 23(b)(3) classes. Applying California’s three-step\ngovernmental interest analysis for choice-of-law questions,\nthe panel concluded that Arizona law should apply to the\nwork performed in Arizona, and Florida law to the work\nperformed in Florida.\n\n The panel reversed the district court’s refusal to certify a\nRule 23(b)(2) class for unpaid work at defendants’ training\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 5\n\nfacilities in Arizona and Florida on the basis that choice-of-\nlaw issues undermined “cohesiveness” and therefore made\ninjunctive and declaratory relief inappropriate. The panel\nconcluded that the district court’s errors in its choice-of-law\nanalysis relating to the proposed Arizona and Florida Rule\n23(b)(3) classes applied equally to its refusal to certify the\nRule 23(b)(2) class. The panel further held that the district\ncourt erred in imposing a “cohesiveness” requirement for the\nproposed Rule 23(b)(2) class. The panel remanded for the\ndistrict court to consider anew whether to certify the Rule\n23(b)(2) class.\n\n The panel held that plaintiffs could meet the\npredominance requirement for the proposed California,\nFlorida, and Arizona Rule 23(b)(3) classes through a\ncombination of representative evidence and application of\nthe “continuous workday” rule. The panel applied the Mt.\nClemens burden-shifting framework and the holding of\nTyson Foods v. Bouaphakeo, 136 S. Ct. 1036 (2016), that\nrepresentative evidence may be used at the class certification\nstage and may be used to establish liability in addition to\ndamages. The panel explained that the continuous workday\nrule presumes that once the beginning of the workday is\ntriggered, an employee performs compensable work\nthroughout the rest of the day until the employee completes\ntheir last principal activity. Any activity that is “integral and\nindispensable” to principal activities triggers the beginning\nof the workday. As to the Arizona and Florida classes,\ncovering alleged minimum wage violations in the lack of any\npay for time spent participating in spring training, extended\nspring training, and instructional leagues, the panel affirmed\nthe determination that the predominance requirement was\nmet. As to the California class, covering overtime and\nminimum wage claims relating to work performed during the\nchampionship season, the panel held that the district court\n\n6 SENNE V. KANSAS CITY ROYALS BASEBALL\n\ndid not abuse its discretion in concluding that defendants’\nuniform pay policy, the team schedules, and representative\nevidence, including an expert survey known as the “Main\nSurvey,” established predominance. The panel held that the\ndistrict court was not required to “rigorously analyze” the\nMain Survey, rather than evaluating its admissibility under\nDaubert and its appropriateness for meeting class\ncertification requirements under Tyson.\n\n Affirming the district court’s certification of the FLSA\ncollective action, the panel applied the standard set forth in\nCampbell v. City of L.A., 903 F.3d 1090 (9th Cir. 2018),\nwhich postdated the district court’s ruling, and held that the\ndistrict court’s use of the ad hoc approach was harmless\nerror. The panel concluded that collective certification was\nproper because plaintiffs shared similar issues of law or fact\nmaterial to the disposition of their FLSA claims and thus\nwere similarly situated.\n\n The panel affirmed the district court’s certification of the\nCalifornia Rule 23(b)(3) class and the FLSA collective\naction, reversed the district court’s refusal to certify Arizona\nand Florida classes and a Rule 23(b)(2) class, and remanded\nfor further proceedings.\n\n Dissenting, Judge Ikuta wrote that the district court\ncorrectly concluded that consideration of plaintiffs’ claims\non a classwide basis would be overwhelmed by\nindividualized choice-of-law inquiries. She wrote that the\nmajority’s rule, applying the law of the jurisdiction where\nthe work took place, was contrary to the court’s framework\nfor analyzing the intersection of class action and choice-of-\nlaw issues, overlooked the complexity of California’s\nchoice-of-law rules, and created significant practical and\nlogistical problems.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 7\n\n COUNSEL\n\nRobert L. King (argued) and Garrett R. Broshuis, Korein\nTillery LLC, St. Louis, Missouri; Bruce L. Simon and\nBenjamin E. Shiftan, Pearson Simon & Warshaw LLP, San\nFrancisco, California; Daniel L. Warshaw and Bobby Pouya,\nPearson Simon & Warshaw LLP, Sherman Oaks, California;\nfor Plaintiffs-Appellants.\n\nElise M. Bloom (argued), Adam M. Lupion, and Mark D.\nHarris, Proskauer Rose LLP, New York, New York; John E.\nRoberts, Proskauer Rose LLP, Boston, Massachusetts; for\nDefendants-Apellees.\n\nDavid C. Frederick and Jeffrey A. Love, Kellogg Hansen\nTodd Figel & Frederick PLLC, Washington, D.C., for Amici\nCuriae Professors Peter Hay and Patrick J. Borchers.\n\nAllan Steyer and Donald Scott MacRae, Steyer Lowenthal\nBoodrookas Alvarez & Smith LLP, San Francisco,\nCalifornia, for Amici Curiae Professional Hockey Players\nAssocation; Association of Minor League Umpires; Office\nand Professional Employee’s International Union; and\nUnited Steel, Paper and Forestry, Rubber, Manufacturing,\nEnergy, Allied and Industrial Service Workers International\nUnion.\n\n8 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n OPINION\n\nPAEZ, Circuit Judge:\n\n It is often said that baseball is America’s pastime. In this\ncase, current and former minor league baseball players\nallege that the American tradition of baseball collides with a\ntradition far less benign: the exploitation of workers. We are\ntasked with deciding whether these minor league players\nmay properly bring their wage-and-hour claims on a\ncollective and classwide basis.\n\n BACKGROUND\n\n I.\n\n Most major professional sports in America have their\nown “farm system” for developing talent: for the National\nBasketball Association, it’s the G-League; for the National\nHockey League, it’s the American Hockey League; and for\nMajor League Baseball (MLB), it’s Minor League Baseball.\nMLB and its thirty franchise teams rely heavily on this\nextensive minor league system, which has nearly 200\naffiliates across the country and employs approximately\n6,000 minor league players. Nearly all MLB players begin\ntheir careers in the minor leagues. Each minor league club\nis associated with one of the thirty franchise MLB teams.\n\n The minor league system is governed by the Major\nLeague Rules (MLRs), which dictate the terms of\nemployment and compensation for both minor and major\nleague players. Under the MLRs, all minor league players\nare required to sign a seven-year Uniform Player Contract\n(UPC). Ostensibly, players are required to sign the UPC for\n“morale” and “to produce the similarity of conditions\nnecessary for keen competition.”\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 9\n\n The UPC “obligates Player[s] to perform professional\nservices on a calendar year basis, regardless of the fact that\nsalary payments are to be made only during the actual\nchampionship playing season.” It describes its scope as\nsetting “the terms and conditions of employment during all\nperiods in which Player is employed by Club as a Minor\nLeague Player.” Players are paid by the MLB franchise\naffiliated with the minor league team for which they play.\nUnder the UPC, first-year players are paid a fixed salary of\n$1,100 per month during the regular (“championship”)\nseason that runs from April through September. In addition\nto their salaries during the championship season, some\nplayers receive signing or performance-related bonuses and\ncollege scholarships.\n\n Beginning in early March each year, the minor league\naffiliates conduct spring training in Arizona and Florida;\nevery MLB franchise operates a minor league training\ncomplex in one of these two states. The parties dispute\nwhether spring training is required, but the UPC strongly\nindicates that it is mandatory. 1 Virtually all players are\nunpaid during spring training.\n\n Spring training lasts approximately four weeks, until the\nchampionship season begins in April. Some players attest\nthat spring training entails working seven days a week, with\nno days off. During spring training, teams typically have\n\n 1\n The UPC provides that “Player’s duties and obligations under [the\nUPC] continue in full force and effect throughout the calendar year,\nincluding Club’s championship playing season, Club’s training season,\nClub’s exhibition games, Club’s instructional, post-season training or\nwinter league games, any official play-off series, any other official post-\nseason series in which Player shall be required to participate . . . and any\nremaining portions of the calendar year.”\n\n10 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nscheduled activities in the morning prior to playing games in\nthe afternoon. For example, a team spring training schedule\nfor one of the San Francisco Giants’ affiliates describes that\nat 6:30 AM, there was an “Early Van for Treatment and\nEarly Work” 2; at 7:00 AM, the “Regular Van” departed; at\n7:45 AM, the “Early Work” began; and then between 9:00\nAM and 11:00 AM, the team would perform activities such\nas “Stretch,” “Throwing Program,” and “Batting Practice.”\nLunch was to be at 11:00 AM, before a 12:10 PM bus to a\nneighboring city for a 1:00 PM away game.\n\n At the conclusion of spring training in early April, some\nplayers are assigned to minor league affiliates, and begin\nplaying games in the championship season. During the\nchampionship season, minor league teams play games either\nsix or seven days per week. The championship season lasts\naround five months, beginning in April and ending in\nSeptember. One of the regular season leagues within minor\nleague baseball is the California League, which—as the\nname implies—plays games exclusively within California.\n\n Players who are not assigned to play for affiliates in the\nchampionship season stay at the Arizona or Florida facilities\nfor “extended spring training.” Extended spring training\ncontinues until June, and involves similar activities to spring\ntraining. Although most players do not get paid during\nextended spring training, as many as seven MLB clubs do\npay for work during extended spring training due to an\nambiguity in the MLRs over when players are permitted to\nbe paid.\n\n\n\n 2\n The schedule instructed players to “CHECK [the] BOARD FOR\nEARLY WORK.”\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 11\n\n After the championship season ends in September, some\nplayers participate in the “instructional leagues,” which run\nfrom approximately mid-September to mid-October. The\nparties dispute whether participation in the instructional\nleagues is mandatory for the players involved, although as\nwith spring training, the UPC strongly implies that\nparticipation is required. Activities and schedules during the\ninstructional league are similar to spring training. And just\nas with spring training, players are virtually never paid for\nparticipation in the instructional league.\n\n II.\n\n Plaintiffs are forty-five current and former minor league\nbaseball players who bring claims under the federal Fair\nLabor Standards Act (FLSA) and the wage-and-hour laws of\nCalifornia, Arizona, and Florida against MLB, MLB\nCommissioner Bud Selig, and a number 3 of MLB franchises.\nPlaintiffs allege that defendants do not pay the players at all\nduring spring training, extended spring training, or the\ninstructional leagues. They further allege that because\nplayers are “employees” and the activities the players\nperform during those periods constitute compensable work,\ndefendants have unlawfully failed to pay them at least\nminimum wage. And according to plaintiffs, while the\nplayers are paid—albeit not much—during the\nchampionship season, they routinely work overtime, for\nwhich they are never compensated as a matter of policy.\n\n In May 2015, plaintiffs filed their Second Amended\nConsolidated Class Action Complaint, which alleged wage-\n\n 3\n Plaintiffs originally named all 30 MLB franchises as defendants,\nbut eight of the franchises were subsequently dismissed for lack of\npersonal jurisdiction.\n\n12 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nand-hour claims under the laws of eight states and the FLSA;\nplaintiffs also sought certification of a FLSA collective\naction. The district court preliminarily certified the FLSA\ncollective in October 2015. Notice was sent to\napproximately 15,000 current and former minor league\nplayers, of which more than 2,200 opted in.\n\n In 2016, defendants moved to decertify the FLSA\ncollective, while plaintiffs moved to certify a Rule 23(b)(2)\nclass as well as Rule 23(b)(3) classes under the laws of eight\nstates. The district court denied certification for all proposed\nRule 23(b)(3) classes, concluding that predominance was not\nsatisfied for two primary reasons. Senne v. Kansas City\nRoyals Baseball Corp., 315 F.R.D. 523, 572, 577–84 (N.D.\nCal. 2016). First, the court concluded that predominance\nwas defeated by the choice-of-law issues presented by the\nproposed classes, given that (1) the winter off-season\ntraining claims entailed work performed in dozens of\ndifferent states with no common schedule or situs; and\n(2) the championship season claims involved frequent travel\nbetween state lines for away games. Id. at 580–81. The\ndistrict court also determined that the inclusion of claims for\nwinter off-season work fatally undermined predominance, as\nthe court would be required to undertake an overwhelming\nnumber of individualized inquiries to determine which\nactivities constituted compensable “work” and how much\ntime was spent doing “work.” Id. at 577–84. For similar\nreasons, the court held that plaintiffs were not “similarly\nsituated” and therefore decertified the FLSA collective. Id.\nat 585–86. The court also granted the defendants’ motion to\nexclude an expert survey (the “Pilot Survey”) submitted by\nplaintiffs, finding that its methodology and results did not\nsatisfy the requirements of Federal Rule of Evidence 702 and\nDaubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S.\n579 (1993). Id. at 586–90. The court further refused to\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 13\n\ncertify the proposed Rule 23(b)(2) class, concluding that\nbecause the plaintiffs were all former—rather than current—\nplayers, they lacked standing to represent a (b)(2) class. Id.\nat 584–85.\n\n Plaintiffs moved for reconsideration, narrowing their\nproposed classes significantly in response to the concerns the\ndistrict court expressed in its initial certification order.\nPlaintiffs requested Rule 23(b)(3) certification of an Arizona\nclass and a Florida class for work performed during spring\ntraining, extended spring training, and the instructional\nleagues in those states. Plaintiffs also moved for\ncertification of a 23(b)(3) California class, covering players\nwho participated in the California League during the\nchampionship season. Additionally, plaintiffs sought to\ncertify a reworked FLSA collective of players who\nparticipated in the California League or in spring training,\nextended spring training, and the instructional leagues. In\naddition to the 23(b)(3) classes and FLSA collective,\nplaintiffs requested certification of a Rule 23(b)(2)\ninjunctive relief class consisting of current minor league\nplayers who participate in spring training, extended spring\ntraining, or the instructional leagues in Florida or Arizona.\nTo cure the court’s earlier concerns about standing, four\ncurrent minor league players moved to intervene to represent\nthe proposed (b)(2) class.\n\n On reconsideration, plaintiffs argued that they could\nmeet Rule 23(b)(3)’s predominance requirement and\nFLSA’s “similarly situated” requirement through a\ncombination of the use of representative evidence and\napplication of the so-called “continuous workday” rule. 4\n\n 4\n As we shall explain, the continuous workday rule presumes that\nonce the beginning of the workday is triggered, an employee performs\n\n14 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nPlaintiffs’ representative evidence took a variety of forms,\nincluding an expert survey (the “Main Survey”), hundreds of\nteam schedules, payroll data, and testimony from both\nplayers and league officials. The most controversial piece of\nevidence was the Main Survey, which plaintiffs argued\nserved as representative evidence of hours worked,\nparticularly when used in concert with a continuous workday\ntheory.\n\n The Main Survey asked players to report the times they\n“most often” arrived and departed from the ballpark or\ntraining facility during the championship season, spring\ntraining, extended spring training, and the instructional\nleagues, and asked players to estimate how much time they\nspent eating meals while at the ballpark. The survey did not,\nhowever, ask players about the kinds of activities they\nperformed at the facilities, or how much time they spent\nperforming particular activities. Given these purported\nshortcomings, defendants moved to exclude the Main\nSurvey, and further argued that even if the survey were\nadmissible under Daubert, it still could not be used to meet\nthe predominance and “similarly situated” requirements due\nto its alleged flaws. The district court denied defendants’\nmotion to exclude the Main Survey, finding it admissible\nunder Daubert and concluding that defendants’ challenges\nwent “to the weight of the Survey and not its admissibility”\nand were “better left to a jury to evaluate.” The district court\nfurther concluded that the Main Survey could be used in\ncombination with other evidence—such as team schedules,\ntestimony, and payroll data—to meet Rule 23(b)(3)’s\n\ncompensable work throughout the rest of the day until the employee\ncompletes their last principal activity or the last activity which is\n“integral and indispensable” to the employee’s principal activities. IBP,\nInc. v. Alvarez, 546 U.S. 21, 28, 32–37 (2005).\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 15\n\npredominance and FLSA’s “similarly situated”\nrequirements, observing that certifying the classes and the\nFLSA collective “will not preclude Defendants from\nchallenging the sufficiency of the Main Survey and\nPlaintiffs’ damages model on summary judgment and/or at\ntrial.”\n\n Because it concluded that the predominance and\n“similarly situated” requirements could be met with the use\nof representative evidence and application of the continuous\nworkday rule, the district court recertified the narrowed\nFLSA collective and certified a California (b)(3) class.\nHowever, the district court denied certification for the\nArizona, Florida, and (b)(2) classes, holding that choice-of-\nlaw concerns defeated predominance for the Arizona and\nFlorida classes and undermined “cohesiveness” for the\n(b)(2) class.\n\n At defendants’ request, the district court certified the\nFLSA collective certification order for interlocutory review\nunder 28 U.S.C. § 1292. Plaintiffs petitioned us for\npermission to appeal the denial of certification for the\nArizona, Florida, and Rule 23(b)(2) classes, and defendants\nlikewise petitioned to appeal the certification of the\nCalifornia class; we granted both petitions, consolidating\nthose cross-appeals with the FLSA collective appeal.\n\n STANDARD OF REVIEW\n\n We review for abuse of discretion the district court’s\nclass certification rulings, and review for clear error any\nfindings of fact the district court relied upon in its\ncertification order. Parsons v. Ryan, 754 F.3d 657, 673 (9th\nCir. 2014). A district court’s choice of law determinations,\nhowever, are reviewed de novo. Mazza v. Am. Honda Motor\nCo., 666 F.3d 581, 589 (9th Cir. 2012). A district court\n\n16 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nabuses its discretion where it commits an error of law, relies\non an improper factor, omits a substantial factor, or engages\nin a clear error of judgment in weighing the correct mix of\nfactors. Stockwell v. City & County of San Francisco, 749\nF.3d 1107, 1113 (9th Cir. 2014) (citing Bateman v. Am.\nMulti-Cinema, Inc., 623 F.3d 708, 712 (9th Cir. 2010)).\nWhen we review a grant of class certification, “we accord\nthe district court noticeably more deference than when we\nreview a denial.” Abdullah v. U.S. Sec. Assocs., Inc., 731\nF.3d 952, 956 (9th Cir. 2013) (quoting Wolin v. Jaguar Land\nRover N. Am., LLC, 617 F.3d 1168, 1171 (9th Cir. 2010)).\n\n ANALYSIS\n\n To paraphrase the Chief Justice, these complex appeals\nrequire us to call a great number of balls and strikes, as both\nparties raise numerous challenges to the district court’s\ncertification order. For their part, plaintiffs challenge the\ndistrict court’s decision to deny certification for the Arizona\nand Florida Rule 23(b)(3) classes and the Rule 23(b)(2) class\non the grounds that choice-of-law issues defeated the\npredominance requirement for the Arizona and Florida\n(b)(3) classes and also thwarted “cohesiveness” for the\nproposed (b)(2) class. Defendants, on the other hand, contest\nthe district court’s certification of the California (b)(3) class,\narguing first that choice-of-law issues defeat both\npredominance and adequacy, and second, that plaintiffs\ncannot meet the predominance requirement through the use\nof their proffered representative evidence: the Main Survey,\nteam schedules, payroll records, deposition testimony, and\ndeclarations. Defendants further charge that the district\ncourt erred in certifying the FLSA collective because\nplaintiffs’ representative evidence does not show that the\ncollective members are “similarly situated.” Defendants\nalso contend that the district court erred by not “rigorously\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 17\n\nanalyzing” plaintiffs’ expert evidence at the class and\ncollective certification stage. We address each argument in\nturn.\n\n I.\n\n Class certification is governed by Federal Rule of Civil\nProcedure 23. As a threshold matter, a party seeking class\ncertification must satisfy the four requirements of Rule\n23(a): (1) numerosity; (2) commonality; (3) typicality; and\n(4) adequacy of representation. 5 “Class certification is\nproper only if the trial court has concluded, after a ‘rigorous\nanalysis,’ that Rule 23(a) has been satisfied.” Parsons, 754\nF.3d at 674 (quoting Wang v. Chinese Daily News, Inc., 737\nF.3d 538, 542–43 (9th Cir. 2013)).\n\n In addition to the requirements of Rule 23(a), a proposed\nclass must also meet the requirements of one or more of the\n“three different types of classes” set forth in Rule 23(b).\nLeyva v. Medline Industries, Inc., 716 F.3d 510, 512 (9th Cir.\n2013). Here, plaintiffs proposed classes under two of Rule\n23(b)’s class types: Rule 23(b)(3) and 23(b)(2). A class may\nbe certified under Rule 23(b)(3) only if the district court\n“finds that the questions of law or fact common to class\nmembers predominate over any questions affecting only\nindividual members, and that a class action is superior to\nother available methods for fairly and efficiently\nadjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). Of\nthese two requirements—predominance and superiority—\nonly predominance is at issue on appeal. “The\npredominance inquiry focuses on ‘the relationship between\n\n 5\n Of Rule 23(a)’s four requirements, defendants contest only\nadequacy on appeal; their arguments pertaining to adequacy have to do\nwith choice-of-law issues.\n\n18 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nthe common and individual issues’ and ‘tests whether\nproposed classes are sufficiently cohesive to warrant\nadjudication by representation.’” Vinole v. Countrywide\nHome Loans, Inc., 571 F.3d 935, 944 (9th Cir. 2009)\n(quoting Hanlon v. Chrysler Corp., 150 F.3d 1011, 1022 (9th\nCir. 1998)). In determining whether the predominance\nrequirement is met, courts have a “duty to take a close look\nat whether common questions predominate over individual\nones” to ensure that individual questions do not “overwhelm\nquestions common to the class.” Comcast Corp. v. Behrend,\n569 U.S. 27, 34 (2013) (internal quotation marks and citation\nomitted).\n\n Rule 23(b)(2), on the other hand, requires only that “the\nparty opposing the class ha[ve] acted or refused to act on\ngrounds that apply generally to the class, so that final\ninjunctive relief or corresponding declaratory relief is\nappropriate respecting the class as a whole.” Fed. R. Civ. P.\n23(b)(2). Although 23(b)(2) classes are most common in the\ncivil rights context, “we have certified many different kinds\nof Rule 23(b)(2) classes.” Parsons, 754 F.3d at 686.\n\n II.\n\n We first address whether choice-of-law issues fatally\nundermine plaintiffs’ proposed Rule 23 classes. The district\ncourt’s decision was split on the impact of choice-of-law\nquestions: as to the proposed Rule 23(b)(3) California class,\nthe court held that choice-of-law concerns defeated neither\nRule 23(b)(3)’s predominance requirement nor Rule 23(a)’s\nadequacy requirement. Yet as to the proposed 23(b)(3)\nArizona and Florida classes, the district court held the\nopposite: that choice-of-law issues posed an insurmountable\nhurdle to meeting both predominance and adequacy.\nSimilarly, the court determined that choice-of-law questions\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 19\n\nmade certification of the proposed Rule 23(b)(2) class\ninappropriate.\n\n Concerns over which state’s laws apply to a proposed\nclass “do not necessarily preclude a 23(b)(3) action.”\nHanlon, 150 F.3d at 1022. But “[u]nderstanding which law\nwill apply before making a predominance determination is\nimportant when there are variations in applicable state law,”\nand potentially varying state laws may defeat predominance\nin certain circumstances. Zinser v. Accufix Research Inst.,\nInc., 253 F.3d 1180, 1189 (9th Cir. 2001) opinion amended\non denial of reh’g, 273 F.3d 1266 (9th Cir. 2001). We have\nbeen particularly concerned about the impact of choice-of-\nlaw inquiries in nationwide consumer class actions and\nproducts liability cases. See, e.g., Mazza, 666 F.3d at 585,\n591–94; Zinser, 253 F.3d at 1184–90.\n\n A district court considering state law claims brought in\nfederal court must utilize the choice-of-law rules of the\nforum state—here, California. Klaxon Co. v. Stentor\nElectric Mfg. Co., 313 U.S. 487, 496–97 (1941). “By\ndefault, California courts apply California law unless a party\nlitigant timely invokes the law of a foreign state, in which\ncase it is the foreign law proponent who must shoulder the\nburden of demonstrating that foreign law, rather than\nCalifornia law, should apply to class claims.” In re Hyundai\n& Kia Fuel Econ. Litig., 926 F.3d 539, 561 (9th Cir. 2019)\n(en banc) (internal quotation marks and citations omitted).\nTo meet their burden, the objectors must satisfy California’s\nthree-step governmental interest test, used to resolve choice\nof law issues. Id.\n\n First, the court determines whether the\n relevant law of each of the potentially\n affected jurisdictions with regard to the\n particular issue in question is the same or\n\n20 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n different. Second, if there is a difference, the\n court examines each jurisdiction’s interest in\n the application of its own law under the\n circumstances of the particular case to\n determine whether a true conflict exists.\n Third, if the court finds that there is a true\n conflict, it carefully evaluates and compares\n the nature and strength of the interest of each\n jurisdiction in the application of its own law\n to determine which state’s interest would be\n more impaired if its policy were subordinated\n to the policy of the other state, and then\n ultimately applies the law of the state whose\n interest would be the more impaired if its law\n were not applied.\n\nKearney v. Salomon Smith Barney, Inc., 137 P.3d 914, 922\n(Cal. 2006) (internal quotation marks and citations omitted);\nsee also Hairu Chen v. Los Angeles Truck Centers, LLC, No.\nS240245, 2019 WL 3281346, at *3 (Cal. July 22, 2019).\n\n In making its choice-of-law determinations, the district\ncourt relied heavily on the California Supreme Court’s\ndecision in Sullivan v. Oracle Corp., 254 P.3d 237 (Cal.\n2011), and the parties do not dispute that Sullivan provides\nthe most helpful guidance for the choice-of-law questions\nbefore us. In Sullivan, the California Supreme Court\nanswered a certified question from this court regarding\nwhether California overtime law applied to non-resident\nemployees of a California corporation who worked primarily\nin their home states of Colorado and Arizona, but also\nworked in California (and several other states) for “entire\ndays or weeks” at a time. Id. at 239, 243. Sullivan first\nconcluded that as a matter of statutory construction,\nCalifornia law applied to all work performed for days or\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 21\n\nweeks at a time within the state’s borders, regardless of\nwhether it was performed by residents or non-residents. Id.\nat 241–43. Next, Sullivan undertook California’s three-step\ngovernmental interest analysis for choice-of-law questions.\nId. at 244–47. At the first step of the analysis—whether the\nrelevant laws differed—the court noted that California’s\novertime law “clearly” differed from the laws of the\nplaintiffs’ home states. Id. at 245.\n\n At the second step—whether a “true” conflict existed—\nthe court held that the existence of a true conflict was\n“doubtful, at best.” Id. The court explained that the second\nstep involves examining “each jurisdiction’s interest in the\napplication of its own law under the circumstances of the\nparticular case,” noting that a court “may make [its] own\ndetermination of the relevant policies and interests, without\ntaking ‘evidence’ as such on the matter.” Id. (internal\nquotation marks, alterations, and citations omitted). Sullivan\nobserved that “California has, and has unambiguously\nasserted, a strong interest in applying its overtime law to all\nnonexempt workers, and all work performed, within its\nborders.” Id. at 245. The court concluded that “neither\nColorado nor Arizona has a legitimate interest in shielding\nOracle from the requirements of California wage law as to\nwork performed here.” Id. at 246.\n\n In so holding, the court rejected two specific arguments\nadvanced by Oracle. First, Oracle contended that because\nArizona and Colorado have workers’ compensation statutes\nwith express extraterritorial application, those statutes\nindicate an interest in extending the protection of their\nemployment laws to their residents working outside the\nstate. Id. Not so, Sullivan held. While “a state has such an\ninterest, at least in the abstract, when the traveling, resident\nemployee of a domestic employer would otherwise be left\n\n22 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nwithout the protection of another state’s law,” the states had\n“expressed no interest in disabling their residents from\nreceiving the full protection of California overtime law when\nworking here, or in requiring their residents to work side-by-\nside with California residents in California for lower pay.”\nId.\n\n Second, Oracle argued that Arizona and Colorado “have\nan interest in providing hospitable regulatory environments\nfor their own businesses” and thus “also have an interest in\nshielding their own businesses from more costly and\nburdensome regulatory environments in other states.” Id.\nRelying on principles of federalism, Sullivan dismissed this\nargument. While “a state can properly choose to create a\nbusiness-friendly environment within its own boundaries,”\nthe federal Constitution does not require a state to substitute\n“‘the conflicting statute of another state’” for its own laws\nthat are “‘applicable to persons and events’” within that\nstate. Id. (quoting Phillips Petrol. Co. v. Shutts, 472 U.S.\n797, 822 (1985)). Nor does the Constitution “permit one\nstate to project its regulatory regime into the jurisdiction of\nanother state.” Id. (citing Healy v. Beer Insti., Inc., 491 U.S.\n324, 336–37 (1989)).\n\n Finally, although Sullivan held that there was almost\ncertainly no true conflict because neither Arizona nor\nColorado had a “legitimate interest” in blocking the\napplication of California law to the work performed in\nCalifornia, the court nonetheless proceeded to the third step\nof the analysis “for the sake of argument.” Id. at 246–47.\nSullivan concluded that the analysis at the third step—\ndetermining which state’s interest would be more impaired\nif its policy were subordinated to the policy of the other\nstate—yielded a straightforward answer:\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 23\n\n [T]o subordinate California’s interests to\n those of Colorado and Arizona\n unquestionably would bring about the greater\n impairment. To permit nonresidents to work\n in California without the protection of our\n overtime law would completely sacrifice, as\n to those employees, the state’s important\n public policy goals of protecting health and\n safety and preventing the evils associated\n with overwork. Not to apply California law\n would also encourage employers to substitute\n lower paid temporary employees from other\n states for California employees, thus\n threatening California’s legitimate interest in\n expanding the job market. By way of\n comparison, not to apply the overtime laws of\n Colorado and Arizona would impact those\n states’ interests negligibly, or not at all . . .\n Alternatively, viewing Colorado’s and\n Arizona's overtime regimens as expressions\n of a general interest in providing hospitable\n regulatory environments to businesses within\n their own boundaries, that interest is not\n perceptibly impaired by requiring a\n California employer to comply with\n California overtime law for work performed\n here.\n\nId. at 247 (citations omitted) (emphasis added).\n\n A.\n\n We first conclude that the district court did not err in\nholding that under Sullivan, California law should apply to\n\n24 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nthe (b)(3) California class. 6 Although defendants correctly\npoint out that Sullivan is not precisely analogous to the case\nat hand, the two principal differences on which defendants\nrely are unpersuasive. Specifically, defendants first rely on\nthe fact that while Sullivan involved a California\ncorporation, “most of the MLB Club Defendants with\naffiliates in the California League are located outside\nCalifornia.” But a close reading of Sullivan indicates that\nCalifornia law should apply to the California class, even\nthough many of the employers are not headquartered in\nCalifornia. For example, Sullivan expressly contemplated\nthat California’s overtime laws may not apply to non-\nresident employees of an out-of-state business who enter\nCalifornia only “temporarily during the course of a\nworkday,” but contrasted such a scenario with employees\nwho work in California for “entire days and weeks,” who are\ncovered by California law. Id. at 243 (emphases, internal\nquotation marks, and citations omitted).\n\n Similarly, Sullivan specifically left open the possibility\nthat other California employment laws, such as pay stub\nrequirements, may not apply to non-resident employees of\nout-of-state employers—with the clear implication that\novertime laws would apply to such employees. See id. at\n243–44. Likewise, far from limiting its holding only to non-\nresident employees of in-state employers, Sullivan merely\nemphasized that employees of in-state employers would\nespecially be covered by California law. See id. at 243.\n\n 6\n Contrary to the dissent’s criticism, Dissent at 74, we do not\nshortcut the governmental interest analysis. As we explain in the text,\nwe believe that Sullivan mandates application of California law to the\nCalifornia class. Rather than repeating Sullivan’s choice of law analysis,\nwe focus on several additional considerations that further support our\ndecision to affirm the district court’s reliance on Sullivan.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 25\n\n Second, defendants characterize Sullivan as resting on\nthe court’s determination that “neither Arizona nor Colorado\n. . . has asserted an interest in regulating overtime work\nperformed in other states.” Defendants argue that here, by\ncontrast, “numerous” states have a competing interest in\nregulating work performed in California. But defendants\nmisread Sullivan by erroneously presuming that its\nconclusion at the third step—that subordinating\n“California’s interests to those of Colorado and Arizona\nunquestionably would bring about the greater\nimpairment”—hinged entirely on whether Arizona or\nColorado law had asserted an interest in extraterritorial\napplication of their wage laws. Id. at 247. It is certainly\naccurate to say that Sullivan’s holding was influenced by the\nfact that neither Arizona nor Colorado law purported to\napply extraterritorially. Yet the court’s discussion at step\nthree cannot fairly be read to support the argument that\nCalifornia’s “strong interest in applying its overtime law to\n. . . all work performed within its borders,” id. at 245, would\nsuddenly become the lesser-impaired interest in the event\nanother state expressed a clear interest in applying its wage\nlaws to work performed in California. Rather, Sullivan\nstrongly indicates that California’s interest in applying its\nlaws to work performed within its borders for days or weeks\nat a time would reign supreme regardless of whether another\nstate expressed an interest in applying its own wage laws\ninstead of California’s.\n\n26 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n Although we read Sullivan as clearly mandating the\napplication of California law to the California class, two\nadditional considerations support our conclusion today. 7\n\n First, because the district court found that plaintiffs had\nmet their burden of showing that California law could\nconstitutionally be applied—a determination defendants do\nnot contest on appeal—the burden shifted to defendants “to\ndemonstrate ‘that foreign law, rather than California law,\nshould apply to class claims.’” Mazza, 666 F.3d at 590\n(quoting Wash. Mut. Bank v. Superior Court, 15 P.3d 1071,\n1081 (Cal. 2001)). The district court held that defendants\nfailed to meet this burden, because they had “not gone\nbeyond speculating in a general manner that the claims of\nsome members of the putative California Class might be\nsubject to the law of another state and that the interests of\nanother state might be more impaired by application of\nCalifornia law.”\n\n Defendants specifically point to one of the named\nplaintiffs—Mitch Hilligoss—as an example of the alleged\n“need to conduct choice of law inquiries as to every member\nof the California class.” The district court found this\nexample unpersuasive for several reasons, and we agree.\nThe defendants argued that Illinois law should apply to\nHilligoss’ work in California because the time he spent in\nCalifornia was a small proportion of his overall career\n(around two months out of a six-year career). The district\ncourt, however, correctly read Sullivan as indicating that\nCalifornia law should nonetheless apply to Hilligoss’\nCalifornia work. Indeed, the proportion of time the non-\n\n 7\n Moreover, as the dissent acknowledges, the California Supreme\nCourt has expressed a strong interest in regulating wage and hour claims\nwithin its borders. Dissent at 77.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 27\n\nresident employees in Sullivan worked in California was\nquite small (and in one case, even less than the proportion of\nHilligoss’ career spent in California): during the relevant\nthree-year period, one worked 20 days, another 74 days, and\nthe third 110 days. 254 P.3d at 239. Put differently, the\nemployees in Sullivan worked in California approximately\n1.8%, 6.7%, and 10% of the time, respectively. Id. What\nmattered in Sullivan—and what matters here—is that when\nthe employees worked in California, they did so for “entire\ndays or weeks” at a time. Id. at 243.\n\n Second, practical considerations strongly support\napplying California law to work performed in California, at\nleast as a general rule; to hold otherwise “would lead to\nbizarre and untenable results.” See Brief for Professors Peter\nHay and Patrick J. Borchers, Dkt. No. 21, at 12–13\n(hereinafter “Professors’ Amicus Brief”). If the law of the\nstate in which work is performed is not the law that generally\napplies, employers and employees alike would be subjected\nto an unworkable scheme. Employers would be required to\nproperly ascertain the residency status—itself not\nnecessarily an easy task, as any student or seasonal worker\ncould attest—of each of its employees. For every non-\nresident employee, employers would then have to determine\nwhether the wage laws of that employee’s state of residence\napply extraterritorially, and then come up with different rules\nfor each of its employees according to their state of residence\nand any extraterritorial application of their home state’s\nlaws. This would mean that at a single worksite, employees\nworking side-by-side in the same position would not only be\nowed vastly different minimum wages, but also that an\nemployer would need to set different rules for meal and rest\nbreaks for different employees, and so on and so forth. It\ncannot be in any state’s legitimate “interest” to foist such an\n\n28 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nadministrative nightmare upon both employers and\nemployees.\n\n Such a scenario would also result in an enormous\ncompetitive advantage—or disadvantage—for prospective\nemployees based solely on their state of residency.\nEmployers would be incentivized to hire residents of states\nwith low minimum wages and otherwise employer-friendly\nwage laws, while residents of states with higher minimum\nwages and more protective employment laws would\nsuddenly be far less appealing. Amici Professors Hays and\nBorchers persuasively point out that as defendants would\nhave it, a college student still domiciled in Seattle while\nattending a Nebraska university would have to be paid $15\nper hour at a part-time job in Nebraska, “nearly double\nNebraska’s minimum wage of $8 per hour.” Professors’\nAmicus Brief at 13. This, of course, would put the student\nat a crushing disadvantage; what rational employer would\nhire her?\n\n Moreover, given the administrative cost involved in\nattempting to comply with a patchwork of multiple states’\nwage laws at a single workplace, some employers might\ninstead choose to stick to hiring only resident employees, or\nperhaps only non-resident employees from a particular state\n(presumably one with a low minimum wage and minimally\nprotective employment laws). 8\n\n\n\n 8\n The California class consisted of those players who participated in\nthe California League, which plays games exclusively within California\nduring the championship season. The Arizona and Florida classes\nconsisted of those who performed during spring training, extended\nspring training, and the instructional leagues in those states. Thus, the\ndissent’s fear that employers will be required to research applicable state\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 29\n\n We do not foreclose the possibility that there could be\nsome circumstances in which a proper application of\nCalifornia’s choice-of-law rules might lead to the\napplication of another state’s wage and hour laws to work\nperformed in California. Nor do we create a per se rule or\nan unrebuttable presumption. We hold only that, given the\nabove considerations, we are more than satisfied that the\ndistrict court did not err in concluding that under Sullivan,\nCalifornia law applies to the California class.\n\n B.\n\n We next address whether the district court erred in\ndetermining that choice-of-law considerations defeated\npredominance and adequacy for the proposed Arizona and\nFlorida Rule 23(b)(3) classes, and conclude that the district\ncourt’s determination must be reversed. Our conclusion is\nanimated in part by several of the considerations outlined\nabove, which apply with equal force to the Arizona and\nFlorida classes. Moreover, the aforementioned enormous\npractical implications of a contrary holding would be just as\nproblematic and unworkable in Arizona and Florida as in\nCalifornia.\n\n 1.\n\n With those considerations in mind, we apply California’s\nthree-step governmental interest analysis, and conclude that\nArizona law should apply to the work performed in Arizona,\nand Florida law to the work performed in Florida. At the\nfirst step, we agree with defendants that the differences in\nstate law are “material,” meaning that “they make a\n\nlaws whenever an employee crosses state lines is overstated. Dissent at\n84.\n\n30 SENNE V. KANSAS CITY ROYALS BASEBALL\n\ndifference in this litigation.” Mazza, 666 F.3d at 590. For\nexample, some states have more expansive definitions of\n“work,” others have differing available defenses, and we\nhave previously held that the elements for a quantum meruit\nclaim—alleged in both the Arizona and Florida classes—\n“vary materially from state to state.” Id. at 591 (citing\nCandace S. Kovacic, A Proposal to Simplify Quantum\nMeruit Litigation, 35 Am. U.L. Rev. 547, 558–60 (1986)).\n\n 2.\n\n “Because the relevant laws differ,” we must “next\nexamine each jurisdiction’s interest in the application of its\nown law under the circumstances of the particular case to\ndetermine whether a true conflict exists.” Sullivan, 254 P.3d\nat 245 (alteration, internal quotation marks, and citation\nomitted). We are not persuaded, as defendants contend, that\na “true” conflict exists.\n\n First, under California’s choice-of-law principles, “a\njurisdiction ordinarily has the predominant interest in\nregulating conduct that occurs within its borders.” 9 Mazza,\n666 F.3d at 592 (internal quotation marks omitted) (quoting\nMcCann v. Foster Wheeler LLC, 225 P.3d 516, 534 (Cal.\n2010)). The dissent contends that “California has long\nrejected” this approach. Dissent at 71. In noting these\nprinciples, we do not ignore the evolution of California’s\nchoice of law doctrine. We recognize that the California\nSupreme Court “renounced the prior rule, adhered to by\ncourts for many years, that in tort actions the law of the place\n\n 9\n Wage and hour laws are typically categorized as “conduct-\nregulating,” as opposed to “loss-allocating.” See Professors’ Amicus\nBrief at 15–16 (citing Hay, Borchers & Symeonides, Conflict of Laws\n874–78 (5th ed. 2010)).\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 31\n\nof the wrong was the applicable law in a California forum\nregardless of the issues before the court” when it adopted the\ngovernmental interest approach. Hurtado v. Superior Court,\n522 P.2d 666 (Cal. 1974). Yet the California Supreme Court\nhas acknowledged that while it “no longer follows the old\nchoice-of-law rule that generally called for application of the\nlaw of the jurisdiction in which a defendant’s allegedly\ntortious conduct occurred without regard to the nature of the\nissue that was before the court . . . California choice-of-law\ncases nonetheless continue to recognize that a jurisdiction\nordinarily has the predominant interest in regulating conduct\nthat occurs within its borders.” McCann 225 P.3d at 534\n(internal quotation marks and citation omitted) (emphasis in\noriginal).\n\n Thus, when conducting the governmental interest\nanalysis, we must also recognize that a state ordinarily has\nthe predominant interest in regulating conduct within its\nborders. We draw this conclusion not from California’s\ninterest in regulating conduct within its own borders, but\nfrom California’s choice-of-law principles. 10 Thus these\n\n 10\n See e.g., McCann, 225 P.3d at 534, 537 (recognizing that although\nCalifornia no longer uniformly applied the law of the jurisdiction in\nwhich the allegedly tortious conduct occurred, Oklahoma’s interests\n“would be more impaired if its law were not applied” as the plaintiff’s\nexposure to asbestos occurred in Oklahoma); Reich v. Purcell, 67 432\nP.2d 727, 730 (Cal. 1967) (“Missouri is concerned with conduct within\nher borders and as to such conduct she has the predominant interest of\nthe states involved.”); Castro v. Budget Rent-A-Car Sys., Inc., 65 Cal.\nRptr. 3d 430, 442 (Cal. Ct. App. 2007) (“The accident and Castro's injury\noccurred within Alabama's borders, thus giving Alabama a presumptive\ninterest in controlling the conduct of those persons who use its roadways,\nabsent some other compelling interest to be served by applying\nCalifornia law.”); Hernandez v. Burger, 162 Cal. Rptr. 564, 568 (Cal. Ct.\nApp. 1980) (“It is true that the place of the wrong is no longer treated as\na controlling factor where application of the law of another jurisdiction\n\n32 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nprinciples are not limited to the California class but apply to\nthe Florida and Arizona classes as well. See Mazza, 666 F.3d\nat 593 (“The district court did not adequately recognize that\neach foreign state [not just California] has an interest in\napplying its law to transactions within its borders.”). The\ndistrict court erred in ignoring these principles as a starting\npoint, instead faulting plaintiffs for not addressing “in detail\nthe interests of either Arizona or Florida in applying their\nlaw” and focusing on the absence of Florida or Arizona cases\nakin to Sullivan—despite the strong indications that Arizona\nand Florida have the “predominant interest” in applying their\nlaws to work performed within their state. See Mazza, 666\nF.3d at 592.\n\n Second, Sullivan relied on several different\nconsiderations to arrive at its conclusion that the existence\nof a true conflict was “doubtful, at best”: (1) the states in\nwhich the employees resided did not express an intent to\napply their laws extraterritorially; (2) the employees’ states\nof residence did not have a “legitimate interest” in shielding\nan employer from California’s wage laws as to work\nperformed in California; and (3) federalism and due process\nmade extraterritorial reach doubtful under the\ncircumstances. See 254 P.3d at 245–47. Although\ndefendants vigorously argue that the first of those rationales\nis inapplicable here—as discussed in greater detail below—\nat a minimum, the second and third rationales do apply, and\nweigh against the existence of a true conflict.\n\n\nhaving a connection with the accident will serve a legitimate interest or\npolicy of the other jurisdiction. However, the situs of the injury remains\na relevant consideration.”); Cable v. Sahara Tahoe Corp., 155 Cal. Rptr.\n770, 777 (Cal. Ct. App. 1979) (“The state with the ‘predominant’ interest\nin controlling conduct normally is the state in which such conduct occurs\nand is most likely to cause injury.”).\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 33\n\n As to the first rationale, both defendants and the dissent\ncontend that several states have expressed an interest in\napplying their wage and hour laws to work performed\noutside the state. In support of their position, they cite to a\nhandful of cases where courts (largely district courts or\nintermediate state courts, with the exceptions of West\nVirginia and Washington) 11 have applied one state’s wage\nlaws to work performed at least partially in another state.\nFor several reasons, we are unpersuaded by defendants’\narguments. For one, we read Sullivan as indicating that\nunder California’s choice-of-law principles, a state has a\nlegitimate interest in applying its wage laws extraterritorially\nonly in two limited circumstances, neither of which apply\nhere: one, when a state’s resident employee of that state’s\nresident employer leaves the state “temporarily during the\ncourse of the normal workday,” and two, “when the\ntraveling, resident employee of a domestic employer would\notherwise be left without the protection of another state’s\n\n\n 11\n In New v. Tac & C Energy, Inc., 355 S.E.2d 629 (W. Va. 1987),\nthe West Virginia Supreme Court of Appeals applied its own conflict-\nof-laws principles—relying on the Restatement (Second) of Conflicts\n§ 196—to conclude that while there was a presumption that the law of\nthe state where services were rendered applies, the presumption could be\novercome by showing that another state had a “more significant\nrelationship to the transaction and the parties.” Id. at 631. Where all\nparties were residents of West Virginia, the employment contract was\nmade and partially performed in West Virginia, and the plaintiffs were\nonly in Kentucky for the duration of the work, the court concluded that\nthe presumption was overcome and that West Virginia “had the more\nsignificant connection to the employment relationship.” Id. California’s\nchoice-of-law test, of course, does not utilize the “more significant\nrelationship” test for choice-of-law questions in the wage and hour\ncontext. See Sullivan, 254 P.3d at 244. New is therefore unpersuasive\nhere. We discuss the Washington Supreme Court case below.\n\n34 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nlaw.” Id. at 242, 246 (citations, internal quotation marks,\nand alterations omitted).\n\n Moreover, the cases on which defendants and the dissent\nrely are, in large part, both factually and procedurally\ninapposite to the circumstances of this case. 12 For example,\ndefendants rely heavily on Bostain v. Food Exp., Inc., 153\nP.3d 846, 851 (Wash. 2007) to argue that Washington has an\ninterest in applying its wage laws extraterritorially. As the\nCalifornia Supreme Court held in Sullivan, however, Bostain\n“says nothing about a case such as this”—that is, a case\nwhich (1) involves work performed entirely in one state, and\n(2) presents an unavoidable conflict-of-laws issue. 254 P.3d\nat 243. In Bostain, by contrast, either Washington law\napplied to the work performed in both Washington and other\nstates, or else no state’s law applied. Id. at 243, 246.\nSignificantly, Bostain interpreted an overtime statute that\nspecifically delineated the circumstances under which its\nprovisions would apply to interstate truck drivers; as the\nWashington Supreme Court noted, interstate truck drivers by\ndefinition perform some of their work out of state. 153 P.3d\nat 848–51. The statute at issue in Bostain did “not limit the\nrequirement for overtime pay to hours worked” within the\nstate’s borders. Id. at 851. Similarly, here, defendants point\n\n 12\n Defendants’ repeated citation to Gonyou v. Tri-Wire Eng'g Sols.,\nInc., 717 F. Supp. 2d 152 (D. Mass. 2010) is illustrative. In Gonyou, a\nMassachusetts resident employee of a Massachusetts employer worked\nlargely, although not entirely, in Connecticut. Id. at 153–54. The\ndefendant filed a motion to dismiss on the ground that the Massachusetts\novertime statute did not apply to work performed in Connecticut. Id. at\n154–55. The court denied the motion but emphasized the limited nature\nof its ruling: “As is eminently clear, this is a motion to dismiss and this\nruling is strictly limited to the facts and circumstances of this case and\nthis motion.” Id. at 155.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 35\n\nto no state statutes potentially applicable to the Arizona and\nFlorida class members that limit their application to work\nperformed within the state.\n\n 3.\n\n Although the existence of a “true” conflict is\nquestionable, we need not decide whether a true conflict\nexists, as the third step of California’s governmental interest\ntest yields a clear answer: the laws of Arizona and Florida\nshould apply to the work performed wholly within their\nrespective boundaries. 13 See Sullivan, 254 F.3d at 247. As\nthe California Supreme Court has explained the step three\ninquiry:\n\n [T]he court does not “weigh” the conflicting\n governmental interests in the sense of\n determining which conflicting law\n manifested the “better” or the “worthier”\n\n 13\n Furthermore, in many of the cases cited by the dissent to\ndemonstrate that some states have asserted an interest in applying their\nwage and hour laws outside of their borders, courts have looked closely\nat where the relevant work is performed. See e.g., Pierre v. Gts\nHoldings, Inc., No. 15 CIV. 143 (PAC), 2015 WL 7736552, at *3–*4\n(S.D.N.Y. Nov. 30, 2015) (concluding that New York labor laws apply\nbecause, among other things, the majority of the plaintiff’s chauffeured\nrides were conducted in New York); Baxi v. Ennis Knupp & Assocs.,\nInc., No. 10-CV-6346, 2011 WL 3898034, at *14 (N.D. Ill. Sept. 2,\n2011) (denying a motion to dismiss Illinois labor law claims because the\nplaintiff, a foreign resident, performed some work in Illinois); Friedrich\nv. U.S. Computer Sys., Inc., No. CIV. A. 90-1615, 1996 WL 32888, at\n*8 (E.D. Pa. Jan. 22, 1996) (concluding that a Pennsylvania labor law\napplies to the plaintiffs because the jury found the plaintiffs were “based\nin Pennsylvania,” even if they were not residents of the state); Dow, 989\nN.E.2d at 914 (concluding that Massachusetts law applied because,\ngiven the nature of the plaintiff’s work, the work “sensibly may be\nviewed as having ‘occurred’ in Massachusetts”).\n\n36 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n social policy on the specific issue. An\n attempted balancing of conflicting state\n policies in that sense is difficult to justify in\n the context of a federal system in which,\n within constitutional limits, states are\n empowered to mold their policies as they\n wish. Instead, the process can accurately be\n described as a problem of allocating domains\n of law-making power in multi-state\n contexts—by determining the appropriate\n limitations on the reach of state policies—as\n distinguished from evaluating the wisdom of\n those policies. Emphasis is placed on the\n appropriate scope of conflicting state policies\n rather than on the “quality” of those policies.\n\nMcCann, 225 P.3d at 533–34 (alterations and citations\nomitted) (emphasis added).\n\n As discussed above, in Mazza, we faithfully applied the\nprinciple under California’s choice-of-law jurisprudence\nthat “a jurisdiction ordinarily has the predominant interest in\nregulating conduct that occurs within its borders.” Mazza,\n666 F.3d at 592 (internal quotation marks omitted) (quoting\nMcCann, 225 P.3d at 534). We thus had no trouble\nconcluding at step three that “each class member’s consumer\nprotection claim should be governed by the consumer\nprotection laws of the jurisdiction in which the transaction\ntook place.” Id. at 594. Notably, we reached this conclusion\nwithout specifically inquiring into the interests potentially\nexpressed by any state’s statutory language or case law.\nRather, our conclusion was dictated by the principle,\ndiscussed above, that a jurisdiction ordinarily has the\npredominant interest in regulating conduct within its own\nborders. Id. at 591–92 (first citing State Farm Mut. Auto.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 37\n\nIns. Co. v. Campbell, 538 U.S. 408, 422 (2003); and then\nciting McCann, 225 P.3d at 534).\n\n Moreover, in Sullivan, the court concluded that to\nsubordinate California’s ability to apply its own wage laws\nto work performed within the state would “unquestionably”\ncause greater impairment to California than to the states that\nmight seek to apply their wage laws to work performed by\ntheir residents within California. 254 P.3d at 247. As\ndescribed previously, while this holding was influenced by\nthe absence of an expression of interest by Arizona or\nColorado in applying their laws extraterritorially, it did not\nrise or fall on that ground. See id. at 244–47. And although\ndefendants point to a handful of cases that have entertained\nthe potential application of one state’s wage laws to work\nperformed in another state, they have not pointed to a single\nstate with a potentially-applicable statute that expresses a\nclear interest in applying to work performed wholly outside\nthe state.\n\n But even if defendants were able to identify any states\nthat had unambiguously expressed an interest in applying\ntheir wage laws to work performed entirely in another state,\nSullivan strongly militates against concluding that such an\nexpression of interest would be adequate to overcome the\nprinciple that the state in which the conduct at issue occurs\nhas the “predominant interest” in applying their own law.\nSee Mazza, 666 F.3d at 592–94; Sullivan, 254 P.3d at 245–\n47. Forcing Arizona or Florida to allow the application of\nother states’ wage laws in this case would be just as\ndestructive to the balance Arizona and Florida have struck\nbetween protecting workers and fostering a hospitable\nbusiness environment within their states as allowing the\napplication of Colorado or Arizona law in Sullivan would\nhave been to the balance California struck between those\n\n38 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nsame interests. See Sullivan, 254 P.3d at 246–47. The\ndistrict court fundamentally misunderstood the proper\napplication of California’s choice-of-law principles—which,\nwhen correctly applied, indicate that Arizona law should\ngovern the Arizona class, and Florida law the Florida class.\n\n C.\n\n We next address whether the district court erred in\nrefusing to certify a Rule 23(b)(2) class for unpaid work at\ndefendants’ training facilities in Arizona and Florida on the\nsole basis that choice-of-law issues undermined\n“cohesiveness” and therefore made injunctive and\ndeclaratory relief inappropriate. Because the district court’s\nerrors in its choice-of-law analysis relating to the proposed\nArizona and Florida Rule 23(b)(3) classes apply equally to\nits refusal to certify the proposed Rule 23(b)(2) class, we\nalso reverse the denial of the (b)(2) class.\n\n We further hold that the district court erred in imposing\na “cohesiveness” requirement for the proposed Rule 23(b)(2)\nclass. Although we have never explicitly addressed whether\n“cohesiveness” is required under Rule 23(b)(2), courts that\nhave imposed such a test treat it similarly to Rule 23(b)(3)’s\npredominance inquiry14—something we have previously\nrejected in no uncertain terms. See Walters v. Reno, 145 F.3d\n1032, 1047 (9th Cir. 1998) (“[W]ith respect to 23(b)(2) in\nparticular, the government’s dogged focus on the factual\ndifferences among the class members appears to\n\n 14\n The similarity between “cohesiveness” and predominance is\nperhaps unsurprising, given that the Supreme Court described the\npredominance inquiry under 23(b)(3) as testing whether a class is\n“sufficiently cohesive to warrant adjudication by representation.”\nAmchem Products, Inc. v. Windsor, 521 U.S. 591, 623 (1997) (emphasis\nadded).\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 39\n\ndemonstrate a fundamental misunderstanding of the rule.\nAlthough common issues must predominate for class\ncertification under Rule 23(b)(3), no such requirement exists\nunder 23(b)(2).”); see also 2 Newberg on Class Actions\n§ 4:34 (5th ed. 2012) (describing similarity between\npredominance under Rule 23(b)(3) and “cohesiveness”\nunder Rule 23(b)(2) in courts that have adopted it). We\ntherefore remand for the district court to consider anew\nwhether to certify the proposed Rule 23(b)(2) class. 15\n\n III.\n\n Having addressed the impact of choice-of-law questions,\nwe turn to the issue next up at bat: whether the district court\nerred in concluding that plaintiffs could meet the\npredominance requirement for the proposed California,\nFlorida, and Arizona (b)(3) classes through a combination of\nrepresentative evidence and application of the “continuous\nworkday” rule.\n\n Rule 23(b)(3)’s predominance requirement requires\ncourts to ask “whether the common, aggregation-enabling\nissues in the case are more prevalent or important than the\nnon-common, aggregation-defeating, individual issues.”\nTyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1045\n(2016) (quoting 2 Newberg on Class Actions § 4:49 (5th ed.\n2012)). A proposed (b)(3) class may be certified as long as\n“one or more of the central issues in the action are common\nto the class and can be said to predominate . . . even though\nother important matters will have to be tried separately, such\n 15\n While the parties advanced numerous arguments regarding (b)(2)\ncertification in the district court, and advance similar arguments—along\nwith a few new ones—before us, we decline to pass on those other issues\nin the first instance. See Stockwell, 749 F.3d at 1113, 1116–17; Davis v.\nNordstrom, Inc., 755 F.3d 1089, 1094–95 (9th Cir. 2014).\n\n40 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nas damages or some affirmative defenses peculiar to some\nindividual class members.” Id. (quoting 7AA C. Wright, A.\nMiller, & M. Kane, Federal Practice and Procedure § 1778\n(3d ed. 2005)).\n\n “[P]redominance in employment cases is rarely defeated\non the grounds of differences among employees so long as\nliability arises from a common practice or policy of an\nemployer.” 7 Newberg on Class Actions § 23:33 (5th ed.\n2012). Although the existence of blanket corporate policies\nis not a guarantee that predominance will be satisfied, such\npolicies “often bear heavily on questions of predominance\nand superiority.” In re Wells Fargo Home Mortg. Overtime\nPay Litig., 571 F.3d 953, 958 (9th Cir. 2009).\n\n Whether the district court was correct in concluding that\nplaintiffs had satisfied the predominance requirement hinges\non the application of two longstanding wage-and-hour\ndoctrines to this case: first, the burden-shifting framework\ninitially set forth in the Supreme Court’s seminal decision in\nAnderson v. Mt. Clemens, 328 U.S. 680 (1946) and recently\nexpanded upon in Tyson Foods v. Bouaphakeo, 136 S. Ct.\n1036 (2016); and second, the so-called “continuous\nworkday” rule. We address each of these doctrines and their\napplication to this case in turn.\n\n A.\n\n In Mt. Clemens, the Supreme Court acknowledged the\ndifficult bind that employees frequently confronted when\nseeking to bring wage-and-hour claims against their\nemployers: if their employers had failed to maintain proper\ntimekeeping records, proving the hours of uncompensated\nwork often posed “an impossible hurdle for the employee.”\n328 U.S. at 687. Mt. Clemens held that such a catch-22 was\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 41\n\nnot in line with “the remedial nature of [the FLSA] 16 and the\ngreat public policy which it embodies.” Id. After all, “[s]uch\na result would place a premium on an employer’s failure to\nkeep proper records in conformity with his statutory duty; it\nwould allow the employer to keep the benefits of an\nemployee’s labors without paying due compensation.” Id.\n\n To address this problem, Mt. Clemens established its\nlandmark burden-shifting framework for actions in which\nthe employer has kept inaccurate or inadequate records: if an\nemployee “proves that he has in fact performed work for\nwhich he was improperly compensated” and “produces\nsufficient evidence to show the amount and extent of that\nwork as a matter of just and reasonable inference,” then the\nburden “shifts to the employer to come forward with\nevidence of the precise amount of work performed or with\nevidence to negative the reasonableness of the inference to\nbe drawn from the employee’s evidence.” Id. at 687–88. If\nthe employer does not rebut the employee’s evidence,\ndamages may then be awarded to the employee, “even\nthough the result be only approximate.” Id. at 688.\n\n Mt. Clemens explicitly rejected the notion that allowing\napproximate damages in such situations would be unfair due\nto its speculative and imprecise nature or because employers\nsometimes make good-faith mistakes over what constitutes\ncompensable “work”:\n\n\n\n 16\n Although Mt. Clemens was decided under the FLSA, its holding\nhas been consistently applied in the context of state wage-and-hour\nclaims as well. See, e.g., Tyson, 136 S. Ct. at 1045–48; Torres v. Mercer\nCanyons Inc., 835 F.3d 1125, 1140 (9th Cir. 2016); Hernandez v.\nMendoza, 245 Cal. Rptr. 36, 39–40 (Cal. Ct. App. 1988) (applying Mt.\nClemens to claims under California wage and hour law).\n\n42 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n The employer cannot be heard to complain\n that the damages lack the exactness and\n precision of measurement that would be\n possible had he kept records in accordance\n with the [statutory] requirements . . . And\n even where the lack of accurate records\n grows out of a bona fide mistake as to\n whether certain activities or non-activities\n constitute work, the employer, having\n received the benefits of such work, cannot\n object to the payment for the work on the\n most accurate basis possible under the\n circumstances . . . In such a case it would be\n a perversion of fundamental principles of\n justice to deny all relief to the injured person,\n and thereby relieve the wrongdoer from\n making any amend for his acts.\n\nId. (internal quotation marks and citation omitted).\n\n Seventy years after Mt. Clemens addressed the use of\nrepresentative evidence at the trial stage to show damages,\nTyson extended Mt. Clemens’ holding to answer two\nimportant questions: whether representative evidence may\nbe used at the class certification stage, and whether\nrepresentative evidence may also be used to establish\nliability in addition to damages. In Tyson, employees who\nworked in more than 400 jobs across three departments at a\nmeat processing plant sued under the FLSA and an Iowa\nwage law, alleging that Tyson had not paid them overtime\nfor time they spent donning and doffing protective gear; the\nemployees also sought certification of a Rule 23 class and a\nFLSA collective action. 136 S. Ct. at 1041–42.\n\n The district court certified the class and collective\nactions, rejecting Tyson’s arguments that the claims were\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 43\n\ninappropriate for resolution on a classwide and collective\nbasis due to the dissimilarity in the types of protective gear\nworn and the variations in time spent donning and doffing\nthat gear. Id. at 1042–43. Because Tyson had not kept\nrecords of the donning and doffing time, plaintiffs relied on\nrepresentative evidence to demonstrate both liability17 and\ndamages: employee testimony, video recordings, and—most\nsignificantly—an expert study that computed an estimated\namount of time spent donning and doffing for each of the\nthree departments based on hundreds of video observations.\nId. at 1043. Although the expert estimated that the time\nspent donning and doffing was 18 minutes per day for two\nof the departments and 21.25 minutes for the other, id., the\nsurvey data showed a great deal of variation in how long it\ntook individual employees to don and doff. Id. at 1055\n(Thomas, J., dissenting). Specifically, the time spent\ndonning ranged from around thirty seconds to more than ten\nminutes, and the time doffing varied from under two minutes\nto over nine minutes. Id. After a jury verdict in the\nemployees’ favor (albeit one that awarded less than half of\nthe damages recommended by the employees’ expert based\non the survey data), Tyson moved to decertify the class and\nset aside the jury verdict, arguing that this variance made\nclass and collective certification inappropriate. Id. at 1044–\n\n\n\n 17\n Because the employees brought only overtime claims (as opposed\nto minimum wage or other wage claims), “each employee had to show\nhe or she worked more than 40 hours a week, inclusive of time spent\ndonning and doffing, in order to recover.” Tyson, 136 S. Ct. at 1043.\nThat the majority permitted the use of representative evidence to\nestablish “an otherwise uncertain element of liability”—i.e., whether\nclass members worked more than 40 hours per week—was one of the\nkey bases for Justice Thomas’s vigorous dissent. See id. at 1057–59\n(Thomas, J., dissenting).\n\n44 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n45. The district court denied the motion, and the Eighth\nCircuit affirmed. Id.\n\n Tyson sought certiorari on the grounds that using\nrepresentative evidence “manufactures predominance by\nassuming away the very differences that make the case\ninappropriate for classwide resolution,” “absolves each\nemployee of the responsibility to prove personal injury,” and\nstrips the employer of their ability to “litigate its defenses to\nindividual claims.” Id. at 1046. Rejecting these arguments,\nthe Supreme Court affirmed the class and collective\ncertifications. Id. at 1046–47. Because of Tyson’s\ndereliction of their recordkeeping duties, the employees\nwere entitled to “introduce a representative sample to fill an\nevidentiary gap created by the employer's failure to keep\nadequate records.” Id. at 1047. The Court held that if the\nrepresentative sample introduced were admissible and\n“could have sustained a reasonable jury finding as to hours\nworked in each employee's individual action, that sample is\na permissible means of establishing the employees' hours\nworked in a class action.” Id. at 1046–47.\n\n Stated another way, Tyson concluded that even where\n“reasonable minds may differ” about whether representative\nevidence is sufficiently probative of the requirements for\nliability for a particular cause of action—in Tyson, whether\nit was probative of the “time actually worked by each\nemployee”—that question is to be resolved by the jury, not\nat the class certification stage. Id. at 1049 (“The District\nCourt could have denied class certification on this ground\n[whether the representative evidence was “probative as to\nthe time actually worked by each employee”] only if it\nconcluded that no reasonable juror could have believed that\nthe employees spent roughly equal time donning and\ndoffing.”) (emphasis added). If the proffered representative\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 45\n\nevidence, however, were “statistically inadequate or based\non implausible assumptions,” it “could not lead to a fair or\naccurate estimate of the uncompensated hours an employee\nhas worked.” Id. at 1048–49. But where the evidence is\nadmissible—for expert evidence, using the Daubert\nstandard—then the “no reasonable juror” standard at the\nclass certification stage applies. See id. at 1049.\n\n B.\n\n Having established the parameters of when\nrepresentative evidence may be used at the class certification\nstage, we address the second significant wage-and-hour\ndoctrine relevant to this case: the “continuous workday”\nrule. The rule was first promulgated by the Department of\nLabor (DOL) consistent with the Supreme Court’s decisions\ninterpreting the FLSA prior to the enactment of the Portal-\nto-Portal Act 18 in 1947. IBP, Inc. v. Alvarez, 546 U.S. 21,\n27–28 (2005). It presumes that once the beginning of the\n\n 18\n In response to what Congress perceived as excessively expansive\njudicial interpretations of what constitutes compensable work under the\nFLSA, IBP, 546 U.S. at 27–28, it passed the Portal-to-Portal Act to\nexempt certain activities as compensable under FLSA:\n\n “(1) walking, riding, or traveling to and from the actual\n place of performance of the principal activity or\n activities which such employee is employed to\n perform, and\n\n (2) activities which are preliminary to or postliminary\n to said principal activity or activities,\n\nwhich occur either prior to the time on any particular workday at which\nsuch employee commences, or subsequent to the time on any particular\nworkday at which he ceases, such principal activity or activities.” 61\nStat. 86–87 (codified at 29 U.S.C. § 254(a)).\n\n46 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nworkday is triggered, an employee performs compensable\nwork throughout the rest of the day until the employee\ncompletes their last principal activity (or the last activity\nwhich is “integral and indispensable” to the employee’s\nprincipal activities)—whether or not the employee actually\nengages in work throughout that entire period. See id. at 28,\n32–37; see also Alvarez v. IBP, Inc., 339 F.3d 894, 907 (9th\nCir. 2003), aff’d, 546 U.S. 21 (2005) (holding that under the\ncontinuous workday rule, “work time [is] continuous, not the\nsum of discrete periods”).\n\n Of course, this rule raises inevitable questions: when\ndoes the workday begin, and when does it end? The DOL\ndefines the “workday” to generally mean “the period\nbetween the commencement and completion on the same\nworkday of an employee's principal activity or activities.”\n29 C.F.R. § 790.6(b). The Supreme Court expanded upon\nthis definition, interpreting “principal activity or activities”\nto also include “all activities which are an integral and\nindispensable part of the principal activities.” IBP, 546 U.S.\nat 29–30 (internal quotation marks and citation omitted).\nThus, any activity which is “integral and indispensable” to\nprincipal activities, even if performed outside of a scheduled\nshift, triggers the beginning of the “workday.” Id. at 31–37.\n“Among the activities included as an integral part of a\nprincipal activity are those closely related activities which\nare indispensable to its performance,” 29 C.F.R. § 790.8(c),\nsuch as knife-sharpening performed outside of a scheduled\nshift by butchers at a meatpacking plant. Mitchell v. King\nPacking Co., 350 U.S. 260, 261–63 (1956).\n\n C.\n\n With all of that in mind, we turn to how these two\ndoctrines impact this case, and more specifically, whether\nthe district court was correct in concluding that the\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 47\n\ncombination of Tyson and the continuous workday rule\nenabled plaintiffs to show that they meet Rule 23(b)(3)’s\npredominance requirement. Defendants contend that the\ndistrict court erred in holding that plaintiffs had\ndemonstrated predominance for two main reasons:\n(1) because the Main Survey asked only about arrival and\ndeparture times at the ballpark and not about what activities\nthe players actually performed while at the ballpark,\nplaintiffs cannot rely on the continuous workday theory\nbecause there is no way to determine the beginning or end of\nthe “workday,” and (2) the Main Survey revealed significant\nvariations in players’ arrival and departure times, even\namong players employed by the same MLB franchise.\n\n This task requires us to address the proposed Arizona\nand Florida classes separately from the California class. As\nan initial matter, however, we note that despite defendants’\nrepeated suggestions to the contrary, the representative\nevidence offered by plaintiffs was not limited to just the\nMain Survey, nor are observational studies the only type of\nevidence permitted to fill in evidentiary gaps under Tyson.\nWe reject defendants’ erroneous view of the record and their\ncramped reading of Tyson.\n\n 1.\n\n As to the Arizona and Florida classes, we easily affirm\nthe district court’s determination. Recall that these two\nclasses cover time spent participating in spring training,\nextended spring training, and the instructional leagues—\nperiods during which virtually all players are completely\nunpaid for their participation. 19 Moreover, these classes do\n\n 19\n Payroll data produced by defendants reveals that of the 21,211\nplayers who participated in spring training between the 2009 and 2015\n\n48 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nnot bring overtime claims, but rather allege minimum wage\nviolations. 20 Therefore—as the district court correctly\nheld—liability can be established simply by showing that the\nclass members performed any compensable work. 21 That is\neasily resolved on a classwide basis by answering two\nquestions: (1) are the players employees of defendants, and\n(2) do the minor league team activities during these periods\nconstitute compensable work under the laws of either\nArizona or Florida? We hold that these two “common,\naggregation-enabling issues in the case are more prevalent\n[and] important than the non-common, aggregation-\ndefeating, individual issues,” therefore making certification\nappropriate. Tyson, 136 S. Ct. at 1045 (quoting 2 Newberg\non Class Actions § 4:49 (5th ed. 2012)).\n\n Defendants do not seriously contest that their policy is to\ndeny players compensation during spring training, extended\nspring training, and the instructional leagues—nor could\nthey credibly do so, given that the MLB’s own mandatory\n\nseasons, only 11 were paid a salary. Put differently, a mere .005% of\nplayers received a salary during spring training, and those 11 players\nmay be identified through payroll records and appropriately excluded\nfrom the class. Likewise, a small number of MLB franchises pay players\nduring extended spring training, but these players are identifiable\nthrough payroll records and may either be excluded from the class or,\npotentially, placed into a subclass.\n 20\n The Arizona and Florida classes also bring quantum meruit\nclaims, and the Arizona class alleges recordkeeping violations, but the\nparties do not dispute that these claims are irrelevant to this portion of\nour predominance analysis.\n\n 21\n We also note that the Arizona class’s claims are bolstered by the\nfact that under Arizona law, failure to keep appropriate records of hours\nworked “raise[s] a rebuttable presumption that the employer did not pay\nthe required minimum wage rate.” Ariz. Rev. Stat. Ann. § 23-364.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 49\n\ncontract “obligates Player[s] to perform professional\nservices on a calendar year basis, regardless of the fact that\nsalary payments are to be made only during the actual\nchampionship playing season.” And as we have long held,\nsuch uniform corporate policies “carry great weight for\ncertification purposes.” In re Wells Fargo Home Mortg.\nOvertime Pay Litig., 571 F.3d at 958. This is not the “rare[]”\ncase where predominance is defeated despite the existence\nof an employer’s “common practice or policy.” 7 Newberg\non Class Actions § 23:33 (5th ed. 2012).\n\n We also agree with the district court that as to these\nclasses, many of defendants’ protests go to damages, not\nliability. Damages may well vary, and may require\nindividualized calculations. But “the rule is clear: the need\nfor individual damages calculations does not, alone, defeat\nclass certification.” Vaquero v. Ashley Furniture Indus.,\nInc., 824 F.3d 1150, 1155 (9th Cir. 2016); see Tyson, 136\nS. Ct. at 1045 (holding that where “one or more of the central\nissues in the action are common to the class and can be said\nto predominate,” certification may be appropriate “even\nthough other important matters will have to be tried\nseparately, such as damages.” (quoting 7AA C. Wright, A.\nMiller, & M. Kane, Federal Practice and Procedure § 1778\n(3d ed. 2005))).\n\n We do not, however, mean to minimize defendants’\ncriticisms of the Main Survey. Indeed, we agree that there\nare a number of legitimate questions about the\npersuasiveness of the Main Survey, especially if it were the\nonly representative evidence submitted in support of\ncertification. But as we have mentioned, the Main Survey\nwas but one piece of the plaintiffs’ representative\nevidence—evidence that also included hundreds of internal\n\n50 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nteam schedules and public game schedules, payroll data, and\nthe testimony of both players and league officials.\n\n At minimum, if the players are “employees” under either\nArizona or Florida law and defendants are unable to prove\nthat any affirmative defenses apply, the team schedules will\nserve to conclusively demonstrate that the players spent time\nworking for which they were uncompensated. See Ariz.\nRev. Stat. Ann. § 23-362; Ariz. Admin. Code. § R20-5-\n1202(19) (“‘Hours worked’ means all hours for which an\nemployee covered under the Act is employed and required\nto give to the employer, including all time during which an\nemployee is on duty or at a prescribed work place and all\ntime the employee is suffered or permitted to work.”);\n29 C.F.R. § 778.223 (“As a general rule the term ‘hours\nworked’ will include: (a) All time during which an employee\nis required to be on duty or to be on the employer's premises\nor at a prescribed workplace and (b) all time during which\nan employee is suffered or permitted to work whether or not\nhe is required to do so.”). 22 Moreover, if plaintiffs can\npersuade a jury that their workday began at a particular\ntime—either because they were required to report at that\ntime, 23 or because they arrived of their own volition but\n\n 22\n We rely on interpretations of the FLSA here because Florida’s\nconstitution provides that “case law, administrative interpretations, and\nother guiding standards developed under the federal FLSA shall guide\nthe construction of [the constitutional amendment providing for a\nminimum wage] and any implementing statutes or regulations.” Fla.\nConst. art. X, § 24.\n\n 23\n See 29 C.F.R. § 790.6 (“If an employee is required to report at the\nactual place of performance of his principal activity at a certain specific\ntime, his ‘workday’ commences at the time he reports there for work in\naccordance with the employer's requirement.”).\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 51\n\nengaged in work activities upon arriving (i.e., were\n“permitted” to work)—the continuous workday doctrine\neliminates the need for plaintiffs to prove which activities\nthey engaged in throughout the day. 24 See IBP, 546 U.S. at\n28, 32–37.\n\n Defendants should not “be heard to complain that the\ndamages lack the exactness and precision of measurement\nthat would be possible had [they] kept records in accordance\nwith the [statutory] requirements,” even if their “lack of\naccurate records grows out of a bona fide mistake as to\nwhether certain activities or non-activities constitute work.”\nMt. Clemens, 328 U.S. at 688. “Having received the benefits\nof such work, [defendants] cannot object to the payment for\nthe work on the most accurate basis possible under the\ncircumstances.” Id.\n\n 2.\n\n We next address whether the district court was correct to\nhold that predominance had been met for the California\nclass. Given the differences in the types of claims brought\nby the California class as compared to the Arizona and\nFlorida classes, certification of the California class is more\ncomplex and requires additional analysis. Unlike the\nArizona and Florida classes, the California class brought\nclaims relating to work performed during the championship\nseason—a time when the players do get paid, albeit not\nmuch. As a result, in order to prove liability on their\novertime claims, the California class must show that its\n 24\n A jury may also decide that for baseball players, activities like\nhitting practice with coaches and supervised weightlifting—much like\nknife-sharpening by butchers at a meatpacking plant— are “integral and\nindispensable” to the principal activity of playing baseball and therefore\ntrigger the start of the “workday.” See Mitchell, 350 U.S. at 261–63.\n\n52 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nmembers worked more than 8 hours in a day, more than 40\nhours in a week, and/or worked 7 days in a workweek. See\nCal. Labor Code § 510; Mendoza v. Nordstrom, Inc., 393\nP.3d 375, 381–82 (Cal. 2017). Likewise, to establish\nliability on their minimum wage claims, the California class\nmust demonstrate that they worked hours for which they\nwere not paid at least minimum wage—but whereas the\nArizona and Florida classes can demonstrate liability simply\nby showing they worked any hours, the California class’s\nburden is made more challenging by the fact that the players\nreceive some pay. See Cal. Labor Code §§ 1182 et seq;\nArmenta v. Osmose, Inc., 37 Cal. Rptr. 3d 460, 466–68 (Cal.\nCt. App. 2005). Nonetheless, a number of considerations\nlead us to affirm the district court’s determination.\n\n First, as with defendants’ uniform policy of not paying\nplayers for participation outside of the championship season,\ndefendants do not credibly dispute that their policy is to\nnever pay overtime and to pay a fixed salary, regardless of\nthe actual number of hours worked. We reiterate that\ncommon corporate policies like this “carry great weight for\ncertification purposes,” In re Wells Fargo Home Mortg.\nOvertime Pay Litig., 571 F.3d at 958, and that predominance\nis “rarely” defeated in cases where such uniform policies\nexist. See 7 Newberg on Class Actions § 23:33 (5th ed.\n2012).\n\n Second, the team schedules alone—independent of the\nMain Survey or any other evidence—may suffice to show\novertime liability. As the district court noted, plaintiffs’\nexpert testified that approximately 65–85% of California\nLeague players had at least one workweek with games on all\nseven days, and that nearly half of all workweeks included\ngames on all seven days. For those workweeks, the players\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 53\n\nwould be entitled to overtime pay for their work on the\nseventh day of the workweek. See Cal. Labor Code § 510.\n\n Third, and most significantly, we are persuaded that\nunder Tyson, the representative evidence plaintiffs offered\nwas adequate to meet their burden at this stage. As we\nobserved in the preceding section, defendants do identify\nmultiple legitimate criticisms of the Main Survey, and it is\ncertainly possible that a jury may not find the Main Survey—\neven in combination with all of plaintiffs’ other evidence—\nadequate proof of liability (or at least not to the extent\nplaintiffs allege). In particular, a jury may be persuaded by\ndefendants’ arguments that players did not begin\ncompensable work upon arriving at the ballpark or that\nplayers stopped engaging in compensable work long before\nthey left the ballpark, such that the Main Survey’s estimated\narrival and departure times are insufficient to clear the\npreponderance hurdle. As we explain below, however,\nTyson counsels that such criticisms do not doom certification\nhere unless no reasonable jury could conclude that the\ncombination of the Main Survey and plaintiffs’ other\nrepresentative evidence was probative of the amount of time\nplayers actually spent performing compensable work. See\nTyson, 136 S. Ct. at 1046–49. And while defendants\ncorrectly point out that the Main Survey revealed meaningful\nvariations in players’ arrival and departure times, the same\nwas true of the employees’ donning and doffing times in\nTyson—yet such variation did not preclude certification\nthere. See id. at 1043; id. at 1055 (Thomas, J., dissenting).\n\n Because defendants do not challenge the district court’s\nruling on admissibility under Daubert, the defects they have\nidentified with the Main Survey could only have defeated\ncertification upon a conclusion that all of the representative\nevidence offered—the Main Survey, schedules, testimony,\n\n54 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nand the like—could not have “sustained a reasonable jury\nfinding as to hours worked in each employee’s individual\naction.” See Tyson, 136 S. Ct. at 1046–47. As in Tyson, the\ndistrict court “made no such finding,” id. at 1049, and indeed\nfound the opposite:\n\n Plaintiffs will be able to use the survey data\n in combination with other evidence that may\n be sufficient to allow a jury to draw\n conclusions based on reasonable inference as\n to when players were required to be at the\n ballpark and how long after games they were\n required to remain at the ballpark. . . . Thus,\n as in Tyson Foods, it appears that\n representative evidence can be combined\n with actual records of time spent engaged in\n the various activities to derive a reasonable\n estimate of the amount of time worked by\n class members.\n\nWe are then left to ask whether “the record here provides [a]\nbasis for [us] to second-guess that conclusion.” Id.\n\n After reviewing the record, we are satisfied that we\nshould not disturb the district court’s determination, in part\ndue to California’s expansive definition of “employ” and\n“hours worked.” 25 Under California law, to “employ”\n\n 25\n Unlike Arizona and Florida law—the former of which is silent on\nthe incorporation of FLSA doctrines, and the latter of which expressly\nincorporates them—we are not persuaded that the continuous workday\nrule should apply to the California class. We view California’s definition\nof “hours worked” as more expansive and more employee-friendly than\nunder the FLSA, even with the incorporation of the continuous workday\nrule. The California Supreme Court has “cautioned against confounding\nfederal and state labor law,” and has consistently held that “absent\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 55\n\nmeans “to engage, suffer, or permit to work.” Cal. Code\nRegs. tit. 8, §§ 11040(2)(E), 11100(2)(E) (emphasis\nadded). 26 “Hours worked” means “the time during which an\nemployee is subject to the control of an employer, and\nincludes all the time the employee is suffered or permitted to\nwork, whether or not required to do so.” Cal. Code Regs. tit.\n8, §§ 11040(2)(K), 11100(2)(H). Inexplicably, however,\ndefendants claim that under California law, “time spent\nengaging in activities that are not required by, or under the\ncontrol of, an employer is not compensable and does not\nbegin or extend a workday.” This is a tortured and wholly\nunsupported reading of the law, and is manifestly contrary\nto one of the cases defendants themselves cite in support of\ntheir argument. See Morillion v. Royal Packing Co., 995\nP.2d 139, 143 (Cal. 2000) (“[T]he two phrases—‘time\nduring which an employee is subject to the control of an\nemployer’ and ‘time the employee is suffered or permitted\nto work, whether or not required to do so’—can also be\ninterpreted as independent factors, each of which defines\nwhether certain time spent is compensable as ‘hours\nworked.’”) (citation omitted).\n\n\n\n\nconvincing evidence of the [California agency’s] intent to adopt the\nfederal standard for determining whether time is compensable under\nstate law, we decline to import any federal standard, which expressly\neliminates substantial protections to employees, by implication.”\nTroester v. Starbucks Corp., 421 P.3d 1114, 1119 (Cal. 2018)\n(alterations, citations, and internal quotation marks omitted).\n\n 26\n The California Industrial Welfare Commission’s wage orders\n“have the force of law.” Alvarado v. Dart Container Corp. of Cal., 411\nP.3d 528, 532 (Cal. 2018). We need not decide today which wage order\napplies to minor league players, as all of the most relevant orders define\n“employ” and “hours worked” the same way.\n\n56 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n Indeed, Morillion counsels that “hours worked” includes\nall time the employer “permit[s]” an employee to work, even\nif the work is not required and the employee is not under the\nemployer’s control. See id. Thus, a player who arrives early\nor stays late at the ballpark of their own volition and\nperforms “work” activities during that time is still owed\ncompensation because the player was “permitted” to work,\ndespite the work not being required.\n\n Likewise, under Morillion, if players were expected to\narrive or depart at a particular time—whether that\nrequirement was de facto or official—it is immaterial what\nactivities the players actually engaged in while at the\nballpark. Even if the players spent their time at the ballpark\ndoing things like eating or showering, they were still under\ntheir employer’s control and unable “to use the time\neffectively for their own purposes,” and thus were owed\ncompensation. See id. at 146. Indeed, Morillion explicitly\nrejected an analogous argument by the employer in that case:\n\n We reject Royal’s contention that plaintiffs\n were not under its control during the required\n bus ride because they could read on the bus,\n or perform other personal activities.\n Permitting plaintiffs to engage in limited\n activities such as reading or sleeping on the\n bus does not allow them to use the time\n effectively for their own purposes . . .\n Plaintiffs were foreclosed from numerous\n activities in which they might otherwise\n engage if they were permitted to travel to the\n fields by their own transportation. Allowing\n plaintiffs the circumscribed activities of\n reading or sleeping does not affect, much less\n eliminate, the control Royal exercises by\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 57\n\n requiring them to travel on its buses and by\n prohibiting them from effectively using their\n travel time for their own purposes. Similarly\n . . . listening to music and drinking coffee\n while working in an office setting can also be\n characterized as personal activities, which\n would not otherwise render the time working\n noncompensable.\n\nId. (alterations, internal quotation marks, and citations\nomitted). Thus, if plaintiffs use their representative\nevidence—especially the Main Survey and the testimony of\nplayers and league officials—to persuade a jury that they\nwere required to be at the ballpark at particular times, they\nneed not show how the players spent that time.\n\n The fourth and final consideration weighing in favor of\naffirming the district court’s determination is our standard of\nreview. Abuse of discretion is always a relatively deferential\nstandard, but when we review a grant of class certification,\n“we accord the district court noticeably more deference than\nwhen we review a denial.” Abdullah, 731 F.3d at 956\n(citation omitted). Were we to review de novo, this would\nlikely be a closer call. But as they say, tie goes to the\nrunner—and, under our deferential standard, to the district\ncourt.\n\n D.\n\n Finally, defendants, citing to Wal-Mart Stores, Inc. v.\nDukes, 564 U.S. 338, 352 (2011), contend that the district\ncourt was required to “rigorously analyze” the Main Survey,\nrather than evaluating its admissibility under Daubert and its\nappropriateness for meeting class certification requirements\nunder Tyson. Tyson requires that we reject this argument.\nThere, the Court explicitly distinguished the use of\n\n58 SENNE V. KANSAS CITY ROYALS BASEBALL\n\nrepresentative evidence to establish hours worked in wage\nand hour claims from the use of representative evidence in\ncases like Wal-Mart. Tyson, 136 S. Ct. at 1048.\nSpecifically—as we have explained—for wage and hour\ncases where the employer has failed to keep proper records,\nTyson holds that once a district court has found expert\nevidence to be admissible, it may only deny its use to meet\nthe requirements of Rule 23 certification if “no reasonable\njuror” could find it probative of whether an element of\nliability was met. Id. at 1049. Given the similarities between\nthis case and Tyson, the rule set forward in Tyson controls,\nand “[defendants’] reliance on Wal-Mart is misplaced.” 27\nId. (citation omitted).\n\n IV.\n\n We next address whether the district court properly\ncertified the FLSA collective action.\n\n FLSA permits employees to bring lawsuits on behalf of\n“themselves and other employees similarly situated.” 29\nU.S.C. § 216(b). We recently delineated the appropriate\nstandard for FLSA collective certification in Campbell v.\nCity of Los Angeles, 903 F.3d 1090 (9th Cir. 2018). As we\nexplained in Campbell, “there is no established definition of\nthe FLSA’s ‘similarly situated’ requirement, nor is there an\nestablished test for enforcing it.” Id. at 1111 (citing Thiessen\nv. Gen. Elec. Capital Corp., 267 F.3d 1095, 1102 (10th Cir.\n2001)). In Campbell, we rejected both the minority\napproach to FLSA collective certification—which treats a\nFLSA collective as analogous to a Rule 23(b)(3) class—and\n\n 27\n Tyson expressly cautioned that this rule should be read narrowly\nand not assumed to apply outside of the wage and hour context. 136\nS. Ct. at 1049.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 59\n\nthe majority “ad hoc” approach. Id. at 1111–1117. The\nformer approach, we observed, is inconsistent with the\nstatute itself, as well as the choice of Congress and the\nAdvisory Committee on Rules to distinguish FLSA\ncollectives from Rule 23 class actions. Id. at 1111–1113.\nAnd while the latter approach—the so-called ad hoc\napproach—is a “significant improvement” over the minority\napproach, it has two major flaws that led us to decline to\nadopt it. Id. at 1113–1116. First, this approach\ninappropriately “focus[es] on differences rather than\nsimilarities among the party plaintiffs,” leading district\ncourts to “treat[] difference as disqualifying,” rather than\n“treat[ing] the requisite kind of similarity as the basis for\nallowing partially distinct cases to proceed together.” Id. at\n1117. Second, because the ad hoc approach allows district\ncourts to weigh “fairness and procedural considerations,” it\n“invites courts to import, through a back door, requirements\nwith no application to the FLSA,” such as Rule 23’s\npredominance, adequacy, and superiority requirements. Id.\nat 1115.\n\n Because of the flaws in the two predominant approaches\nto FLSA collective certification, we instead developed our\nown standard: “[p]arty plaintiffs are similarly situated, and\nmay proceed in a collective, to the extent they share a similar\nissue of law or fact material to the disposition of their FLSA\nclaims.” Id. at 1117. Significantly, as long as the proposed\ncollective’s “factual or legal similarities are material to the\nresolution of their case, dissimilarities in other respects\nshould not defeat collective treatment.” Id. at 1114\n(emphasis omitted).\n\n The district court here did not have the benefit of our\nopinion in Campbell, and instead followed the vast majority\nof district courts in this circuit by applying the ad hoc\n\n60 SENNE V. KANSAS CITY ROYALS BASEBALL\n\napproach. See, e.g., Lewis v. Wells Fargo & Co., 669\nF. Supp. 2d 1124, 1127 (N.D. Cal. 2009) (“Although various\napproaches have been taken to determine whether plaintiffs\nare ‘similarly situated,’ district courts in this circuit have\nused the ad hoc, two-tiered approach.”). While legally\nincorrect, we conclude that the district court’s erroneous use\nof the ad hoc approach was harmless under the\ncircumstances, 28 and we affirm the collective’s certification.\n\n The district court found that plaintiffs met their burden\nof demonstrating they were “similarly situated,” reasoning:\n\n First, by eliminating the winter conditioning\n claims and pursuing on a classwide basis only\n claims that are based on the continuous\n workday doctrine, Plaintiffs have\n significantly reduced the need to engage in\n individualized inquiries relating to the type of\n work performed. Second, the Court is now\n persuaded that the payroll records maintained\n by Defendants will allow any variations in\n compensation to be analyzed without\n burdensome individualized inquiries. This is\n especially true as to the spring training,\n extended spring training and instructional\n league claims because players generally were\n not compensated for their participation in\n these activities and the small fraction of\n players who did receive compensation for\n 28\n As we explained in Campbell, the ad hoc approach imposes a\nhigher bar for certification than the FLSA requires. See Campbell, 903\nF.3d at 1114–1116. Thus, if the collective was appropriately certified\nunder the more stringent ad hoc approach, a fortiori the collective would\nbe appropriately certified under Campbell’s more lenient approach to\n“similarly situated.” See id.\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 61\n\n these activities can be identified using payroll\n records maintained by Defendants. Third, as\n discussed above, the Court finds that the\n defenses asserted by Defendants to the FLSA\n present common questions that are not likely\n to be overwhelmed by the need to conduct\n individualized inquiries. Finally, the\n possibility that the Court will be required to\n apply the laws of numerous states (or at a\n minimum, conduct numerous choice of law\n inquiries) is not present as to the FLSA class,\n which will require the Court to apply only\n federal wage and hour law.\n\n Defendants’ arguments in support of reversal echo those\nthey make in relation to the Rule 23(b)(3) classes, and we\nreject them for largely the same reasons. Cf. Tyson, 136\nS. Ct. at 1036 (“For purposes of this case . . . if certification\nof respondents’ class action under [Rule 23] was proper,\ncertification of the collective action was proper as well.”).\nWe therefore expand on our earlier reasoning only briefly.\n\n Because the FLSA collective covers work performed\nduring spring training, extended spring training, and the\ninstructional leagues—that is, work for which the players\nreceived no pay—we affirm the certification of the collective\nfor that work. Specifically, for these time periods, two\ncommon legal questions drive the litigation: are the players\nemployees, and do the activities they perform during those\ntimes constitute compensable work? As nearly all players\nare unpaid during these time periods, if the answers to those\ntwo questions are resolved in plaintiffs’ favor, liability may\nbe established by showing that the players performed any\nwork.\n\n62 SENNE V. KANSAS CITY ROYALS BASEBALL\n\n We also affirm the district court’s certification of the\nFLSA collective as to plaintiffs’ overtime claims, although\nthis holding requires additional explanation. Critical to our\ndecision is that plaintiffs allege a single, FLSA-violating\npolicy—the failure to pay overtime under any\ncircumstances—and argue a common theory of defendants’\nstatutory violations: that defendants “suffer or permit”\nplaintiffs to perform compensable work before and after\nscheduled practice and game times. These are “similar\nissue[s] of law or fact material to the disposition of their\nFLSA claims,” thus making plaintiffs “similarly situated.”\nCampbell, 903 F.3d at 1117. And as previously discussed,\nwe believe a reasonable jury could find that all of plaintiffs’\nevidence—not just the Main Survey, but also the schedules,\ntestimony, and payroll data—sustains a “just and reasonable\ninference” as to the hours players actually worked. See\nTyson, 136 S. Ct. at 1046–47.\n\n Specifically, there are several overlapping ways that\nplaintiffs may be able to rely on their representative evidence\nto persuade a jury that they have worked overtime hours for\nwhich they were not compensated. Under any of these\nscenarios, the continuous workday rule lends significant\nassistance to plaintiffs by eliminating the need for plaintiffs\nto prove exactly which activities they engaged in throughout\nthe day. See IBP, 546 U.S. at 28, 32–37.\n\n First, plaintiffs could potentially use their evidence—\nparticularly the Main Survey, but also the testimony of\nplayers and league officials—to establish approximate times\nthat they were required to arrive at and depart from the\nballpark. This would obviate the need for plaintiffs to\ndemonstrate which activities they engaged in upon arrival or\nprior to departure. See 29 C.F.R. § 790.6 (“If an employee\nis required to report at the actual place of performance of his\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 63\n\nprincipal activity at a certain specific time, his ‘workday’\ncommences at the time he reports there for work in\naccordance with the employer's requirement.”); 29 C.F.R.\n§ 778.223 (“As a general rule the term ‘hours worked’ will\ninclude . . . [a]ll time during which an employee is required\nto be on duty or to be on the employer’s premises or at a\nprescribed workplace.”) (emphasis added).\n\n Second, plaintiffs could rely on their representative\nevidence to demonstrate that before and after the times they\nwere required to be at the ballpark, they still performed\nactivities at the ballpark that were “an integral and\nindispensable part of [their] principal activities” and were\ntherefore compensable. See IBP, 546 U.S. at 29–30. As\nmentioned previously, a jury may well determine that\nactivities like batting practice or supervised weightlifting are\nto baseball players what knife-sharpening is to butchers at a\nmeatpacking plant—that is, activities that are “integral and\nindispensable” to the principal activity of playing baseball.\nSee Mitchell, 350 U.S. at 261–63. If so, such activities\nwould trigger the start of the “workday” within the meaning\nof the FLSA. Plaintiffs may have somewhat of an uphill\nbattle proceeding under this second theory on a collective-\nwide basis, but we are certainly not prepared to say that no\nreasonable jury could find defendants liable for overtime\nviolations under this theory. See Tyson, 136 S. Ct. at 1048–\n49; cf. Campbell, 903 F.3d at 1117–1119 (explaining that\npost-discovery decertification motions should be evaluated\nunder the summary judgment standard where “overlap exists\nbetween the availability of the collective action mechanism\nand the merits of the underlying claim”).\n\n Finally, if internal team schedules establish that\nplaintiffs had required team-related activities for forty hours\n\n64 SENNE V. KANSAS CITY ROYALS BASEBALL\n\na week, 29 then plaintiffs can establish liability simply by\nshowing that they performed any additional work beyond\nthose officially-scheduled times. Cf. Tyson, 136 S. Ct. at\n1036 (Thomas, J., dissenting) (explaining that in Mt.\nClemens, the employer was “presumptively liable to all\nemployees because they all claimed to work 40 hours per\nweek. All additional uncompensated work was necessarily\nunpaid overtime.”) (citation omitted).\n\n Under any of these theories, damages will inevitably be\nindividualized, at least to some extent. But just as the need\nfor individualized damage calculations is insufficient to\ndefeat Rule 23 certification, “[i]ndividual damages amounts\ncannot defeat collective treatment under the more forgiving\nstandard” for FLSA collective certification. See Campbell,\n903 F.3d at 1117 (citing Leyva v. Medline Indus. Inc., 716\nF.3d 510, 514 (9th Cir. 2013)). District courts are well-\nequipped to deal with issues of individualized calculations\nin the wage-and-hour context, and may use “any of the\npractices developed to deal with Rule 23 classes facing\nsimilar issues.” Id. at 18 (citing Jimenez v. Allstate Ins. Co.,\n765 F.3d 1161, 1167 (9th Cir. 2014)).\n\n\n\n 29\n Given the internal team schedules in the record, this may be an\neasy task, particularly for spring training and extended spring training.\nFor example, a spring training schedule for one of the San Francisco\nGiants’ affiliates involved a workday beginning at 6:30 AM on the day\nof a 1:00 PM away game, with a 50 minute window provided for transit\nbetween the training facility and the ballpark. Assuming for the sake of\nargument that the 1:00 PM game lasted 2.5 hours and that the return trip\nto the training facility took the same amount of time—50 minutes—as\nthe outgoing trip, that day alone entailed approximately 10 hours of work\nif the players left the training facility immediately upon their return (and\nbased on the testimony in the record, that assumption seems\nimplausible).\n\n SENNE V. KANSAS CITY ROYALS BASEBALL 65\n\n As is true in all FLSA cases, underlying our decision\ntoday is the background principle that “because the FLSA is\na remedial statute, it must be interpreted broadly.” Lambert\nv. Ackerley, 180 F.3d 997, 1003 (9th Cir. 1999) (en banc)\n(citing Tennessee Coal, Iron & R. Co. v. Muscoda Local No.\n123, 321 U.S. 590, 597 (1944)). After all, the FLSA does\nnot deal “with mere chattels or articles of trade but with the\nrights of those who toil.” Tennessee Coal, 321 U.S. at 597.\nWe are satisfied that certification of the collective is not only\nappropriate under our interpretation of “similarly situated,”\nbut also that it is consistent with “the great public policy”\nembodied by the FLSA. Mt. Clemens, 328 U.S. at 687.\n\n V.\n\n For the reasons explained above, we AFFIRM in part,\nREVERSE in part, and REMAND for further proceedings\nconsistent with this opinion.\n\n Plaintiffs-Appellants/Cross-Appellees shall recover\ntheir costs on appeal.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Aaron Senne v. Kansas City Royals Baseball\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-16\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nAARON SENNE; MICHAEL LIBERTO; Nos. 17-16245\\nOLIVER ODLE; BRAD MCATEE; CRAIG 17-16267\\nBENNIGSON; MATT LAWSON; KYLE 17-16276\\nWOODRUFF; RYAN KIEL; KYLE\\nNICHOLSON; BRAD STONE; MATT D.C. No.\\nDALY; AARON MEADE; JUSTIN 3:14-cv-00608-\\nMURRAY; JAKE KAHAULELIO; RYAN JCS\\nKHOURY; DUSTIN PEASE; JEFF\\nNADEAU; JON GASTON; BRANDON\\nHENDERSON; TIM PAHUTA; LEE OPINION\\nSMITH; JOSEPH NEWBY; RYAN\\nHUTSON; MATT FREVERT; ROBERTO\\nORTIZ; WITER JIMENEZ; KRIS WATTS;\\nMITCH HILLIGOSS; DANIEL BRITT;\\nYADEL MARTI; HELDER VELAQUEZ;\\nJORGE JIMENEZ; JORGE MINYETY;\\nEDWIN MAYSONET; JOSE DIAZ; NICK\\nGIARRAPUTO; LAUREN GAGNIER;\\nLEONARD DAVIS; GASPAR SANTIAGO;\\nGRANT DUFF; OMAR AGUILAR; MARK\\nWAGNER; DAVID QUINOWSKI;\\nBRANDON PINCKNEY, Individually\\nand on Behalf of All Those Similarly\\nSituated; JAKE OPITZ; BRETT\\nNEWSOME,\\n Plaintiffs-Appellants,\\n\\n v.\\n\\n2 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n\\nKANSAS CITY ROYALS BASEBALL\\nCORP.; MARLINS TEAMCO LLC; SAN\\nFRANCISCO BASEBALL ASSOCIATES,\\nLLC; OFFICE OF THE COMMISSIONER\\nOF BASEBALL, DBA Major League\\nBaseball, an unincorporated\\nassociation; ALLAN HUBER SELIG,\\n“BUD”; ANGELS BASEBALL LP; ST.\\nLOUIS CARDINALS, LLC; COLORADO\\nROCKIES BASEBALL CLUB, LTD.;\\nCINCINNATI REDS, LLC; HOUSTON\\nBASEBALL PARTNERS LLC;\\nATHLETICS INVESTMENT GROUP,\\nLLC; ROGERS BLUE JAYS BASEBALL\\nPARTNERSHIP; PADRES L.P.; SAN\\nDIEGO PADRES BASEBALL CLUB, L.P.;\\nMINNESOTA TWINS, LLC; DETROIT\\nTIGERS, INC.; LOS ANGELES\\nDODGERS LLC; STERLING METS L.P.;\\nAZPB L.P.; NEW YORK YANKEES\\nP’SHIP; RANGERS BASEBALL EXPRESS,\\nLLC; MILWAUKEE BREWERS\\nBASEBALL CLUB, INC.; CHICAGO\\nCUBS BASEBALL CLUB, LLC;\\nPITTSBURGH ASSOCIATES, LP;\\nBASEBALL CLUB OF SEATTLE, LLP;\\nLOS ANGELES DODGERS HOLDING\\nCOMPANY LLC; RANGERS BASEBALL,\\nLLC,\\n Defendants-Appellees.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 3\\n\\n Appeals from the United States District Court\\n for the Northern District of California\\n Joseph C. Spero, Magistrate Judge, Presiding\\n\\n Argued and Submitted June 13, 2018\\n San Francisco, California\\n\\n Filed August 16, 2019\\n\\n Before: Michael R. Murphy, * Richard A. Paez,\\n and Sandra S. Ikuta, Circuit Judges.\\n\\n Opinion by Judge Paez;\\n Dissent by Judge Ikuta\\n\\n\\n\\n\\n *\\n The Honorable Michael R. Murphy, United States Circuit Judge\\nfor the U.S. Court of Appeals for the Tenth Circuit, sitting by\\ndesignation.\\n\\n4 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n SUMMARY **\\n\\n\\n Labor Law / Class and Collective Certification\\n\\n The panel affirmed in part and reversed in part the\\ndistrict court’s orders certifying a class and a collective\\naction for wage-and-hour claims brought by minor league\\nbaseball players under the Fair Labor Standards Act and\\nstate law.\\n\\n The district court certified a California class under\\nFederal Rule of Civil Procedure 23(b)(3) but denied\\ncertification for Arizona and Florida classes and for a Rule\\n23(b)(2) class. The district court also certified an FLSA\\ncollective.\\n\\n The panel held that, as to the state law claims, California\\nchoice-of-law rules applied. The panel held that under\\nSullivan v. Oracle Corp., 254 P.3d 237 (Cal. 2011),\\nCalifornia law applied to the Rule 23(b)(3) California class.\\nThe panel reversed the district court’s determination that\\nchoice-of-law considerations defeated the predominance and\\nadequacy requirements for the proposed Arizona and Florida\\nRule 23(b)(3) classes. Applying California’s three-step\\ngovernmental interest analysis for choice-of-law questions,\\nthe panel concluded that Arizona law should apply to the\\nwork performed in Arizona, and Florida law to the work\\nperformed in Florida.\\n\\n The panel reversed the district court’s refusal to certify a\\nRule 23(b)(2) class for unpaid work at defendants’ training\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 5\\n\\nfacilities in Arizona and Florida on the basis that choice-of-\\nlaw issues undermined “cohesiveness” and therefore made\\ninjunctive and declaratory relief inappropriate. The panel\\nconcluded that the district court’s errors in its choice-of-law\\nanalysis relating to the proposed Arizona and Florida Rule\\n23(b)(3) classes applied equally to its refusal to certify the\\nRule 23(b)(2) class. The panel further held that the district\\ncourt erred in imposing a “cohesiveness” requirement for the\\nproposed Rule 23(b)(2) class. The panel remanded for the\\ndistrict court to consider anew whether to certify the Rule\\n23(b)(2) class.\\n\\n The panel held that plaintiffs could meet the\\npredominance requirement for the proposed California,\\nFlorida, and Arizona Rule 23(b)(3) classes through a\\ncombination of representative evidence and application of\\nthe “continuous workday” rule. The panel applied the Mt.\\nClemens burden-shifting framework and the holding of\\nTyson Foods v. Bouaphakeo, 136 S. Ct. 1036 (2016), that\\nrepresentative evidence may be used at the class certification\\nstage and may be used to establish liability in addition to\\ndamages. The panel explained that the continuous workday\\nrule presumes that once the beginning of the workday is\\ntriggered, an employee performs compensable work\\nthroughout the rest of the day until the employee completes\\ntheir last principal activity. Any activity that is “integral and\\nindispensable” to principal activities triggers the beginning\\nof the workday. As to the Arizona and Florida classes,\\ncovering alleged minimum wage violations in the lack of any\\npay for time spent participating in spring training, extended\\nspring training, and instructional leagues, the panel affirmed\\nthe determination that the predominance requirement was\\nmet. As to the California class, covering overtime and\\nminimum wage claims relating to work performed during the\\nchampionship season, the panel held that the district court\\n\\n6 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\ndid not abuse its discretion in concluding that defendants’\\nuniform pay policy, the team schedules, and representative\\nevidence, including an expert survey known as the “Main\\nSurvey,” established predominance. The panel held that the\\ndistrict court was not required to “rigorously analyze” the\\nMain Survey, rather than evaluating its admissibility under\\nDaubert and its appropriateness for meeting class\\ncertification requirements under Tyson.\\n\\n Affirming the district court’s certification of the FLSA\\ncollective action, the panel applied the standard set forth in\\nCampbell v. City of L.A., 903 F.3d 1090 (9th Cir. 2018),\\nwhich postdated the district court’s ruling, and held that the\\ndistrict court’s use of the ad hoc approach was harmless\\nerror. The panel concluded that collective certification was\\nproper because plaintiffs shared similar issues of law or fact\\nmaterial to the disposition of their FLSA claims and thus\\nwere similarly situated.\\n\\n The panel affirmed the district court’s certification of the\\nCalifornia Rule 23(b)(3) class and the FLSA collective\\naction, reversed the district court’s refusal to certify Arizona\\nand Florida classes and a Rule 23(b)(2) class, and remanded\\nfor further proceedings.\\n\\n Dissenting, Judge Ikuta wrote that the district court\\ncorrectly concluded that consideration of plaintiffs’ claims\\non a classwide basis would be overwhelmed by\\nindividualized choice-of-law inquiries. She wrote that the\\nmajority’s rule, applying the law of the jurisdiction where\\nthe work took place, was contrary to the court’s framework\\nfor analyzing the intersection of class action and choice-of-\\nlaw issues, overlooked the complexity of California’s\\nchoice-of-law rules, and created significant practical and\\nlogistical problems.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 7\\n\\n COUNSEL\\n\\nRobert L. King (argued) and Garrett R. Broshuis, Korein\\nTillery LLC, St. Louis, Missouri; Bruce L. Simon and\\nBenjamin E. Shiftan, Pearson Simon & Warshaw LLP, San\\nFrancisco, California; Daniel L. Warshaw and Bobby Pouya,\\nPearson Simon & Warshaw LLP, Sherman Oaks, California;\\nfor Plaintiffs-Appellants.\\n\\nElise M. Bloom (argued), Adam M. Lupion, and Mark D.\\nHarris, Proskauer Rose LLP, New York, New York; John E.\\nRoberts, Proskauer Rose LLP, Boston, Massachusetts; for\\nDefendants-Apellees.\\n\\nDavid C. Frederick and Jeffrey A. Love, Kellogg Hansen\\nTodd Figel & Frederick PLLC, Washington, D.C., for Amici\\nCuriae Professors Peter Hay and Patrick J. Borchers.\\n\\nAllan Steyer and Donald Scott MacRae, Steyer Lowenthal\\nBoodrookas Alvarez & Smith LLP, San Francisco,\\nCalifornia, for Amici Curiae Professional Hockey Players\\nAssocation; Association of Minor League Umpires; Office\\nand Professional Employee’s International Union; and\\nUnited Steel, Paper and Forestry, Rubber, Manufacturing,\\nEnergy, Allied and Industrial Service Workers International\\nUnion.\\n\\n8 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n OPINION\\n\\nPAEZ, Circuit Judge:\\n\\n It is often said that baseball is America’s pastime. In this\\ncase, current and former minor league baseball players\\nallege that the American tradition of baseball collides with a\\ntradition far less benign: the exploitation of workers. We are\\ntasked with deciding whether these minor league players\\nmay properly bring their wage-and-hour claims on a\\ncollective and classwide basis.\\n\\n BACKGROUND\\n\\n I.\\n\\n Most major professional sports in America have their\\nown “farm system” for developing talent: for the National\\nBasketball Association, it’s the G-League; for the National\\nHockey League, it’s the American Hockey League; and for\\nMajor League Baseball (MLB), it’s Minor League Baseball.\\nMLB and its thirty franchise teams rely heavily on this\\nextensive minor league system, which has nearly 200\\naffiliates across the country and employs approximately\\n6,000 minor league players. Nearly all MLB players begin\\ntheir careers in the minor leagues. Each minor league club\\nis associated with one of the thirty franchise MLB teams.\\n\\n The minor league system is governed by the Major\\nLeague Rules (MLRs), which dictate the terms of\\nemployment and compensation for both minor and major\\nleague players. Under the MLRs, all minor league players\\nare required to sign a seven-year Uniform Player Contract\\n(UPC). Ostensibly, players are required to sign the UPC for\\n“morale” and “to produce the similarity of conditions\\nnecessary for keen competition.”\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 9\\n\\n The UPC “obligates Player[s] to perform professional\\nservices on a calendar year basis, regardless of the fact that\\nsalary payments are to be made only during the actual\\nchampionship playing season.” It describes its scope as\\nsetting “the terms and conditions of employment during all\\nperiods in which Player is employed by Club as a Minor\\nLeague Player.” Players are paid by the MLB franchise\\naffiliated with the minor league team for which they play.\\nUnder the UPC, first-year players are paid a fixed salary of\\n$1,100 per month during the regular (“championship”)\\nseason that runs from April through September. In addition\\nto their salaries during the championship season, some\\nplayers receive signing or performance-related bonuses and\\ncollege scholarships.\\n\\n Beginning in early March each year, the minor league\\naffiliates conduct spring training in Arizona and Florida;\\nevery MLB franchise operates a minor league training\\ncomplex in one of these two states. The parties dispute\\nwhether spring training is required, but the UPC strongly\\nindicates that it is mandatory. 1 Virtually all players are\\nunpaid during spring training.\\n\\n Spring training lasts approximately four weeks, until the\\nchampionship season begins in April. Some players attest\\nthat spring training entails working seven days a week, with\\nno days off. During spring training, teams typically have\\n\\n 1\\n The UPC provides that “Player’s duties and obligations under [the\\nUPC] continue in full force and effect throughout the calendar year,\\nincluding Club’s championship playing season, Club’s training season,\\nClub’s exhibition games, Club’s instructional, post-season training or\\nwinter league games, any official play-off series, any other official post-\\nseason series in which Player shall be required to participate . . . and any\\nremaining portions of the calendar year.”\\n\\n10 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nscheduled activities in the morning prior to playing games in\\nthe afternoon. For example, a team spring training schedule\\nfor one of the San Francisco Giants’ affiliates describes that\\nat 6:30 AM, there was an “Early Van for Treatment and\\nEarly Work” 2; at 7:00 AM, the “Regular Van” departed; at\\n7:45 AM, the “Early Work” began; and then between 9:00\\nAM and 11:00 AM, the team would perform activities such\\nas “Stretch,” “Throwing Program,” and “Batting Practice.”\\nLunch was to be at 11:00 AM, before a 12:10 PM bus to a\\nneighboring city for a 1:00 PM away game.\\n\\n At the conclusion of spring training in early April, some\\nplayers are assigned to minor league affiliates, and begin\\nplaying games in the championship season. During the\\nchampionship season, minor league teams play games either\\nsix or seven days per week. The championship season lasts\\naround five months, beginning in April and ending in\\nSeptember. One of the regular season leagues within minor\\nleague baseball is the California League, which—as the\\nname implies—plays games exclusively within California.\\n\\n Players who are not assigned to play for affiliates in the\\nchampionship season stay at the Arizona or Florida facilities\\nfor “extended spring training.” Extended spring training\\ncontinues until June, and involves similar activities to spring\\ntraining. Although most players do not get paid during\\nextended spring training, as many as seven MLB clubs do\\npay for work during extended spring training due to an\\nambiguity in the MLRs over when players are permitted to\\nbe paid.\\n\\n\\n\\n 2\\n The schedule instructed players to “CHECK [the] BOARD FOR\\nEARLY WORK.”\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 11\\n\\n After the championship season ends in September, some\\nplayers participate in the “instructional leagues,” which run\\nfrom approximately mid-September to mid-October. The\\nparties dispute whether participation in the instructional\\nleagues is mandatory for the players involved, although as\\nwith spring training, the UPC strongly implies that\\nparticipation is required. Activities and schedules during the\\ninstructional league are similar to spring training. And just\\nas with spring training, players are virtually never paid for\\nparticipation in the instructional league.\\n\\n II.\\n\\n Plaintiffs are forty-five current and former minor league\\nbaseball players who bring claims under the federal Fair\\nLabor Standards Act (FLSA) and the wage-and-hour laws of\\nCalifornia, Arizona, and Florida against MLB, MLB\\nCommissioner Bud Selig, and a number 3 of MLB franchises.\\nPlaintiffs allege that defendants do not pay the players at all\\nduring spring training, extended spring training, or the\\ninstructional leagues. They further allege that because\\nplayers are “employees” and the activities the players\\nperform during those periods constitute compensable work,\\ndefendants have unlawfully failed to pay them at least\\nminimum wage. And according to plaintiffs, while the\\nplayers are paid—albeit not much—during the\\nchampionship season, they routinely work overtime, for\\nwhich they are never compensated as a matter of policy.\\n\\n In May 2015, plaintiffs filed their Second Amended\\nConsolidated Class Action Complaint, which alleged wage-\\n\\n 3\\n Plaintiffs originally named all 30 MLB franchises as defendants,\\nbut eight of the franchises were subsequently dismissed for lack of\\npersonal jurisdiction.\\n\\n12 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nand-hour claims under the laws of eight states and the FLSA;\\nplaintiffs also sought certification of a FLSA collective\\naction. The district court preliminarily certified the FLSA\\ncollective in October 2015. Notice was sent to\\napproximately 15,000 current and former minor league\\nplayers, of which more than 2,200 opted in.\\n\\n In 2016, defendants moved to decertify the FLSA\\ncollective, while plaintiffs moved to certify a Rule 23(b)(2)\\nclass as well as Rule 23(b)(3) classes under the laws of eight\\nstates. The district court denied certification for all proposed\\nRule 23(b)(3) classes, concluding that predominance was not\\nsatisfied for two primary reasons. Senne v. Kansas City\\nRoyals Baseball Corp., 315 F.R.D. 523, 572, 577–84 (N.D.\\nCal. 2016). First, the court concluded that predominance\\nwas defeated by the choice-of-law issues presented by the\\nproposed classes, given that (1) the winter off-season\\ntraining claims entailed work performed in dozens of\\ndifferent states with no common schedule or situs; and\\n(2) the championship season claims involved frequent travel\\nbetween state lines for away games. Id. at 580–81. The\\ndistrict court also determined that the inclusion of claims for\\nwinter off-season work fatally undermined predominance, as\\nthe court would be required to undertake an overwhelming\\nnumber of individualized inquiries to determine which\\nactivities constituted compensable “work” and how much\\ntime was spent doing “work.” Id. at 577–84. For similar\\nreasons, the court held that plaintiffs were not “similarly\\nsituated” and therefore decertified the FLSA collective. Id.\\nat 585–86. The court also granted the defendants’ motion to\\nexclude an expert survey (the “Pilot Survey”) submitted by\\nplaintiffs, finding that its methodology and results did not\\nsatisfy the requirements of Federal Rule of Evidence 702 and\\nDaubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S.\\n579 (1993). Id. at 586–90. The court further refused to\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 13\\n\\ncertify the proposed Rule 23(b)(2) class, concluding that\\nbecause the plaintiffs were all former—rather than current—\\nplayers, they lacked standing to represent a (b)(2) class. Id.\\nat 584–85.\\n\\n Plaintiffs moved for reconsideration, narrowing their\\nproposed classes significantly in response to the concerns the\\ndistrict court expressed in its initial certification order.\\nPlaintiffs requested Rule 23(b)(3) certification of an Arizona\\nclass and a Florida class for work performed during spring\\ntraining, extended spring training, and the instructional\\nleagues in those states. Plaintiffs also moved for\\ncertification of a 23(b)(3) California class, covering players\\nwho participated in the California League during the\\nchampionship season. Additionally, plaintiffs sought to\\ncertify a reworked FLSA collective of players who\\nparticipated in the California League or in spring training,\\nextended spring training, and the instructional leagues. In\\naddition to the 23(b)(3) classes and FLSA collective,\\nplaintiffs requested certification of a Rule 23(b)(2)\\ninjunctive relief class consisting of current minor league\\nplayers who participate in spring training, extended spring\\ntraining, or the instructional leagues in Florida or Arizona.\\nTo cure the court’s earlier concerns about standing, four\\ncurrent minor league players moved to intervene to represent\\nthe proposed (b)(2) class.\\n\\n On reconsideration, plaintiffs argued that they could\\nmeet Rule 23(b)(3)’s predominance requirement and\\nFLSA’s “similarly situated” requirement through a\\ncombination of the use of representative evidence and\\napplication of the so-called “continuous workday” rule. 4\\n\\n 4\\n As we shall explain, the continuous workday rule presumes that\\nonce the beginning of the workday is triggered, an employee performs\\n\\n14 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nPlaintiffs’ representative evidence took a variety of forms,\\nincluding an expert survey (the “Main Survey”), hundreds of\\nteam schedules, payroll data, and testimony from both\\nplayers and league officials. The most controversial piece of\\nevidence was the Main Survey, which plaintiffs argued\\nserved as representative evidence of hours worked,\\nparticularly when used in concert with a continuous workday\\ntheory.\\n\\n The Main Survey asked players to report the times they\\n“most often” arrived and departed from the ballpark or\\ntraining facility during the championship season, spring\\ntraining, extended spring training, and the instructional\\nleagues, and asked players to estimate how much time they\\nspent eating meals while at the ballpark. The survey did not,\\nhowever, ask players about the kinds of activities they\\nperformed at the facilities, or how much time they spent\\nperforming particular activities. Given these purported\\nshortcomings, defendants moved to exclude the Main\\nSurvey, and further argued that even if the survey were\\nadmissible under Daubert, it still could not be used to meet\\nthe predominance and “similarly situated” requirements due\\nto its alleged flaws. The district court denied defendants’\\nmotion to exclude the Main Survey, finding it admissible\\nunder Daubert and concluding that defendants’ challenges\\nwent “to the weight of the Survey and not its admissibility”\\nand were “better left to a jury to evaluate.” The district court\\nfurther concluded that the Main Survey could be used in\\ncombination with other evidence—such as team schedules,\\ntestimony, and payroll data—to meet Rule 23(b)(3)’s\\n\\ncompensable work throughout the rest of the day until the employee\\ncompletes their last principal activity or the last activity which is\\n“integral and indispensable” to the employee’s principal activities. IBP,\\nInc. v. Alvarez, 546 U.S. 21, 28, 32–37 (2005).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 15\\n\\npredominance and FLSA’s “similarly situated”\\nrequirements, observing that certifying the classes and the\\nFLSA collective “will not preclude Defendants from\\nchallenging the sufficiency of the Main Survey and\\nPlaintiffs’ damages model on summary judgment and/or at\\ntrial.”\\n\\n Because it concluded that the predominance and\\n“similarly situated” requirements could be met with the use\\nof representative evidence and application of the continuous\\nworkday rule, the district court recertified the narrowed\\nFLSA collective and certified a California (b)(3) class.\\nHowever, the district court denied certification for the\\nArizona, Florida, and (b)(2) classes, holding that choice-of-\\nlaw concerns defeated predominance for the Arizona and\\nFlorida classes and undermined “cohesiveness” for the\\n(b)(2) class.\\n\\n At defendants’ request, the district court certified the\\nFLSA collective certification order for interlocutory review\\nunder 28 U.S.C. § 1292. Plaintiffs petitioned us for\\npermission to appeal the denial of certification for the\\nArizona, Florida, and Rule 23(b)(2) classes, and defendants\\nlikewise petitioned to appeal the certification of the\\nCalifornia class; we granted both petitions, consolidating\\nthose cross-appeals with the FLSA collective appeal.\\n\\n STANDARD OF REVIEW\\n\\n We review for abuse of discretion the district court’s\\nclass certification rulings, and review for clear error any\\nfindings of fact the district court relied upon in its\\ncertification order. Parsons v. Ryan, 754 F.3d 657, 673 (9th\\nCir. 2014). A district court’s choice of law determinations,\\nhowever, are reviewed de novo. Mazza v. Am. Honda Motor\\nCo., 666 F.3d 581, 589 (9th Cir. 2012). A district court\\n\\n16 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nabuses its discretion where it commits an error of law, relies\\non an improper factor, omits a substantial factor, or engages\\nin a clear error of judgment in weighing the correct mix of\\nfactors. Stockwell v. City & County of San Francisco, 749\\nF.3d 1107, 1113 (9th Cir. 2014) (citing Bateman v. Am.\\nMulti-Cinema, Inc., 623 F.3d 708, 712 (9th Cir. 2010)).\\nWhen we review a grant of class certification, “we accord\\nthe district court noticeably more deference than when we\\nreview a denial.” Abdullah v. U.S. Sec. Assocs., Inc., 731\\nF.3d 952, 956 (9th Cir. 2013) (quoting Wolin v. Jaguar Land\\nRover N. Am., LLC, 617 F.3d 1168, 1171 (9th Cir. 2010)).\\n\\n ANALYSIS\\n\\n To paraphrase the Chief Justice, these complex appeals\\nrequire us to call a great number of balls and strikes, as both\\nparties raise numerous challenges to the district court’s\\ncertification order. For their part, plaintiffs challenge the\\ndistrict court’s decision to deny certification for the Arizona\\nand Florida Rule 23(b)(3) classes and the Rule 23(b)(2) class\\non the grounds that choice-of-law issues defeated the\\npredominance requirement for the Arizona and Florida\\n(b)(3) classes and also thwarted “cohesiveness” for the\\nproposed (b)(2) class. Defendants, on the other hand, contest\\nthe district court’s certification of the California (b)(3) class,\\narguing first that choice-of-law issues defeat both\\npredominance and adequacy, and second, that plaintiffs\\ncannot meet the predominance requirement through the use\\nof their proffered representative evidence: the Main Survey,\\nteam schedules, payroll records, deposition testimony, and\\ndeclarations. Defendants further charge that the district\\ncourt erred in certifying the FLSA collective because\\nplaintiffs’ representative evidence does not show that the\\ncollective members are “similarly situated.” Defendants\\nalso contend that the district court erred by not “rigorously\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 17\\n\\nanalyzing” plaintiffs’ expert evidence at the class and\\ncollective certification stage. We address each argument in\\nturn.\\n\\n I.\\n\\n Class certification is governed by Federal Rule of Civil\\nProcedure 23. As a threshold matter, a party seeking class\\ncertification must satisfy the four requirements of Rule\\n23(a): (1) numerosity; (2) commonality; (3) typicality; and\\n(4) adequacy of representation. 5 “Class certification is\\nproper only if the trial court has concluded, after a ‘rigorous\\nanalysis,’ that Rule 23(a) has been satisfied.” Parsons, 754\\nF.3d at 674 (quoting Wang v. Chinese Daily News, Inc., 737\\nF.3d 538, 542–43 (9th Cir. 2013)).\\n\\n In addition to the requirements of Rule 23(a), a proposed\\nclass must also meet the requirements of one or more of the\\n“three different types of classes” set forth in Rule 23(b).\\nLeyva v. Medline Industries, Inc., 716 F.3d 510, 512 (9th Cir.\\n2013). Here, plaintiffs proposed classes under two of Rule\\n23(b)’s class types: Rule 23(b)(3) and 23(b)(2). A class may\\nbe certified under Rule 23(b)(3) only if the district court\\n“finds that the questions of law or fact common to class\\nmembers predominate over any questions affecting only\\nindividual members, and that a class action is superior to\\nother available methods for fairly and efficiently\\nadjudicating the controversy.” Fed. R. Civ. P. 23(b)(3). Of\\nthese two requirements—predominance and superiority—\\nonly predominance is at issue on appeal. “The\\npredominance inquiry focuses on ‘the relationship between\\n\\n 5\\n Of Rule 23(a)’s four requirements, defendants contest only\\nadequacy on appeal; their arguments pertaining to adequacy have to do\\nwith choice-of-law issues.\\n\\n18 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nthe common and individual issues’ and ‘tests whether\\nproposed classes are sufficiently cohesive to warrant\\nadjudication by representation.’” Vinole v. Countrywide\\nHome Loans, Inc., 571 F.3d 935, 944 (9th Cir. 2009)\\n(quoting Hanlon v. Chrysler Corp., 150 F.3d 1011, 1022 (9th\\nCir. 1998)). In determining whether the predominance\\nrequirement is met, courts have a “duty to take a close look\\nat whether common questions predominate over individual\\nones” to ensure that individual questions do not “overwhelm\\nquestions common to the class.” Comcast Corp. v. Behrend,\\n569 U.S. 27, 34 (2013) (internal quotation marks and citation\\nomitted).\\n\\n Rule 23(b)(2), on the other hand, requires only that “the\\nparty opposing the class ha[ve] acted or refused to act on\\ngrounds that apply generally to the class, so that final\\ninjunctive relief or corresponding declaratory relief is\\nappropriate respecting the class as a whole.” Fed. R. Civ. P.\\n23(b)(2). Although 23(b)(2) classes are most common in the\\ncivil rights context, “we have certified many different kinds\\nof Rule 23(b)(2) classes.” Parsons, 754 F.3d at 686.\\n\\n II.\\n\\n We first address whether choice-of-law issues fatally\\nundermine plaintiffs’ proposed Rule 23 classes. The district\\ncourt’s decision was split on the impact of choice-of-law\\nquestions: as to the proposed Rule 23(b)(3) California class,\\nthe court held that choice-of-law concerns defeated neither\\nRule 23(b)(3)’s predominance requirement nor Rule 23(a)’s\\nadequacy requirement. Yet as to the proposed 23(b)(3)\\nArizona and Florida classes, the district court held the\\nopposite: that choice-of-law issues posed an insurmountable\\nhurdle to meeting both predominance and adequacy.\\nSimilarly, the court determined that choice-of-law questions\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 19\\n\\nmade certification of the proposed Rule 23(b)(2) class\\ninappropriate.\\n\\n Concerns over which state’s laws apply to a proposed\\nclass “do not necessarily preclude a 23(b)(3) action.”\\nHanlon, 150 F.3d at 1022. But “[u]nderstanding which law\\nwill apply before making a predominance determination is\\nimportant when there are variations in applicable state law,”\\nand potentially varying state laws may defeat predominance\\nin certain circumstances. Zinser v. Accufix Research Inst.,\\nInc., 253 F.3d 1180, 1189 (9th Cir. 2001) opinion amended\\non denial of reh’g, 273 F.3d 1266 (9th Cir. 2001). We have\\nbeen particularly concerned about the impact of choice-of-\\nlaw inquiries in nationwide consumer class actions and\\nproducts liability cases. See, e.g., Mazza, 666 F.3d at 585,\\n591–94; Zinser, 253 F.3d at 1184–90.\\n\\n A district court considering state law claims brought in\\nfederal court must utilize the choice-of-law rules of the\\nforum state—here, California. Klaxon Co. v. Stentor\\nElectric Mfg. Co., 313 U.S. 487, 496–97 (1941). “By\\ndefault, California courts apply California law unless a party\\nlitigant timely invokes the law of a foreign state, in which\\ncase it is the foreign law proponent who must shoulder the\\nburden of demonstrating that foreign law, rather than\\nCalifornia law, should apply to class claims.” In re Hyundai\\n& Kia Fuel Econ. Litig., 926 F.3d 539, 561 (9th Cir. 2019)\\n(en banc) (internal quotation marks and citations omitted).\\nTo meet their burden, the objectors must satisfy California’s\\nthree-step governmental interest test, used to resolve choice\\nof law issues. Id.\\n\\n First, the court determines whether the\\n relevant law of each of the potentially\\n affected jurisdictions with regard to the\\n particular issue in question is the same or\\n\\n20 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n different. Second, if there is a difference, the\\n court examines each jurisdiction’s interest in\\n the application of its own law under the\\n circumstances of the particular case to\\n determine whether a true conflict exists.\\n Third, if the court finds that there is a true\\n conflict, it carefully evaluates and compares\\n the nature and strength of the interest of each\\n jurisdiction in the application of its own law\\n to determine which state’s interest would be\\n more impaired if its policy were subordinated\\n to the policy of the other state, and then\\n ultimately applies the law of the state whose\\n interest would be the more impaired if its law\\n were not applied.\\n\\nKearney v. Salomon Smith Barney, Inc., 137 P.3d 914, 922\\n(Cal. 2006) (internal quotation marks and citations omitted);\\nsee also Hairu Chen v. Los Angeles Truck Centers, LLC, No.\\nS240245, 2019 WL 3281346, at *3 (Cal. July 22, 2019).\\n\\n In making its choice-of-law determinations, the district\\ncourt relied heavily on the California Supreme Court’s\\ndecision in Sullivan v. Oracle Corp., 254 P.3d 237 (Cal.\\n2011), and the parties do not dispute that Sullivan provides\\nthe most helpful guidance for the choice-of-law questions\\nbefore us. In Sullivan, the California Supreme Court\\nanswered a certified question from this court regarding\\nwhether California overtime law applied to non-resident\\nemployees of a California corporation who worked primarily\\nin their home states of Colorado and Arizona, but also\\nworked in California (and several other states) for “entire\\ndays or weeks” at a time. Id. at 239, 243. Sullivan first\\nconcluded that as a matter of statutory construction,\\nCalifornia law applied to all work performed for days or\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 21\\n\\nweeks at a time within the state’s borders, regardless of\\nwhether it was performed by residents or non-residents. Id.\\nat 241–43. Next, Sullivan undertook California’s three-step\\ngovernmental interest analysis for choice-of-law questions.\\nId. at 244–47. At the first step of the analysis—whether the\\nrelevant laws differed—the court noted that California’s\\novertime law “clearly” differed from the laws of the\\nplaintiffs’ home states. Id. at 245.\\n\\n At the second step—whether a “true” conflict existed—\\nthe court held that the existence of a true conflict was\\n“doubtful, at best.” Id. The court explained that the second\\nstep involves examining “each jurisdiction’s interest in the\\napplication of its own law under the circumstances of the\\nparticular case,” noting that a court “may make [its] own\\ndetermination of the relevant policies and interests, without\\ntaking ‘evidence’ as such on the matter.” Id. (internal\\nquotation marks, alterations, and citations omitted). Sullivan\\nobserved that “California has, and has unambiguously\\nasserted, a strong interest in applying its overtime law to all\\nnonexempt workers, and all work performed, within its\\nborders.” Id. at 245. The court concluded that “neither\\nColorado nor Arizona has a legitimate interest in shielding\\nOracle from the requirements of California wage law as to\\nwork performed here.” Id. at 246.\\n\\n In so holding, the court rejected two specific arguments\\nadvanced by Oracle. First, Oracle contended that because\\nArizona and Colorado have workers’ compensation statutes\\nwith express extraterritorial application, those statutes\\nindicate an interest in extending the protection of their\\nemployment laws to their residents working outside the\\nstate. Id. Not so, Sullivan held. While “a state has such an\\ninterest, at least in the abstract, when the traveling, resident\\nemployee of a domestic employer would otherwise be left\\n\\n22 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nwithout the protection of another state’s law,” the states had\\n“expressed no interest in disabling their residents from\\nreceiving the full protection of California overtime law when\\nworking here, or in requiring their residents to work side-by-\\nside with California residents in California for lower pay.”\\nId.\\n\\n Second, Oracle argued that Arizona and Colorado “have\\nan interest in providing hospitable regulatory environments\\nfor their own businesses” and thus “also have an interest in\\nshielding their own businesses from more costly and\\nburdensome regulatory environments in other states.” Id.\\nRelying on principles of federalism, Sullivan dismissed this\\nargument. While “a state can properly choose to create a\\nbusiness-friendly environment within its own boundaries,”\\nthe federal Constitution does not require a state to substitute\\n“‘the conflicting statute of another state’” for its own laws\\nthat are “‘applicable to persons and events’” within that\\nstate. Id. (quoting Phillips Petrol. Co. v. Shutts, 472 U.S.\\n797, 822 (1985)). Nor does the Constitution “permit one\\nstate to project its regulatory regime into the jurisdiction of\\nanother state.” Id. (citing Healy v. Beer Insti., Inc., 491 U.S.\\n324, 336–37 (1989)).\\n\\n Finally, although Sullivan held that there was almost\\ncertainly no true conflict because neither Arizona nor\\nColorado had a “legitimate interest” in blocking the\\napplication of California law to the work performed in\\nCalifornia, the court nonetheless proceeded to the third step\\nof the analysis “for the sake of argument.” Id. at 246–47.\\nSullivan concluded that the analysis at the third step—\\ndetermining which state’s interest would be more impaired\\nif its policy were subordinated to the policy of the other\\nstate—yielded a straightforward answer:\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 23\\n\\n [T]o subordinate California’s interests to\\n those of Colorado and Arizona\\n unquestionably would bring about the greater\\n impairment. To permit nonresidents to work\\n in California without the protection of our\\n overtime law would completely sacrifice, as\\n to those employees, the state’s important\\n public policy goals of protecting health and\\n safety and preventing the evils associated\\n with overwork. Not to apply California law\\n would also encourage employers to substitute\\n lower paid temporary employees from other\\n states for California employees, thus\\n threatening California’s legitimate interest in\\n expanding the job market. By way of\\n comparison, not to apply the overtime laws of\\n Colorado and Arizona would impact those\\n states’ interests negligibly, or not at all . . .\\n Alternatively, viewing Colorado’s and\\n Arizona's overtime regimens as expressions\\n of a general interest in providing hospitable\\n regulatory environments to businesses within\\n their own boundaries, that interest is not\\n perceptibly impaired by requiring a\\n California employer to comply with\\n California overtime law for work performed\\n here.\\n\\nId. at 247 (citations omitted) (emphasis added).\\n\\n A.\\n\\n We first conclude that the district court did not err in\\nholding that under Sullivan, California law should apply to\\n\\n24 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nthe (b)(3) California class. 6 Although defendants correctly\\npoint out that Sullivan is not precisely analogous to the case\\nat hand, the two principal differences on which defendants\\nrely are unpersuasive. Specifically, defendants first rely on\\nthe fact that while Sullivan involved a California\\ncorporation, “most of the MLB Club Defendants with\\naffiliates in the California League are located outside\\nCalifornia.” But a close reading of Sullivan indicates that\\nCalifornia law should apply to the California class, even\\nthough many of the employers are not headquartered in\\nCalifornia. For example, Sullivan expressly contemplated\\nthat California’s overtime laws may not apply to non-\\nresident employees of an out-of-state business who enter\\nCalifornia only “temporarily during the course of a\\nworkday,” but contrasted such a scenario with employees\\nwho work in California for “entire days and weeks,” who are\\ncovered by California law. Id. at 243 (emphases, internal\\nquotation marks, and citations omitted).\\n\\n Similarly, Sullivan specifically left open the possibility\\nthat other California employment laws, such as pay stub\\nrequirements, may not apply to non-resident employees of\\nout-of-state employers—with the clear implication that\\novertime laws would apply to such employees. See id. at\\n243–44. Likewise, far from limiting its holding only to non-\\nresident employees of in-state employers, Sullivan merely\\nemphasized that employees of in-state employers would\\nespecially be covered by California law. See id. at 243.\\n\\n 6\\n Contrary to the dissent’s criticism, Dissent at 74, we do not\\nshortcut the governmental interest analysis. As we explain in the text,\\nwe believe that Sullivan mandates application of California law to the\\nCalifornia class. Rather than repeating Sullivan’s choice of law analysis,\\nwe focus on several additional considerations that further support our\\ndecision to affirm the district court’s reliance on Sullivan.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 25\\n\\n Second, defendants characterize Sullivan as resting on\\nthe court’s determination that “neither Arizona nor Colorado\\n. . . has asserted an interest in regulating overtime work\\nperformed in other states.” Defendants argue that here, by\\ncontrast, “numerous” states have a competing interest in\\nregulating work performed in California. But defendants\\nmisread Sullivan by erroneously presuming that its\\nconclusion at the third step—that subordinating\\n“California’s interests to those of Colorado and Arizona\\nunquestionably would bring about the greater\\nimpairment”—hinged entirely on whether Arizona or\\nColorado law had asserted an interest in extraterritorial\\napplication of their wage laws. Id. at 247. It is certainly\\naccurate to say that Sullivan’s holding was influenced by the\\nfact that neither Arizona nor Colorado law purported to\\napply extraterritorially. Yet the court’s discussion at step\\nthree cannot fairly be read to support the argument that\\nCalifornia’s “strong interest in applying its overtime law to\\n. . . all work performed within its borders,” id. at 245, would\\nsuddenly become the lesser-impaired interest in the event\\nanother state expressed a clear interest in applying its wage\\nlaws to work performed in California. Rather, Sullivan\\nstrongly indicates that California’s interest in applying its\\nlaws to work performed within its borders for days or weeks\\nat a time would reign supreme regardless of whether another\\nstate expressed an interest in applying its own wage laws\\ninstead of California’s.\\n\\n26 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n Although we read Sullivan as clearly mandating the\\napplication of California law to the California class, two\\nadditional considerations support our conclusion today. 7\\n\\n First, because the district court found that plaintiffs had\\nmet their burden of showing that California law could\\nconstitutionally be applied—a determination defendants do\\nnot contest on appeal—the burden shifted to defendants “to\\ndemonstrate ‘that foreign law, rather than California law,\\nshould apply to class claims.’” Mazza, 666 F.3d at 590\\n(quoting Wash. Mut. Bank v. Superior Court, 15 P.3d 1071,\\n1081 (Cal. 2001)). The district court held that defendants\\nfailed to meet this burden, because they had “not gone\\nbeyond speculating in a general manner that the claims of\\nsome members of the putative California Class might be\\nsubject to the law of another state and that the interests of\\nanother state might be more impaired by application of\\nCalifornia law.”\\n\\n Defendants specifically point to one of the named\\nplaintiffs—Mitch Hilligoss—as an example of the alleged\\n“need to conduct choice of law inquiries as to every member\\nof the California class.” The district court found this\\nexample unpersuasive for several reasons, and we agree.\\nThe defendants argued that Illinois law should apply to\\nHilligoss’ work in California because the time he spent in\\nCalifornia was a small proportion of his overall career\\n(around two months out of a six-year career). The district\\ncourt, however, correctly read Sullivan as indicating that\\nCalifornia law should nonetheless apply to Hilligoss’\\nCalifornia work. Indeed, the proportion of time the non-\\n\\n 7\\n Moreover, as the dissent acknowledges, the California Supreme\\nCourt has expressed a strong interest in regulating wage and hour claims\\nwithin its borders. Dissent at 77.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 27\\n\\nresident employees in Sullivan worked in California was\\nquite small (and in one case, even less than the proportion of\\nHilligoss’ career spent in California): during the relevant\\nthree-year period, one worked 20 days, another 74 days, and\\nthe third 110 days. 254 P.3d at 239. Put differently, the\\nemployees in Sullivan worked in California approximately\\n1.8%, 6.7%, and 10% of the time, respectively. Id. What\\nmattered in Sullivan—and what matters here—is that when\\nthe employees worked in California, they did so for “entire\\ndays or weeks” at a time. Id. at 243.\\n\\n Second, practical considerations strongly support\\napplying California law to work performed in California, at\\nleast as a general rule; to hold otherwise “would lead to\\nbizarre and untenable results.” See Brief for Professors Peter\\nHay and Patrick J. Borchers, Dkt. No. 21, at 12–13\\n(hereinafter “Professors’ Amicus Brief”). If the law of the\\nstate in which work is performed is not the law that generally\\napplies, employers and employees alike would be subjected\\nto an unworkable scheme. Employers would be required to\\nproperly ascertain the residency status—itself not\\nnecessarily an easy task, as any student or seasonal worker\\ncould attest—of each of its employees. For every non-\\nresident employee, employers would then have to determine\\nwhether the wage laws of that employee’s state of residence\\napply extraterritorially, and then come up with different rules\\nfor each of its employees according to their state of residence\\nand any extraterritorial application of their home state’s\\nlaws. This would mean that at a single worksite, employees\\nworking side-by-side in the same position would not only be\\nowed vastly different minimum wages, but also that an\\nemployer would need to set different rules for meal and rest\\nbreaks for different employees, and so on and so forth. It\\ncannot be in any state’s legitimate “interest” to foist such an\\n\\n28 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nadministrative nightmare upon both employers and\\nemployees.\\n\\n Such a scenario would also result in an enormous\\ncompetitive advantage—or disadvantage—for prospective\\nemployees based solely on their state of residency.\\nEmployers would be incentivized to hire residents of states\\nwith low minimum wages and otherwise employer-friendly\\nwage laws, while residents of states with higher minimum\\nwages and more protective employment laws would\\nsuddenly be far less appealing. Amici Professors Hays and\\nBorchers persuasively point out that as defendants would\\nhave it, a college student still domiciled in Seattle while\\nattending a Nebraska university would have to be paid $15\\nper hour at a part-time job in Nebraska, “nearly double\\nNebraska’s minimum wage of $8 per hour.” Professors’\\nAmicus Brief at 13. This, of course, would put the student\\nat a crushing disadvantage; what rational employer would\\nhire her?\\n\\n Moreover, given the administrative cost involved in\\nattempting to comply with a patchwork of multiple states’\\nwage laws at a single workplace, some employers might\\ninstead choose to stick to hiring only resident employees, or\\nperhaps only non-resident employees from a particular state\\n(presumably one with a low minimum wage and minimally\\nprotective employment laws). 8\\n\\n\\n\\n 8\\n The California class consisted of those players who participated in\\nthe California League, which plays games exclusively within California\\nduring the championship season. The Arizona and Florida classes\\nconsisted of those who performed during spring training, extended\\nspring training, and the instructional leagues in those states. Thus, the\\ndissent’s fear that employers will be required to research applicable state\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 29\\n\\n We do not foreclose the possibility that there could be\\nsome circumstances in which a proper application of\\nCalifornia’s choice-of-law rules might lead to the\\napplication of another state’s wage and hour laws to work\\nperformed in California. Nor do we create a per se rule or\\nan unrebuttable presumption. We hold only that, given the\\nabove considerations, we are more than satisfied that the\\ndistrict court did not err in concluding that under Sullivan,\\nCalifornia law applies to the California class.\\n\\n B.\\n\\n We next address whether the district court erred in\\ndetermining that choice-of-law considerations defeated\\npredominance and adequacy for the proposed Arizona and\\nFlorida Rule 23(b)(3) classes, and conclude that the district\\ncourt’s determination must be reversed. Our conclusion is\\nanimated in part by several of the considerations outlined\\nabove, which apply with equal force to the Arizona and\\nFlorida classes. Moreover, the aforementioned enormous\\npractical implications of a contrary holding would be just as\\nproblematic and unworkable in Arizona and Florida as in\\nCalifornia.\\n\\n 1.\\n\\n With those considerations in mind, we apply California’s\\nthree-step governmental interest analysis, and conclude that\\nArizona law should apply to the work performed in Arizona,\\nand Florida law to the work performed in Florida. At the\\nfirst step, we agree with defendants that the differences in\\nstate law are “material,” meaning that “they make a\\n\\nlaws whenever an employee crosses state lines is overstated. Dissent at\\n84.\\n\\n30 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\ndifference in this litigation.” Mazza, 666 F.3d at 590. For\\nexample, some states have more expansive definitions of\\n“work,” others have differing available defenses, and we\\nhave previously held that the elements for a quantum meruit\\nclaim—alleged in both the Arizona and Florida classes—\\n“vary materially from state to state.” Id. at 591 (citing\\nCandace S. Kovacic, A Proposal to Simplify Quantum\\nMeruit Litigation, 35 Am. U.L. Rev. 547, 558–60 (1986)).\\n\\n 2.\\n\\n “Because the relevant laws differ,” we must “next\\nexamine each jurisdiction’s interest in the application of its\\nown law under the circumstances of the particular case to\\ndetermine whether a true conflict exists.” Sullivan, 254 P.3d\\nat 245 (alteration, internal quotation marks, and citation\\nomitted). We are not persuaded, as defendants contend, that\\na “true” conflict exists.\\n\\n First, under California’s choice-of-law principles, “a\\njurisdiction ordinarily has the predominant interest in\\nregulating conduct that occurs within its borders.” 9 Mazza,\\n666 F.3d at 592 (internal quotation marks omitted) (quoting\\nMcCann v. Foster Wheeler LLC, 225 P.3d 516, 534 (Cal.\\n2010)). The dissent contends that “California has long\\nrejected” this approach. Dissent at 71. In noting these\\nprinciples, we do not ignore the evolution of California’s\\nchoice of law doctrine. We recognize that the California\\nSupreme Court “renounced the prior rule, adhered to by\\ncourts for many years, that in tort actions the law of the place\\n\\n 9\\n Wage and hour laws are typically categorized as “conduct-\\nregulating,” as opposed to “loss-allocating.” See Professors’ Amicus\\nBrief at 15–16 (citing Hay, Borchers & Symeonides, Conflict of Laws\\n874–78 (5th ed. 2010)).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 31\\n\\nof the wrong was the applicable law in a California forum\\nregardless of the issues before the court” when it adopted the\\ngovernmental interest approach. Hurtado v. Superior Court,\\n522 P.2d 666 (Cal. 1974). Yet the California Supreme Court\\nhas acknowledged that while it “no longer follows the old\\nchoice-of-law rule that generally called for application of the\\nlaw of the jurisdiction in which a defendant’s allegedly\\ntortious conduct occurred without regard to the nature of the\\nissue that was before the court . . . California choice-of-law\\ncases nonetheless continue to recognize that a jurisdiction\\nordinarily has the predominant interest in regulating conduct\\nthat occurs within its borders.” McCann 225 P.3d at 534\\n(internal quotation marks and citation omitted) (emphasis in\\noriginal).\\n\\n Thus, when conducting the governmental interest\\nanalysis, we must also recognize that a state ordinarily has\\nthe predominant interest in regulating conduct within its\\nborders. We draw this conclusion not from California’s\\ninterest in regulating conduct within its own borders, but\\nfrom California’s choice-of-law principles. 10 Thus these\\n\\n 10\\n See e.g., McCann, 225 P.3d at 534, 537 (recognizing that although\\nCalifornia no longer uniformly applied the law of the jurisdiction in\\nwhich the allegedly tortious conduct occurred, Oklahoma’s interests\\n“would be more impaired if its law were not applied” as the plaintiff’s\\nexposure to asbestos occurred in Oklahoma); Reich v. Purcell, 67 432\\nP.2d 727, 730 (Cal. 1967) (“Missouri is concerned with conduct within\\nher borders and as to such conduct she has the predominant interest of\\nthe states involved.”); Castro v. Budget Rent-A-Car Sys., Inc., 65 Cal.\\nRptr. 3d 430, 442 (Cal. Ct. App. 2007) (“The accident and Castro's injury\\noccurred within Alabama's borders, thus giving Alabama a presumptive\\ninterest in controlling the conduct of those persons who use its roadways,\\nabsent some other compelling interest to be served by applying\\nCalifornia law.”); Hernandez v. Burger, 162 Cal. Rptr. 564, 568 (Cal. Ct.\\nApp. 1980) (“It is true that the place of the wrong is no longer treated as\\na controlling factor where application of the law of another jurisdiction\\n\\n32 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nprinciples are not limited to the California class but apply to\\nthe Florida and Arizona classes as well. See Mazza, 666 F.3d\\nat 593 (“The district court did not adequately recognize that\\neach foreign state [not just California] has an interest in\\napplying its law to transactions within its borders.”). The\\ndistrict court erred in ignoring these principles as a starting\\npoint, instead faulting plaintiffs for not addressing “in detail\\nthe interests of either Arizona or Florida in applying their\\nlaw” and focusing on the absence of Florida or Arizona cases\\nakin to Sullivan—despite the strong indications that Arizona\\nand Florida have the “predominant interest” in applying their\\nlaws to work performed within their state. See Mazza, 666\\nF.3d at 592.\\n\\n Second, Sullivan relied on several different\\nconsiderations to arrive at its conclusion that the existence\\nof a true conflict was “doubtful, at best”: (1) the states in\\nwhich the employees resided did not express an intent to\\napply their laws extraterritorially; (2) the employees’ states\\nof residence did not have a “legitimate interest” in shielding\\nan employer from California’s wage laws as to work\\nperformed in California; and (3) federalism and due process\\nmade extraterritorial reach doubtful under the\\ncircumstances. See 254 P.3d at 245–47. Although\\ndefendants vigorously argue that the first of those rationales\\nis inapplicable here—as discussed in greater detail below—\\nat a minimum, the second and third rationales do apply, and\\nweigh against the existence of a true conflict.\\n\\n\\nhaving a connection with the accident will serve a legitimate interest or\\npolicy of the other jurisdiction. However, the situs of the injury remains\\na relevant consideration.”); Cable v. Sahara Tahoe Corp., 155 Cal. Rptr.\\n770, 777 (Cal. Ct. App. 1979) (“The state with the ‘predominant’ interest\\nin controlling conduct normally is the state in which such conduct occurs\\nand is most likely to cause injury.”).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 33\\n\\n As to the first rationale, both defendants and the dissent\\ncontend that several states have expressed an interest in\\napplying their wage and hour laws to work performed\\noutside the state. In support of their position, they cite to a\\nhandful of cases where courts (largely district courts or\\nintermediate state courts, with the exceptions of West\\nVirginia and Washington) 11 have applied one state’s wage\\nlaws to work performed at least partially in another state.\\nFor several reasons, we are unpersuaded by defendants’\\narguments. For one, we read Sullivan as indicating that\\nunder California’s choice-of-law principles, a state has a\\nlegitimate interest in applying its wage laws extraterritorially\\nonly in two limited circumstances, neither of which apply\\nhere: one, when a state’s resident employee of that state’s\\nresident employer leaves the state “temporarily during the\\ncourse of the normal workday,” and two, “when the\\ntraveling, resident employee of a domestic employer would\\notherwise be left without the protection of another state’s\\n\\n\\n 11\\n In New v. Tac & C Energy, Inc., 355 S.E.2d 629 (W. Va. 1987),\\nthe West Virginia Supreme Court of Appeals applied its own conflict-\\nof-laws principles—relying on the Restatement (Second) of Conflicts\\n§ 196—to conclude that while there was a presumption that the law of\\nthe state where services were rendered applies, the presumption could be\\novercome by showing that another state had a “more significant\\nrelationship to the transaction and the parties.” Id. at 631. Where all\\nparties were residents of West Virginia, the employment contract was\\nmade and partially performed in West Virginia, and the plaintiffs were\\nonly in Kentucky for the duration of the work, the court concluded that\\nthe presumption was overcome and that West Virginia “had the more\\nsignificant connection to the employment relationship.” Id. California’s\\nchoice-of-law test, of course, does not utilize the “more significant\\nrelationship” test for choice-of-law questions in the wage and hour\\ncontext. See Sullivan, 254 P.3d at 244. New is therefore unpersuasive\\nhere. We discuss the Washington Supreme Court case below.\\n\\n34 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nlaw.” Id. at 242, 246 (citations, internal quotation marks,\\nand alterations omitted).\\n\\n Moreover, the cases on which defendants and the dissent\\nrely are, in large part, both factually and procedurally\\ninapposite to the circumstances of this case. 12 For example,\\ndefendants rely heavily on Bostain v. Food Exp., Inc., 153\\nP.3d 846, 851 (Wash. 2007) to argue that Washington has an\\ninterest in applying its wage laws extraterritorially. As the\\nCalifornia Supreme Court held in Sullivan, however, Bostain\\n“says nothing about a case such as this”—that is, a case\\nwhich (1) involves work performed entirely in one state, and\\n(2) presents an unavoidable conflict-of-laws issue. 254 P.3d\\nat 243. In Bostain, by contrast, either Washington law\\napplied to the work performed in both Washington and other\\nstates, or else no state’s law applied. Id. at 243, 246.\\nSignificantly, Bostain interpreted an overtime statute that\\nspecifically delineated the circumstances under which its\\nprovisions would apply to interstate truck drivers; as the\\nWashington Supreme Court noted, interstate truck drivers by\\ndefinition perform some of their work out of state. 153 P.3d\\nat 848–51. The statute at issue in Bostain did “not limit the\\nrequirement for overtime pay to hours worked” within the\\nstate’s borders. Id. at 851. Similarly, here, defendants point\\n\\n 12\\n Defendants’ repeated citation to Gonyou v. Tri-Wire Eng'g Sols.,\\nInc., 717 F. Supp. 2d 152 (D. Mass. 2010) is illustrative. In Gonyou, a\\nMassachusetts resident employee of a Massachusetts employer worked\\nlargely, although not entirely, in Connecticut. Id. at 153–54. The\\ndefendant filed a motion to dismiss on the ground that the Massachusetts\\novertime statute did not apply to work performed in Connecticut. Id. at\\n154–55. The court denied the motion but emphasized the limited nature\\nof its ruling: “As is eminently clear, this is a motion to dismiss and this\\nruling is strictly limited to the facts and circumstances of this case and\\nthis motion.” Id. at 155.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 35\\n\\nto no state statutes potentially applicable to the Arizona and\\nFlorida class members that limit their application to work\\nperformed within the state.\\n\\n 3.\\n\\n Although the existence of a “true” conflict is\\nquestionable, we need not decide whether a true conflict\\nexists, as the third step of California’s governmental interest\\ntest yields a clear answer: the laws of Arizona and Florida\\nshould apply to the work performed wholly within their\\nrespective boundaries. 13 See Sullivan, 254 F.3d at 247. As\\nthe California Supreme Court has explained the step three\\ninquiry:\\n\\n [T]he court does not “weigh” the conflicting\\n governmental interests in the sense of\\n determining which conflicting law\\n manifested the “better” or the “worthier”\\n\\n 13\\n Furthermore, in many of the cases cited by the dissent to\\ndemonstrate that some states have asserted an interest in applying their\\nwage and hour laws outside of their borders, courts have looked closely\\nat where the relevant work is performed. See e.g., Pierre v. Gts\\nHoldings, Inc., No. 15 CIV. 143 (PAC), 2015 WL 7736552, at *3–*4\\n(S.D.N.Y. Nov. 30, 2015) (concluding that New York labor laws apply\\nbecause, among other things, the majority of the plaintiff’s chauffeured\\nrides were conducted in New York); Baxi v. Ennis Knupp & Assocs.,\\nInc., No. 10-CV-6346, 2011 WL 3898034, at *14 (N.D. Ill. Sept. 2,\\n2011) (denying a motion to dismiss Illinois labor law claims because the\\nplaintiff, a foreign resident, performed some work in Illinois); Friedrich\\nv. U.S. Computer Sys., Inc., No. CIV. A. 90-1615, 1996 WL 32888, at\\n*8 (E.D. Pa. Jan. 22, 1996) (concluding that a Pennsylvania labor law\\napplies to the plaintiffs because the jury found the plaintiffs were “based\\nin Pennsylvania,” even if they were not residents of the state); Dow, 989\\nN.E.2d at 914 (concluding that Massachusetts law applied because,\\ngiven the nature of the plaintiff’s work, the work “sensibly may be\\nviewed as having ‘occurred’ in Massachusetts”).\\n\\n36 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n social policy on the specific issue. An\\n attempted balancing of conflicting state\\n policies in that sense is difficult to justify in\\n the context of a federal system in which,\\n within constitutional limits, states are\\n empowered to mold their policies as they\\n wish. Instead, the process can accurately be\\n described as a problem of allocating domains\\n of law-making power in multi-state\\n contexts—by determining the appropriate\\n limitations on the reach of state policies—as\\n distinguished from evaluating the wisdom of\\n those policies. Emphasis is placed on the\\n appropriate scope of conflicting state policies\\n rather than on the “quality” of those policies.\\n\\nMcCann, 225 P.3d at 533–34 (alterations and citations\\nomitted) (emphasis added).\\n\\n As discussed above, in Mazza, we faithfully applied the\\nprinciple under California’s choice-of-law jurisprudence\\nthat “a jurisdiction ordinarily has the predominant interest in\\nregulating conduct that occurs within its borders.” Mazza,\\n666 F.3d at 592 (internal quotation marks omitted) (quoting\\nMcCann, 225 P.3d at 534). We thus had no trouble\\nconcluding at step three that “each class member’s consumer\\nprotection claim should be governed by the consumer\\nprotection laws of the jurisdiction in which the transaction\\ntook place.” Id. at 594. Notably, we reached this conclusion\\nwithout specifically inquiring into the interests potentially\\nexpressed by any state’s statutory language or case law.\\nRather, our conclusion was dictated by the principle,\\ndiscussed above, that a jurisdiction ordinarily has the\\npredominant interest in regulating conduct within its own\\nborders. Id. at 591–92 (first citing State Farm Mut. Auto.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 37\\n\\nIns. Co. v. Campbell, 538 U.S. 408, 422 (2003); and then\\nciting McCann, 225 P.3d at 534).\\n\\n Moreover, in Sullivan, the court concluded that to\\nsubordinate California’s ability to apply its own wage laws\\nto work performed within the state would “unquestionably”\\ncause greater impairment to California than to the states that\\nmight seek to apply their wage laws to work performed by\\ntheir residents within California. 254 P.3d at 247. As\\ndescribed previously, while this holding was influenced by\\nthe absence of an expression of interest by Arizona or\\nColorado in applying their laws extraterritorially, it did not\\nrise or fall on that ground. See id. at 244–47. And although\\ndefendants point to a handful of cases that have entertained\\nthe potential application of one state’s wage laws to work\\nperformed in another state, they have not pointed to a single\\nstate with a potentially-applicable statute that expresses a\\nclear interest in applying to work performed wholly outside\\nthe state.\\n\\n But even if defendants were able to identify any states\\nthat had unambiguously expressed an interest in applying\\ntheir wage laws to work performed entirely in another state,\\nSullivan strongly militates against concluding that such an\\nexpression of interest would be adequate to overcome the\\nprinciple that the state in which the conduct at issue occurs\\nhas the “predominant interest” in applying their own law.\\nSee Mazza, 666 F.3d at 592–94; Sullivan, 254 P.3d at 245–\\n47. Forcing Arizona or Florida to allow the application of\\nother states’ wage laws in this case would be just as\\ndestructive to the balance Arizona and Florida have struck\\nbetween protecting workers and fostering a hospitable\\nbusiness environment within their states as allowing the\\napplication of Colorado or Arizona law in Sullivan would\\nhave been to the balance California struck between those\\n\\n38 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nsame interests. See Sullivan, 254 P.3d at 246–47. The\\ndistrict court fundamentally misunderstood the proper\\napplication of California’s choice-of-law principles—which,\\nwhen correctly applied, indicate that Arizona law should\\ngovern the Arizona class, and Florida law the Florida class.\\n\\n C.\\n\\n We next address whether the district court erred in\\nrefusing to certify a Rule 23(b)(2) class for unpaid work at\\ndefendants’ training facilities in Arizona and Florida on the\\nsole basis that choice-of-law issues undermined\\n“cohesiveness” and therefore made injunctive and\\ndeclaratory relief inappropriate. Because the district court’s\\nerrors in its choice-of-law analysis relating to the proposed\\nArizona and Florida Rule 23(b)(3) classes apply equally to\\nits refusal to certify the proposed Rule 23(b)(2) class, we\\nalso reverse the denial of the (b)(2) class.\\n\\n We further hold that the district court erred in imposing\\na “cohesiveness” requirement for the proposed Rule 23(b)(2)\\nclass. Although we have never explicitly addressed whether\\n“cohesiveness” is required under Rule 23(b)(2), courts that\\nhave imposed such a test treat it similarly to Rule 23(b)(3)’s\\npredominance inquiry14—something we have previously\\nrejected in no uncertain terms. See Walters v. Reno, 145 F.3d\\n1032, 1047 (9th Cir. 1998) (“[W]ith respect to 23(b)(2) in\\nparticular, the government’s dogged focus on the factual\\ndifferences among the class members appears to\\n\\n 14\\n The similarity between “cohesiveness” and predominance is\\nperhaps unsurprising, given that the Supreme Court described the\\npredominance inquiry under 23(b)(3) as testing whether a class is\\n“sufficiently cohesive to warrant adjudication by representation.”\\nAmchem Products, Inc. v. Windsor, 521 U.S. 591, 623 (1997) (emphasis\\nadded).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 39\\n\\ndemonstrate a fundamental misunderstanding of the rule.\\nAlthough common issues must predominate for class\\ncertification under Rule 23(b)(3), no such requirement exists\\nunder 23(b)(2).”); see also 2 Newberg on Class Actions\\n§ 4:34 (5th ed. 2012) (describing similarity between\\npredominance under Rule 23(b)(3) and “cohesiveness”\\nunder Rule 23(b)(2) in courts that have adopted it). We\\ntherefore remand for the district court to consider anew\\nwhether to certify the proposed Rule 23(b)(2) class. 15\\n\\n III.\\n\\n Having addressed the impact of choice-of-law questions,\\nwe turn to the issue next up at bat: whether the district court\\nerred in concluding that plaintiffs could meet the\\npredominance requirement for the proposed California,\\nFlorida, and Arizona (b)(3) classes through a combination of\\nrepresentative evidence and application of the “continuous\\nworkday” rule.\\n\\n Rule 23(b)(3)’s predominance requirement requires\\ncourts to ask “whether the common, aggregation-enabling\\nissues in the case are more prevalent or important than the\\nnon-common, aggregation-defeating, individual issues.”\\nTyson Foods, Inc. v. Bouaphakeo, 136 S. Ct. 1036, 1045\\n(2016) (quoting 2 Newberg on Class Actions § 4:49 (5th ed.\\n2012)). A proposed (b)(3) class may be certified as long as\\n“one or more of the central issues in the action are common\\nto the class and can be said to predominate . . . even though\\nother important matters will have to be tried separately, such\\n 15\\n While the parties advanced numerous arguments regarding (b)(2)\\ncertification in the district court, and advance similar arguments—along\\nwith a few new ones—before us, we decline to pass on those other issues\\nin the first instance. See Stockwell, 749 F.3d at 1113, 1116–17; Davis v.\\nNordstrom, Inc., 755 F.3d 1089, 1094–95 (9th Cir. 2014).\\n\\n40 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nas damages or some affirmative defenses peculiar to some\\nindividual class members.” Id. (quoting 7AA C. Wright, A.\\nMiller, & M. Kane, Federal Practice and Procedure § 1778\\n(3d ed. 2005)).\\n\\n “[P]redominance in employment cases is rarely defeated\\non the grounds of differences among employees so long as\\nliability arises from a common practice or policy of an\\nemployer.” 7 Newberg on Class Actions § 23:33 (5th ed.\\n2012). Although the existence of blanket corporate policies\\nis not a guarantee that predominance will be satisfied, such\\npolicies “often bear heavily on questions of predominance\\nand superiority.” In re Wells Fargo Home Mortg. Overtime\\nPay Litig., 571 F.3d 953, 958 (9th Cir. 2009).\\n\\n Whether the district court was correct in concluding that\\nplaintiffs had satisfied the predominance requirement hinges\\non the application of two longstanding wage-and-hour\\ndoctrines to this case: first, the burden-shifting framework\\ninitially set forth in the Supreme Court’s seminal decision in\\nAnderson v. Mt. Clemens, 328 U.S. 680 (1946) and recently\\nexpanded upon in Tyson Foods v. Bouaphakeo, 136 S. Ct.\\n1036 (2016); and second, the so-called “continuous\\nworkday” rule. We address each of these doctrines and their\\napplication to this case in turn.\\n\\n A.\\n\\n In Mt. Clemens, the Supreme Court acknowledged the\\ndifficult bind that employees frequently confronted when\\nseeking to bring wage-and-hour claims against their\\nemployers: if their employers had failed to maintain proper\\ntimekeeping records, proving the hours of uncompensated\\nwork often posed “an impossible hurdle for the employee.”\\n328 U.S. at 687. Mt. Clemens held that such a catch-22 was\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 41\\n\\nnot in line with “the remedial nature of [the FLSA] 16 and the\\ngreat public policy which it embodies.” Id. After all, “[s]uch\\na result would place a premium on an employer’s failure to\\nkeep proper records in conformity with his statutory duty; it\\nwould allow the employer to keep the benefits of an\\nemployee’s labors without paying due compensation.” Id.\\n\\n To address this problem, Mt. Clemens established its\\nlandmark burden-shifting framework for actions in which\\nthe employer has kept inaccurate or inadequate records: if an\\nemployee “proves that he has in fact performed work for\\nwhich he was improperly compensated” and “produces\\nsufficient evidence to show the amount and extent of that\\nwork as a matter of just and reasonable inference,” then the\\nburden “shifts to the employer to come forward with\\nevidence of the precise amount of work performed or with\\nevidence to negative the reasonableness of the inference to\\nbe drawn from the employee’s evidence.” Id. at 687–88. If\\nthe employer does not rebut the employee’s evidence,\\ndamages may then be awarded to the employee, “even\\nthough the result be only approximate.” Id. at 688.\\n\\n Mt. Clemens explicitly rejected the notion that allowing\\napproximate damages in such situations would be unfair due\\nto its speculative and imprecise nature or because employers\\nsometimes make good-faith mistakes over what constitutes\\ncompensable “work”:\\n\\n\\n\\n 16\\n Although Mt. Clemens was decided under the FLSA, its holding\\nhas been consistently applied in the context of state wage-and-hour\\nclaims as well. See, e.g., Tyson, 136 S. Ct. at 1045–48; Torres v. Mercer\\nCanyons Inc., 835 F.3d 1125, 1140 (9th Cir. 2016); Hernandez v.\\nMendoza, 245 Cal. Rptr. 36, 39–40 (Cal. Ct. App. 1988) (applying Mt.\\nClemens to claims under California wage and hour law).\\n\\n42 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n The employer cannot be heard to complain\\n that the damages lack the exactness and\\n precision of measurement that would be\\n possible had he kept records in accordance\\n with the [statutory] requirements . . . And\\n even where the lack of accurate records\\n grows out of a bona fide mistake as to\\n whether certain activities or non-activities\\n constitute work, the employer, having\\n received the benefits of such work, cannot\\n object to the payment for the work on the\\n most accurate basis possible under the\\n circumstances . . . In such a case it would be\\n a perversion of fundamental principles of\\n justice to deny all relief to the injured person,\\n and thereby relieve the wrongdoer from\\n making any amend for his acts.\\n\\nId. (internal quotation marks and citation omitted).\\n\\n Seventy years after Mt. Clemens addressed the use of\\nrepresentative evidence at the trial stage to show damages,\\nTyson extended Mt. Clemens’ holding to answer two\\nimportant questions: whether representative evidence may\\nbe used at the class certification stage, and whether\\nrepresentative evidence may also be used to establish\\nliability in addition to damages. In Tyson, employees who\\nworked in more than 400 jobs across three departments at a\\nmeat processing plant sued under the FLSA and an Iowa\\nwage law, alleging that Tyson had not paid them overtime\\nfor time they spent donning and doffing protective gear; the\\nemployees also sought certification of a Rule 23 class and a\\nFLSA collective action. 136 S. Ct. at 1041–42.\\n\\n The district court certified the class and collective\\nactions, rejecting Tyson’s arguments that the claims were\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 43\\n\\ninappropriate for resolution on a classwide and collective\\nbasis due to the dissimilarity in the types of protective gear\\nworn and the variations in time spent donning and doffing\\nthat gear. Id. at 1042–43. Because Tyson had not kept\\nrecords of the donning and doffing time, plaintiffs relied on\\nrepresentative evidence to demonstrate both liability17 and\\ndamages: employee testimony, video recordings, and—most\\nsignificantly—an expert study that computed an estimated\\namount of time spent donning and doffing for each of the\\nthree departments based on hundreds of video observations.\\nId. at 1043. Although the expert estimated that the time\\nspent donning and doffing was 18 minutes per day for two\\nof the departments and 21.25 minutes for the other, id., the\\nsurvey data showed a great deal of variation in how long it\\ntook individual employees to don and doff. Id. at 1055\\n(Thomas, J., dissenting). Specifically, the time spent\\ndonning ranged from around thirty seconds to more than ten\\nminutes, and the time doffing varied from under two minutes\\nto over nine minutes. Id. After a jury verdict in the\\nemployees’ favor (albeit one that awarded less than half of\\nthe damages recommended by the employees’ expert based\\non the survey data), Tyson moved to decertify the class and\\nset aside the jury verdict, arguing that this variance made\\nclass and collective certification inappropriate. Id. at 1044–\\n\\n\\n\\n 17\\n Because the employees brought only overtime claims (as opposed\\nto minimum wage or other wage claims), “each employee had to show\\nhe or she worked more than 40 hours a week, inclusive of time spent\\ndonning and doffing, in order to recover.” Tyson, 136 S. Ct. at 1043.\\nThat the majority permitted the use of representative evidence to\\nestablish “an otherwise uncertain element of liability”—i.e., whether\\nclass members worked more than 40 hours per week—was one of the\\nkey bases for Justice Thomas’s vigorous dissent. See id. at 1057–59\\n(Thomas, J., dissenting).\\n\\n44 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n45. The district court denied the motion, and the Eighth\\nCircuit affirmed. Id.\\n\\n Tyson sought certiorari on the grounds that using\\nrepresentative evidence “manufactures predominance by\\nassuming away the very differences that make the case\\ninappropriate for classwide resolution,” “absolves each\\nemployee of the responsibility to prove personal injury,” and\\nstrips the employer of their ability to “litigate its defenses to\\nindividual claims.” Id. at 1046. Rejecting these arguments,\\nthe Supreme Court affirmed the class and collective\\ncertifications. Id. at 1046–47. Because of Tyson’s\\ndereliction of their recordkeeping duties, the employees\\nwere entitled to “introduce a representative sample to fill an\\nevidentiary gap created by the employer's failure to keep\\nadequate records.” Id. at 1047. The Court held that if the\\nrepresentative sample introduced were admissible and\\n“could have sustained a reasonable jury finding as to hours\\nworked in each employee's individual action, that sample is\\na permissible means of establishing the employees' hours\\nworked in a class action.” Id. at 1046–47.\\n\\n Stated another way, Tyson concluded that even where\\n“reasonable minds may differ” about whether representative\\nevidence is sufficiently probative of the requirements for\\nliability for a particular cause of action—in Tyson, whether\\nit was probative of the “time actually worked by each\\nemployee”—that question is to be resolved by the jury, not\\nat the class certification stage. Id. at 1049 (“The District\\nCourt could have denied class certification on this ground\\n[whether the representative evidence was “probative as to\\nthe time actually worked by each employee”] only if it\\nconcluded that no reasonable juror could have believed that\\nthe employees spent roughly equal time donning and\\ndoffing.”) (emphasis added). If the proffered representative\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 45\\n\\nevidence, however, were “statistically inadequate or based\\non implausible assumptions,” it “could not lead to a fair or\\naccurate estimate of the uncompensated hours an employee\\nhas worked.” Id. at 1048–49. But where the evidence is\\nadmissible—for expert evidence, using the Daubert\\nstandard—then the “no reasonable juror” standard at the\\nclass certification stage applies. See id. at 1049.\\n\\n B.\\n\\n Having established the parameters of when\\nrepresentative evidence may be used at the class certification\\nstage, we address the second significant wage-and-hour\\ndoctrine relevant to this case: the “continuous workday”\\nrule. The rule was first promulgated by the Department of\\nLabor (DOL) consistent with the Supreme Court’s decisions\\ninterpreting the FLSA prior to the enactment of the Portal-\\nto-Portal Act 18 in 1947. IBP, Inc. v. Alvarez, 546 U.S. 21,\\n27–28 (2005). It presumes that once the beginning of the\\n\\n 18\\n In response to what Congress perceived as excessively expansive\\njudicial interpretations of what constitutes compensable work under the\\nFLSA, IBP, 546 U.S. at 27–28, it passed the Portal-to-Portal Act to\\nexempt certain activities as compensable under FLSA:\\n\\n “(1) walking, riding, or traveling to and from the actual\\n place of performance of the principal activity or\\n activities which such employee is employed to\\n perform, and\\n\\n (2) activities which are preliminary to or postliminary\\n to said principal activity or activities,\\n\\nwhich occur either prior to the time on any particular workday at which\\nsuch employee commences, or subsequent to the time on any particular\\nworkday at which he ceases, such principal activity or activities.” 61\\nStat. 86–87 (codified at 29 U.S.C. § 254(a)).\\n\\n46 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nworkday is triggered, an employee performs compensable\\nwork throughout the rest of the day until the employee\\ncompletes their last principal activity (or the last activity\\nwhich is “integral and indispensable” to the employee’s\\nprincipal activities)—whether or not the employee actually\\nengages in work throughout that entire period. See id. at 28,\\n32–37; see also Alvarez v. IBP, Inc., 339 F.3d 894, 907 (9th\\nCir. 2003), aff’d, 546 U.S. 21 (2005) (holding that under the\\ncontinuous workday rule, “work time [is] continuous, not the\\nsum of discrete periods”).\\n\\n Of course, this rule raises inevitable questions: when\\ndoes the workday begin, and when does it end? The DOL\\ndefines the “workday” to generally mean “the period\\nbetween the commencement and completion on the same\\nworkday of an employee's principal activity or activities.”\\n29 C.F.R. § 790.6(b). The Supreme Court expanded upon\\nthis definition, interpreting “principal activity or activities”\\nto also include “all activities which are an integral and\\nindispensable part of the principal activities.” IBP, 546 U.S.\\nat 29–30 (internal quotation marks and citation omitted).\\nThus, any activity which is “integral and indispensable” to\\nprincipal activities, even if performed outside of a scheduled\\nshift, triggers the beginning of the “workday.” Id. at 31–37.\\n“Among the activities included as an integral part of a\\nprincipal activity are those closely related activities which\\nare indispensable to its performance,” 29 C.F.R. § 790.8(c),\\nsuch as knife-sharpening performed outside of a scheduled\\nshift by butchers at a meatpacking plant. Mitchell v. King\\nPacking Co., 350 U.S. 260, 261–63 (1956).\\n\\n C.\\n\\n With all of that in mind, we turn to how these two\\ndoctrines impact this case, and more specifically, whether\\nthe district court was correct in concluding that the\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 47\\n\\ncombination of Tyson and the continuous workday rule\\nenabled plaintiffs to show that they meet Rule 23(b)(3)’s\\npredominance requirement. Defendants contend that the\\ndistrict court erred in holding that plaintiffs had\\ndemonstrated predominance for two main reasons:\\n(1) because the Main Survey asked only about arrival and\\ndeparture times at the ballpark and not about what activities\\nthe players actually performed while at the ballpark,\\nplaintiffs cannot rely on the continuous workday theory\\nbecause there is no way to determine the beginning or end of\\nthe “workday,” and (2) the Main Survey revealed significant\\nvariations in players’ arrival and departure times, even\\namong players employed by the same MLB franchise.\\n\\n This task requires us to address the proposed Arizona\\nand Florida classes separately from the California class. As\\nan initial matter, however, we note that despite defendants’\\nrepeated suggestions to the contrary, the representative\\nevidence offered by plaintiffs was not limited to just the\\nMain Survey, nor are observational studies the only type of\\nevidence permitted to fill in evidentiary gaps under Tyson.\\nWe reject defendants’ erroneous view of the record and their\\ncramped reading of Tyson.\\n\\n 1.\\n\\n As to the Arizona and Florida classes, we easily affirm\\nthe district court’s determination. Recall that these two\\nclasses cover time spent participating in spring training,\\nextended spring training, and the instructional leagues—\\nperiods during which virtually all players are completely\\nunpaid for their participation. 19 Moreover, these classes do\\n\\n 19\\n Payroll data produced by defendants reveals that of the 21,211\\nplayers who participated in spring training between the 2009 and 2015\\n\\n48 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nnot bring overtime claims, but rather allege minimum wage\\nviolations. 20 Therefore—as the district court correctly\\nheld—liability can be established simply by showing that the\\nclass members performed any compensable work. 21 That is\\neasily resolved on a classwide basis by answering two\\nquestions: (1) are the players employees of defendants, and\\n(2) do the minor league team activities during these periods\\nconstitute compensable work under the laws of either\\nArizona or Florida? We hold that these two “common,\\naggregation-enabling issues in the case are more prevalent\\n[and] important than the non-common, aggregation-\\ndefeating, individual issues,” therefore making certification\\nappropriate. Tyson, 136 S. Ct. at 1045 (quoting 2 Newberg\\non Class Actions § 4:49 (5th ed. 2012)).\\n\\n Defendants do not seriously contest that their policy is to\\ndeny players compensation during spring training, extended\\nspring training, and the instructional leagues—nor could\\nthey credibly do so, given that the MLB’s own mandatory\\n\\nseasons, only 11 were paid a salary. Put differently, a mere .005% of\\nplayers received a salary during spring training, and those 11 players\\nmay be identified through payroll records and appropriately excluded\\nfrom the class. Likewise, a small number of MLB franchises pay players\\nduring extended spring training, but these players are identifiable\\nthrough payroll records and may either be excluded from the class or,\\npotentially, placed into a subclass.\\n 20\\n The Arizona and Florida classes also bring quantum meruit\\nclaims, and the Arizona class alleges recordkeeping violations, but the\\nparties do not dispute that these claims are irrelevant to this portion of\\nour predominance analysis.\\n\\n 21\\n We also note that the Arizona class’s claims are bolstered by the\\nfact that under Arizona law, failure to keep appropriate records of hours\\nworked “raise[s] a rebuttable presumption that the employer did not pay\\nthe required minimum wage rate.” Ariz. Rev. Stat. Ann. § 23-364.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 49\\n\\ncontract “obligates Player[s] to perform professional\\nservices on a calendar year basis, regardless of the fact that\\nsalary payments are to be made only during the actual\\nchampionship playing season.” And as we have long held,\\nsuch uniform corporate policies “carry great weight for\\ncertification purposes.” In re Wells Fargo Home Mortg.\\nOvertime Pay Litig., 571 F.3d at 958. This is not the “rare[]”\\ncase where predominance is defeated despite the existence\\nof an employer’s “common practice or policy.” 7 Newberg\\non Class Actions § 23:33 (5th ed. 2012).\\n\\n We also agree with the district court that as to these\\nclasses, many of defendants’ protests go to damages, not\\nliability. Damages may well vary, and may require\\nindividualized calculations. But “the rule is clear: the need\\nfor individual damages calculations does not, alone, defeat\\nclass certification.” Vaquero v. Ashley Furniture Indus.,\\nInc., 824 F.3d 1150, 1155 (9th Cir. 2016); see Tyson, 136\\nS. Ct. at 1045 (holding that where “one or more of the central\\nissues in the action are common to the class and can be said\\nto predominate,” certification may be appropriate “even\\nthough other important matters will have to be tried\\nseparately, such as damages.” (quoting 7AA C. Wright, A.\\nMiller, & M. Kane, Federal Practice and Procedure § 1778\\n(3d ed. 2005))).\\n\\n We do not, however, mean to minimize defendants’\\ncriticisms of the Main Survey. Indeed, we agree that there\\nare a number of legitimate questions about the\\npersuasiveness of the Main Survey, especially if it were the\\nonly representative evidence submitted in support of\\ncertification. But as we have mentioned, the Main Survey\\nwas but one piece of the plaintiffs’ representative\\nevidence—evidence that also included hundreds of internal\\n\\n50 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nteam schedules and public game schedules, payroll data, and\\nthe testimony of both players and league officials.\\n\\n At minimum, if the players are “employees” under either\\nArizona or Florida law and defendants are unable to prove\\nthat any affirmative defenses apply, the team schedules will\\nserve to conclusively demonstrate that the players spent time\\nworking for which they were uncompensated. See Ariz.\\nRev. Stat. Ann. § 23-362; Ariz. Admin. Code. § R20-5-\\n1202(19) (“‘Hours worked’ means all hours for which an\\nemployee covered under the Act is employed and required\\nto give to the employer, including all time during which an\\nemployee is on duty or at a prescribed work place and all\\ntime the employee is suffered or permitted to work.”);\\n29 C.F.R. § 778.223 (“As a general rule the term ‘hours\\nworked’ will include: (a) All time during which an employee\\nis required to be on duty or to be on the employer's premises\\nor at a prescribed workplace and (b) all time during which\\nan employee is suffered or permitted to work whether or not\\nhe is required to do so.”). 22 Moreover, if plaintiffs can\\npersuade a jury that their workday began at a particular\\ntime—either because they were required to report at that\\ntime, 23 or because they arrived of their own volition but\\n\\n 22\\n We rely on interpretations of the FLSA here because Florida’s\\nconstitution provides that “case law, administrative interpretations, and\\nother guiding standards developed under the federal FLSA shall guide\\nthe construction of [the constitutional amendment providing for a\\nminimum wage] and any implementing statutes or regulations.” Fla.\\nConst. art. X, § 24.\\n\\n 23\\n See 29 C.F.R. § 790.6 (“If an employee is required to report at the\\nactual place of performance of his principal activity at a certain specific\\ntime, his ‘workday’ commences at the time he reports there for work in\\naccordance with the employer's requirement.”).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 51\\n\\nengaged in work activities upon arriving (i.e., were\\n“permitted” to work)—the continuous workday doctrine\\neliminates the need for plaintiffs to prove which activities\\nthey engaged in throughout the day. 24 See IBP, 546 U.S. at\\n28, 32–37.\\n\\n Defendants should not “be heard to complain that the\\ndamages lack the exactness and precision of measurement\\nthat would be possible had [they] kept records in accordance\\nwith the [statutory] requirements,” even if their “lack of\\naccurate records grows out of a bona fide mistake as to\\nwhether certain activities or non-activities constitute work.”\\nMt. Clemens, 328 U.S. at 688. “Having received the benefits\\nof such work, [defendants] cannot object to the payment for\\nthe work on the most accurate basis possible under the\\ncircumstances.” Id.\\n\\n 2.\\n\\n We next address whether the district court was correct to\\nhold that predominance had been met for the California\\nclass. Given the differences in the types of claims brought\\nby the California class as compared to the Arizona and\\nFlorida classes, certification of the California class is more\\ncomplex and requires additional analysis. Unlike the\\nArizona and Florida classes, the California class brought\\nclaims relating to work performed during the championship\\nseason—a time when the players do get paid, albeit not\\nmuch. As a result, in order to prove liability on their\\novertime claims, the California class must show that its\\n 24\\n A jury may also decide that for baseball players, activities like\\nhitting practice with coaches and supervised weightlifting—much like\\nknife-sharpening by butchers at a meatpacking plant— are “integral and\\nindispensable” to the principal activity of playing baseball and therefore\\ntrigger the start of the “workday.” See Mitchell, 350 U.S. at 261–63.\\n\\n52 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nmembers worked more than 8 hours in a day, more than 40\\nhours in a week, and/or worked 7 days in a workweek. See\\nCal. Labor Code § 510; Mendoza v. Nordstrom, Inc., 393\\nP.3d 375, 381–82 (Cal. 2017). Likewise, to establish\\nliability on their minimum wage claims, the California class\\nmust demonstrate that they worked hours for which they\\nwere not paid at least minimum wage—but whereas the\\nArizona and Florida classes can demonstrate liability simply\\nby showing they worked any hours, the California class’s\\nburden is made more challenging by the fact that the players\\nreceive some pay. See Cal. Labor Code §§ 1182 et seq;\\nArmenta v. Osmose, Inc., 37 Cal. Rptr. 3d 460, 466–68 (Cal.\\nCt. App. 2005). Nonetheless, a number of considerations\\nlead us to affirm the district court’s determination.\\n\\n First, as with defendants’ uniform policy of not paying\\nplayers for participation outside of the championship season,\\ndefendants do not credibly dispute that their policy is to\\nnever pay overtime and to pay a fixed salary, regardless of\\nthe actual number of hours worked. We reiterate that\\ncommon corporate policies like this “carry great weight for\\ncertification purposes,” In re Wells Fargo Home Mortg.\\nOvertime Pay Litig., 571 F.3d at 958, and that predominance\\nis “rarely” defeated in cases where such uniform policies\\nexist. See 7 Newberg on Class Actions § 23:33 (5th ed.\\n2012).\\n\\n Second, the team schedules alone—independent of the\\nMain Survey or any other evidence—may suffice to show\\novertime liability. As the district court noted, plaintiffs’\\nexpert testified that approximately 65–85% of California\\nLeague players had at least one workweek with games on all\\nseven days, and that nearly half of all workweeks included\\ngames on all seven days. For those workweeks, the players\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 53\\n\\nwould be entitled to overtime pay for their work on the\\nseventh day of the workweek. See Cal. Labor Code § 510.\\n\\n Third, and most significantly, we are persuaded that\\nunder Tyson, the representative evidence plaintiffs offered\\nwas adequate to meet their burden at this stage. As we\\nobserved in the preceding section, defendants do identify\\nmultiple legitimate criticisms of the Main Survey, and it is\\ncertainly possible that a jury may not find the Main Survey—\\neven in combination with all of plaintiffs’ other evidence—\\nadequate proof of liability (or at least not to the extent\\nplaintiffs allege). In particular, a jury may be persuaded by\\ndefendants’ arguments that players did not begin\\ncompensable work upon arriving at the ballpark or that\\nplayers stopped engaging in compensable work long before\\nthey left the ballpark, such that the Main Survey’s estimated\\narrival and departure times are insufficient to clear the\\npreponderance hurdle. As we explain below, however,\\nTyson counsels that such criticisms do not doom certification\\nhere unless no reasonable jury could conclude that the\\ncombination of the Main Survey and plaintiffs’ other\\nrepresentative evidence was probative of the amount of time\\nplayers actually spent performing compensable work. See\\nTyson, 136 S. Ct. at 1046–49. And while defendants\\ncorrectly point out that the Main Survey revealed meaningful\\nvariations in players’ arrival and departure times, the same\\nwas true of the employees’ donning and doffing times in\\nTyson—yet such variation did not preclude certification\\nthere. See id. at 1043; id. at 1055 (Thomas, J., dissenting).\\n\\n Because defendants do not challenge the district court’s\\nruling on admissibility under Daubert, the defects they have\\nidentified with the Main Survey could only have defeated\\ncertification upon a conclusion that all of the representative\\nevidence offered—the Main Survey, schedules, testimony,\\n\\n54 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nand the like—could not have “sustained a reasonable jury\\nfinding as to hours worked in each employee’s individual\\naction.” See Tyson, 136 S. Ct. at 1046–47. As in Tyson, the\\ndistrict court “made no such finding,” id. at 1049, and indeed\\nfound the opposite:\\n\\n Plaintiffs will be able to use the survey data\\n in combination with other evidence that may\\n be sufficient to allow a jury to draw\\n conclusions based on reasonable inference as\\n to when players were required to be at the\\n ballpark and how long after games they were\\n required to remain at the ballpark. . . . Thus,\\n as in Tyson Foods, it appears that\\n representative evidence can be combined\\n with actual records of time spent engaged in\\n the various activities to derive a reasonable\\n estimate of the amount of time worked by\\n class members.\\n\\nWe are then left to ask whether “the record here provides [a]\\nbasis for [us] to second-guess that conclusion.” Id.\\n\\n After reviewing the record, we are satisfied that we\\nshould not disturb the district court’s determination, in part\\ndue to California’s expansive definition of “employ” and\\n“hours worked.” 25 Under California law, to “employ”\\n\\n 25\\n Unlike Arizona and Florida law—the former of which is silent on\\nthe incorporation of FLSA doctrines, and the latter of which expressly\\nincorporates them—we are not persuaded that the continuous workday\\nrule should apply to the California class. We view California’s definition\\nof “hours worked” as more expansive and more employee-friendly than\\nunder the FLSA, even with the incorporation of the continuous workday\\nrule. The California Supreme Court has “cautioned against confounding\\nfederal and state labor law,” and has consistently held that “absent\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 55\\n\\nmeans “to engage, suffer, or permit to work.” Cal. Code\\nRegs. tit. 8, §§ 11040(2)(E), 11100(2)(E) (emphasis\\nadded). 26 “Hours worked” means “the time during which an\\nemployee is subject to the control of an employer, and\\nincludes all the time the employee is suffered or permitted to\\nwork, whether or not required to do so.” Cal. Code Regs. tit.\\n8, §§ 11040(2)(K), 11100(2)(H). Inexplicably, however,\\ndefendants claim that under California law, “time spent\\nengaging in activities that are not required by, or under the\\ncontrol of, an employer is not compensable and does not\\nbegin or extend a workday.” This is a tortured and wholly\\nunsupported reading of the law, and is manifestly contrary\\nto one of the cases defendants themselves cite in support of\\ntheir argument. See Morillion v. Royal Packing Co., 995\\nP.2d 139, 143 (Cal. 2000) (“[T]he two phrases—‘time\\nduring which an employee is subject to the control of an\\nemployer’ and ‘time the employee is suffered or permitted\\nto work, whether or not required to do so’—can also be\\ninterpreted as independent factors, each of which defines\\nwhether certain time spent is compensable as ‘hours\\nworked.’”) (citation omitted).\\n\\n\\n\\n\\nconvincing evidence of the [California agency’s] intent to adopt the\\nfederal standard for determining whether time is compensable under\\nstate law, we decline to import any federal standard, which expressly\\neliminates substantial protections to employees, by implication.”\\nTroester v. Starbucks Corp., 421 P.3d 1114, 1119 (Cal. 2018)\\n(alterations, citations, and internal quotation marks omitted).\\n\\n 26\\n The California Industrial Welfare Commission’s wage orders\\n“have the force of law.” Alvarado v. Dart Container Corp. of Cal., 411\\nP.3d 528, 532 (Cal. 2018). We need not decide today which wage order\\napplies to minor league players, as all of the most relevant orders define\\n“employ” and “hours worked” the same way.\\n\\n56 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n Indeed, Morillion counsels that “hours worked” includes\\nall time the employer “permit[s]” an employee to work, even\\nif the work is not required and the employee is not under the\\nemployer’s control. See id. Thus, a player who arrives early\\nor stays late at the ballpark of their own volition and\\nperforms “work” activities during that time is still owed\\ncompensation because the player was “permitted” to work,\\ndespite the work not being required.\\n\\n Likewise, under Morillion, if players were expected to\\narrive or depart at a particular time—whether that\\nrequirement was de facto or official—it is immaterial what\\nactivities the players actually engaged in while at the\\nballpark. Even if the players spent their time at the ballpark\\ndoing things like eating or showering, they were still under\\ntheir employer’s control and unable “to use the time\\neffectively for their own purposes,” and thus were owed\\ncompensation. See id. at 146. Indeed, Morillion explicitly\\nrejected an analogous argument by the employer in that case:\\n\\n We reject Royal’s contention that plaintiffs\\n were not under its control during the required\\n bus ride because they could read on the bus,\\n or perform other personal activities.\\n Permitting plaintiffs to engage in limited\\n activities such as reading or sleeping on the\\n bus does not allow them to use the time\\n effectively for their own purposes . . .\\n Plaintiffs were foreclosed from numerous\\n activities in which they might otherwise\\n engage if they were permitted to travel to the\\n fields by their own transportation. Allowing\\n plaintiffs the circumscribed activities of\\n reading or sleeping does not affect, much less\\n eliminate, the control Royal exercises by\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 57\\n\\n requiring them to travel on its buses and by\\n prohibiting them from effectively using their\\n travel time for their own purposes. Similarly\\n . . . listening to music and drinking coffee\\n while working in an office setting can also be\\n characterized as personal activities, which\\n would not otherwise render the time working\\n noncompensable.\\n\\nId. (alterations, internal quotation marks, and citations\\nomitted). Thus, if plaintiffs use their representative\\nevidence—especially the Main Survey and the testimony of\\nplayers and league officials—to persuade a jury that they\\nwere required to be at the ballpark at particular times, they\\nneed not show how the players spent that time.\\n\\n The fourth and final consideration weighing in favor of\\naffirming the district court’s determination is our standard of\\nreview. Abuse of discretion is always a relatively deferential\\nstandard, but when we review a grant of class certification,\\n“we accord the district court noticeably more deference than\\nwhen we review a denial.” Abdullah, 731 F.3d at 956\\n(citation omitted). Were we to review de novo, this would\\nlikely be a closer call. But as they say, tie goes to the\\nrunner—and, under our deferential standard, to the district\\ncourt.\\n\\n D.\\n\\n Finally, defendants, citing to Wal-Mart Stores, Inc. v.\\nDukes, 564 U.S. 338, 352 (2011), contend that the district\\ncourt was required to “rigorously analyze” the Main Survey,\\nrather than evaluating its admissibility under Daubert and its\\nappropriateness for meeting class certification requirements\\nunder Tyson. Tyson requires that we reject this argument.\\nThere, the Court explicitly distinguished the use of\\n\\n58 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\nrepresentative evidence to establish hours worked in wage\\nand hour claims from the use of representative evidence in\\ncases like Wal-Mart. Tyson, 136 S. Ct. at 1048.\\nSpecifically—as we have explained—for wage and hour\\ncases where the employer has failed to keep proper records,\\nTyson holds that once a district court has found expert\\nevidence to be admissible, it may only deny its use to meet\\nthe requirements of Rule 23 certification if “no reasonable\\njuror” could find it probative of whether an element of\\nliability was met. Id. at 1049. Given the similarities between\\nthis case and Tyson, the rule set forward in Tyson controls,\\nand “[defendants’] reliance on Wal-Mart is misplaced.” 27\\nId. (citation omitted).\\n\\n IV.\\n\\n We next address whether the district court properly\\ncertified the FLSA collective action.\\n\\n FLSA permits employees to bring lawsuits on behalf of\\n“themselves and other employees similarly situated.” 29\\nU.S.C. § 216(b). We recently delineated the appropriate\\nstandard for FLSA collective certification in Campbell v.\\nCity of Los Angeles, 903 F.3d 1090 (9th Cir. 2018). As we\\nexplained in Campbell, “there is no established definition of\\nthe FLSA’s ‘similarly situated’ requirement, nor is there an\\nestablished test for enforcing it.” Id. at 1111 (citing Thiessen\\nv. Gen. Elec. Capital Corp., 267 F.3d 1095, 1102 (10th Cir.\\n2001)). In Campbell, we rejected both the minority\\napproach to FLSA collective certification—which treats a\\nFLSA collective as analogous to a Rule 23(b)(3) class—and\\n\\n 27\\n Tyson expressly cautioned that this rule should be read narrowly\\nand not assumed to apply outside of the wage and hour context. 136\\nS. Ct. at 1049.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 59\\n\\nthe majority “ad hoc” approach. Id. at 1111–1117. The\\nformer approach, we observed, is inconsistent with the\\nstatute itself, as well as the choice of Congress and the\\nAdvisory Committee on Rules to distinguish FLSA\\ncollectives from Rule 23 class actions. Id. at 1111–1113.\\nAnd while the latter approach—the so-called ad hoc\\napproach—is a “significant improvement” over the minority\\napproach, it has two major flaws that led us to decline to\\nadopt it. Id. at 1113–1116. First, this approach\\ninappropriately “focus[es] on differences rather than\\nsimilarities among the party plaintiffs,” leading district\\ncourts to “treat[] difference as disqualifying,” rather than\\n“treat[ing] the requisite kind of similarity as the basis for\\nallowing partially distinct cases to proceed together.” Id. at\\n1117. Second, because the ad hoc approach allows district\\ncourts to weigh “fairness and procedural considerations,” it\\n“invites courts to import, through a back door, requirements\\nwith no application to the FLSA,” such as Rule 23’s\\npredominance, adequacy, and superiority requirements. Id.\\nat 1115.\\n\\n Because of the flaws in the two predominant approaches\\nto FLSA collective certification, we instead developed our\\nown standard: “[p]arty plaintiffs are similarly situated, and\\nmay proceed in a collective, to the extent they share a similar\\nissue of law or fact material to the disposition of their FLSA\\nclaims.” Id. at 1117. Significantly, as long as the proposed\\ncollective’s “factual or legal similarities are material to the\\nresolution of their case, dissimilarities in other respects\\nshould not defeat collective treatment.” Id. at 1114\\n(emphasis omitted).\\n\\n The district court here did not have the benefit of our\\nopinion in Campbell, and instead followed the vast majority\\nof district courts in this circuit by applying the ad hoc\\n\\n60 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\napproach. See, e.g., Lewis v. Wells Fargo & Co., 669\\nF. Supp. 2d 1124, 1127 (N.D. Cal. 2009) (“Although various\\napproaches have been taken to determine whether plaintiffs\\nare ‘similarly situated,’ district courts in this circuit have\\nused the ad hoc, two-tiered approach.”). While legally\\nincorrect, we conclude that the district court’s erroneous use\\nof the ad hoc approach was harmless under the\\ncircumstances, 28 and we affirm the collective’s certification.\\n\\n The district court found that plaintiffs met their burden\\nof demonstrating they were “similarly situated,” reasoning:\\n\\n First, by eliminating the winter conditioning\\n claims and pursuing on a classwide basis only\\n claims that are based on the continuous\\n workday doctrine, Plaintiffs have\\n significantly reduced the need to engage in\\n individualized inquiries relating to the type of\\n work performed. Second, the Court is now\\n persuaded that the payroll records maintained\\n by Defendants will allow any variations in\\n compensation to be analyzed without\\n burdensome individualized inquiries. This is\\n especially true as to the spring training,\\n extended spring training and instructional\\n league claims because players generally were\\n not compensated for their participation in\\n these activities and the small fraction of\\n players who did receive compensation for\\n 28\\n As we explained in Campbell, the ad hoc approach imposes a\\nhigher bar for certification than the FLSA requires. See Campbell, 903\\nF.3d at 1114–1116. Thus, if the collective was appropriately certified\\nunder the more stringent ad hoc approach, a fortiori the collective would\\nbe appropriately certified under Campbell’s more lenient approach to\\n“similarly situated.” See id.\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 61\\n\\n these activities can be identified using payroll\\n records maintained by Defendants. Third, as\\n discussed above, the Court finds that the\\n defenses asserted by Defendants to the FLSA\\n present common questions that are not likely\\n to be overwhelmed by the need to conduct\\n individualized inquiries. Finally, the\\n possibility that the Court will be required to\\n apply the laws of numerous states (or at a\\n minimum, conduct numerous choice of law\\n inquiries) is not present as to the FLSA class,\\n which will require the Court to apply only\\n federal wage and hour law.\\n\\n Defendants’ arguments in support of reversal echo those\\nthey make in relation to the Rule 23(b)(3) classes, and we\\nreject them for largely the same reasons. Cf. Tyson, 136\\nS. Ct. at 1036 (“For purposes of this case . . . if certification\\nof respondents’ class action under [Rule 23] was proper,\\ncertification of the collective action was proper as well.”).\\nWe therefore expand on our earlier reasoning only briefly.\\n\\n Because the FLSA collective covers work performed\\nduring spring training, extended spring training, and the\\ninstructional leagues—that is, work for which the players\\nreceived no pay—we affirm the certification of the collective\\nfor that work. Specifically, for these time periods, two\\ncommon legal questions drive the litigation: are the players\\nemployees, and do the activities they perform during those\\ntimes constitute compensable work? As nearly all players\\nare unpaid during these time periods, if the answers to those\\ntwo questions are resolved in plaintiffs’ favor, liability may\\nbe established by showing that the players performed any\\nwork.\\n\\n62 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\n We also affirm the district court’s certification of the\\nFLSA collective as to plaintiffs’ overtime claims, although\\nthis holding requires additional explanation. Critical to our\\ndecision is that plaintiffs allege a single, FLSA-violating\\npolicy—the failure to pay overtime under any\\ncircumstances—and argue a common theory of defendants’\\nstatutory violations: that defendants “suffer or permit”\\nplaintiffs to perform compensable work before and after\\nscheduled practice and game times. These are “similar\\nissue[s] of law or fact material to the disposition of their\\nFLSA claims,” thus making plaintiffs “similarly situated.”\\nCampbell, 903 F.3d at 1117. And as previously discussed,\\nwe believe a reasonable jury could find that all of plaintiffs’\\nevidence—not just the Main Survey, but also the schedules,\\ntestimony, and payroll data—sustains a “just and reasonable\\ninference” as to the hours players actually worked. See\\nTyson, 136 S. Ct. at 1046–47.\\n\\n Specifically, there are several overlapping ways that\\nplaintiffs may be able to rely on their representative evidence\\nto persuade a jury that they have worked overtime hours for\\nwhich they were not compensated. Under any of these\\nscenarios, the continuous workday rule lends significant\\nassistance to plaintiffs by eliminating the need for plaintiffs\\nto prove exactly which activities they engaged in throughout\\nthe day. See IBP, 546 U.S. at 28, 32–37.\\n\\n First, plaintiffs could potentially use their evidence—\\nparticularly the Main Survey, but also the testimony of\\nplayers and league officials—to establish approximate times\\nthat they were required to arrive at and depart from the\\nballpark. This would obviate the need for plaintiffs to\\ndemonstrate which activities they engaged in upon arrival or\\nprior to departure. See 29 C.F.R. § 790.6 (“If an employee\\nis required to report at the actual place of performance of his\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 63\\n\\nprincipal activity at a certain specific time, his ‘workday’\\ncommences at the time he reports there for work in\\naccordance with the employer's requirement.”); 29 C.F.R.\\n§ 778.223 (“As a general rule the term ‘hours worked’ will\\ninclude . . . [a]ll time during which an employee is required\\nto be on duty or to be on the employer’s premises or at a\\nprescribed workplace.”) (emphasis added).\\n\\n Second, plaintiffs could rely on their representative\\nevidence to demonstrate that before and after the times they\\nwere required to be at the ballpark, they still performed\\nactivities at the ballpark that were “an integral and\\nindispensable part of [their] principal activities” and were\\ntherefore compensable. See IBP, 546 U.S. at 29–30. As\\nmentioned previously, a jury may well determine that\\nactivities like batting practice or supervised weightlifting are\\nto baseball players what knife-sharpening is to butchers at a\\nmeatpacking plant—that is, activities that are “integral and\\nindispensable” to the principal activity of playing baseball.\\nSee Mitchell, 350 U.S. at 261–63. If so, such activities\\nwould trigger the start of the “workday” within the meaning\\nof the FLSA. Plaintiffs may have somewhat of an uphill\\nbattle proceeding under this second theory on a collective-\\nwide basis, but we are certainly not prepared to say that no\\nreasonable jury could find defendants liable for overtime\\nviolations under this theory. See Tyson, 136 S. Ct. at 1048–\\n49; cf. Campbell, 903 F.3d at 1117–1119 (explaining that\\npost-discovery decertification motions should be evaluated\\nunder the summary judgment standard where “overlap exists\\nbetween the availability of the collective action mechanism\\nand the merits of the underlying claim”).\\n\\n Finally, if internal team schedules establish that\\nplaintiffs had required team-related activities for forty hours\\n\\n64 SENNE V. KANSAS CITY ROYALS BASEBALL\\n\\na week, 29 then plaintiffs can establish liability simply by\\nshowing that they performed any additional work beyond\\nthose officially-scheduled times. Cf. Tyson, 136 S. Ct. at\\n1036 (Thomas, J., dissenting) (explaining that in Mt.\\nClemens, the employer was “presumptively liable to all\\nemployees because they all claimed to work 40 hours per\\nweek. All additional uncompensated work was necessarily\\nunpaid overtime.”) (citation omitted).\\n\\n Under any of these theories, damages will inevitably be\\nindividualized, at least to some extent. But just as the need\\nfor individualized damage calculations is insufficient to\\ndefeat Rule 23 certification, “[i]ndividual damages amounts\\ncannot defeat collective treatment under the more forgiving\\nstandard” for FLSA collective certification. See Campbell,\\n903 F.3d at 1117 (citing Leyva v. Medline Indus. Inc., 716\\nF.3d 510, 514 (9th Cir. 2013)). District courts are well-\\nequipped to deal with issues of individualized calculations\\nin the wage-and-hour context, and may use “any of the\\npractices developed to deal with Rule 23 classes facing\\nsimilar issues.” Id. at 18 (citing Jimenez v. Allstate Ins. Co.,\\n765 F.3d 1161, 1167 (9th Cir. 2014)).\\n\\n\\n\\n 29\\n Given the internal team schedules in the record, this may be an\\neasy task, particularly for spring training and extended spring training.\\nFor example, a spring training schedule for one of the San Francisco\\nGiants’ affiliates involved a workday beginning at 6:30 AM on the day\\nof a 1:00 PM away game, with a 50 minute window provided for transit\\nbetween the training facility and the ballpark. Assuming for the sake of\\nargument that the 1:00 PM game lasted 2.5 hours and that the return trip\\nto the training facility took the same amount of time—50 minutes—as\\nthe outgoing trip, that day alone entailed approximately 10 hours of work\\nif the players left the training facility immediately upon their return (and\\nbased on the testimony in the record, that assumption seems\\nimplausible).\\n\\n SENNE V. KANSAS CITY ROYALS BASEBALL 65\\n\\n As is true in all FLSA cases, underlying our decision\\ntoday is the background principle that “because the FLSA is\\na remedial statute, it must be interpreted broadly.” Lambert\\nv. Ackerley, 180 F.3d 997, 1003 (9th Cir. 1999) (en banc)\\n(citing Tennessee Coal, Iron & R. Co. v. Muscoda Local No.\\n123, 321 U.S. 590, 597 (1944)). After all, the FLSA does\\nnot deal “with mere chattels or articles of trade but with the\\nrights of those who toil.” Tennessee Coal, 321 U.S. at 597.\\nWe are satisfied that certification of the collective is not only\\nappropriate under our interpretation of “similarly situated,”\\nbut also that it is consistent with “the great public policy”\\nembodied by the FLSA. Mt. Clemens, 328 U.S. at 687.\\n\\n V.\\n\\n For the reasons explained above, we AFFIRM in part,\\nREVERSE in part, and REMAND for further proceedings\\nconsistent with this opinion.\\n\\n Plaintiffs-Appellants/Cross-Appellees shall recover\\ntheir costs on appeal.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4078,"cells":{"id":{"kind":"number","value":4649426,"string":"4,649,426"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Vicente Cuevas-Lopez"},"decision_date":{"kind":"string","value":"2019-08-19"},"docket_number":{"kind":"string","value":"17-10438"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Richard R. Clifton, Sandra S. Ikuta, and Michelle T. Friedland, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"FRIEDLAND, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10438\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 4:17-cr-00306-\\n JGZ-DTF-1\\nVICENTE CUEVAS-LOPEZ, AKA\\nVicente Cuevas Lopez,\\n Defendant-Appellant. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Jennifer G. Zipps, District Judge, Presiding\\n\\n Argued and Submitted March 6, 2019\\n Phoenix, Arizona\\n\\n Filed August 19, 2019\\n\\n Before: Richard R. Clifton, Sandra S. Ikuta,\\n and Michelle T. Friedland, Circuit Judges.\\n\\n Opinion by Judge Friedland;\\n Dissent by Judge Ikuta\\n\\n2 UNITED STATES V. CUEVAS-LOPEZ\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n Affirming a sentence for attempted illegal reentry after\\ndeportation in violation of 8 U.S.C. § 1326, the panel held\\nthat the “single sentence rule” in U.S.S.G. § 4A1.2(a)(2)\\napplies to the enhancements in U.S.S.G. § 2L1.2(b)(2) and\\n(b)(3).\\n\\n The single sentence rule instructs that whether to treat\\nmultiple prior sentences as a single sentence depends on\\nwhether they were separated by an intervening arrest,\\ncharged in the same instrument, or imposed on the same day;\\nand provides that if prior sentences are treated as a single\\nsentence, a court should use the longest sentence of\\nimprisonment if concurrent sentences were imposed and use\\nthe aggregate sentence of imprisonment if consecutive\\nsentences were imposed. A state court had previously\\nsentenced the defendant to two consecutive 3.5-year terms\\nimposed on the same day for two second-degree burglary\\nconvictions.\\n\\n Because the single sentence rule applies to § 2L1.2, the\\npanel concluded that the district court properly relied on the\\nrule to aggregate the defendant’s two consecutive 3.5-year\\nsentences in applying a ten-level enhancement pursuant to\\nU.S.S.G. § 2L1.2(b)(3)(A), which applies to a defendant\\ncharged under § 1326 who was previously ordered deported\\nor removed and who subsequently committed a felony\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 3\\n\\noffense for which the sentence imposed was five years or\\nmore.\\n\\n Dissenting, Judge Ikuta wrote that under the plain\\nlanguage of the Sentencing Guidelines, the defendant does\\nnot have “a conviction for a felony offense . . . for which the\\nsentence imposed was five years or more,” U.S.S.G.\\n§ 2L1.2; and that the applicable Guidelines range should not\\nbe increased based solely on inferences regarding the\\nSentencing Commission’s unspoken intent.\\n\\n\\n COUNSEL\\n\\nJeffrey G. Buchella (argued), Tucson, Arizona, for\\nDefendant-Appellant.\\n\\nCorey J. Mantei (argued), Assistant United States Attorney;\\nElizabeth A. Strange, First Assistant United States Attorney;\\nRobert L. Miskell, Appellate Chief; United States Attorney’s\\nOffice, Tucson, Arizona; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nFRIEDLAND, Circuit Judge:\\n\\n Defendant-Appellant Vicente Cuevas-Lopez pleaded\\nguilty to attempted illegal reentry after deportation in\\nviolation of 8 U.S.C. § 1326. In determining Cuevas-\\nLopez’s sentence, the district court applied a ten-level\\nenhancement to his base offense level pursuant to United\\nStates Sentencing Guidelines Manual (“U.S.S.G.” or\\n“Guidelines Manual”) § 2L1.2(b)(3)(A) (“the\\nEnhancement”), as had been recommended by the Pre-\\n\\n4 UNITED STATES V. CUEVAS-LOPEZ\\n\\nSentence Report. 1 The Enhancement applies to a defendant\\ncharged under 8 U.S.C. § 1326 who was previously ordered\\ndeported or removed and who subsequently committed “a\\nfelony offense . . . for which the sentence imposed was five\\nyears or more.” U.S.S.G. § 2L1.2(b)(3)(A).\\n\\n A state court had previously sentenced Cuevas-Lopez to\\ntwo consecutive 3.5-year terms imposed on the same day for\\ntwo second-degree burglary convictions. The district court\\naggregated Cuevas-Lopez’s two 3.5-year sentences to\\nproduce a seven-year sentence for purposes of applying the\\nEnhancement, relying on § 4A1.2(a)(2) of the Guidelines\\nManual, which is known as the “single sentence rule.”\\nCuevas-Lopez, who did not object at sentencing, now argues\\nthat the district court erred in adhering to the single sentence\\nrule and thus in aggregating the two sentences when\\nconsidering what level of enhancement to apply. We affirm,\\njoining the Fifth Circuit in holding that the single sentence\\nrule in § 4A1.2(a)(2) governs the determination whether an\\nenhancement applies under § 2L1.2(b).\\n\\n I.\\n\\n A.\\n\\n The Sentencing Reform Act of 1984 established “factors\\nto guide [federal] district courts in exercising their traditional\\n\\n 1\\n We review Cuevas-Lopez’s sentence based on the 2016 Guidelines\\nManual, which was in effect at the time of Cuevas-Lopez’s sentencing.\\nU.S.S.G. § 1B1.11 (2018) (“The court shall use the Guidelines Manual\\nin effect on the date that the defendant is sentenced,” unless doing so\\n“would violate the ex post facto clause of the United States\\nConstitution.”); see also United States v. Thomsen, 830 F.3d 1049, 1071\\n(9th Cir. 2016). All section references, and all citations to the Guidelines\\nManual, are to the 2016 version unless otherwise specified.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 5\\n\\nsentencing discretion.” Beckles v. United States, 137 S. Ct.\\n886, 893 (2017). Congress simultaneously “created the\\nUnited States Sentencing Commission and charged it with\\nestablishing guidelines to be used for sentencing.” Id.\\nAlthough “[t]he Guidelines were initially binding on district\\ncourts,” the Supreme Court in United States v. Booker,\\n543 U.S. 220 (2005), “rendered them ‘effectively\\nadvisory.’” Beckles, 137 S. Ct. at 894 (quoting Booker,\\n543 U.S. at 245). The Guidelines Manual is nonetheless\\n‘“the starting point and the initial benchmark’ for\\nsentencing.” Id. (quoting Gall v. United States, 552 U.S. 38,\\n49 (2007)).\\n\\n The Guidelines Manual provides sentencing ranges\\ndetermined by a combination of “the seriousness of a\\ndefendant’s offense . . . and his [or her] criminal history.”\\nMolina-Martinez v. United States, 136 S. Ct. 1338, 1342\\n(2016). The offense seriousness is reflected in an “offense\\nlevel” comprised of a base offense level, which is assigned\\nby the Guidelines Manual to each type of conviction;\\nspecific offense characteristics, which can increase or\\ndecrease the offense level for each offense; and upward and\\ndownward adjustments, which can be applied to any offense.\\nSee U.S.S.G. § 1B1.1(a)(1)–(5). A defendant is assigned\\ncriminal history points based on his or her past criminal\\nconduct, which then places the defendant in a criminal\\nhistory category between I and VI. See U.S.S.G. § 4A1.1;\\nU.S.S.G. ch. 5, pt. A, Sentencing Table. The Guidelines\\nManual combines this criminal history category with a\\ndefendant’s calculated offense level to produce a\\nrecommended sentencing range. U.S.S.G. ch. 5, pt. A,\\nSentencing Table.\\n\\n6 UNITED STATES V. CUEVAS-LOPEZ\\n\\n B.\\n\\n Cuevas-Lopez, who is a citizen of Mexico, was ordered\\ndeported from the United States in 2004, and was deported\\nseveral times between 2004 and 2015. He unsuccessfully\\nattempted to reenter the country in 2017. He was then\\ncharged with attempted illegal reentry after deportation, in\\nviolation of 8 U.S.C. § 1326(a), as enhanced by 8 U.S.C.\\n§ 1326(b)(1). Cuevas-Lopez pled guilty to the charge, and\\nthe district court held a sentencing hearing in October 2017.\\n\\n In advance of the hearing, the U.S. Probation Office\\nprepared a Pre-Sentence Report (“PSR”). Applying\\n§ 2L1.2(a), the relevant offense guideline for illegal reentry\\noffenses, the PSR recommended a base offense level of eight\\nfor a violation of 8 U.S.C. § 1326. With regard to specific\\noffense characteristics, the PSR added a ten-level\\nenhancement pursuant to § 2L1.2(b)(3)(A) based on the fact\\nthat “[a]fter [Cuevas-Lopez] was first ordered deported or\\nremoved from the United States, he sustained a felony\\nconviction for which the sentences imposed in a two-count\\nindictment were ordered to run consecutive,” and which\\n“resulted in a cumulative sentence that was five years or\\nmore.” 2\\n\\n The ten-level enhancement arose out of Cuevas-Lopez’s\\nNovember 3, 2007 arrest for two felony burglaries. The first\\nburglary was reported on November 1, 2007, and the second\\nwas committed on November 3, 2007. Cuevas-Lopez was\\n\\n 2\\n Although the PSR referred to Cuevas-Lopez as having “sustained\\na felony conviction” after first being deported, the relevant question\\nunder § 2L1.2(b)(3) is whether a defendant “engaged in criminal\\nconduct” after deportation that later resulted in a felony conviction. This\\ndifference is immaterial here because Cuevas-Lopez’s relevant conduct\\nand resulting convictions all occurred after he was first ordered deported.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 7\\n\\nconvicted of both burglaries, and on March 10, 2008 an\\nArizona state court sentenced him to 3.5 years in prison for\\neach offense, ordered to run consecutively.\\n\\n The PSR applied the single sentence rule to treat Cuevas-\\nLopez’s two consecutive 3.5-year sentences as a single\\nseven-year sentence. This aggregated seven-year sentence\\ntriggered the ten-level enhancement under § 2L1.2(b)(3)(A)\\n(again, the “Enhancement”), so the PSR assigned Cuevas-\\nLopez an adjusted offense level of 18 for the instant illegal\\nreentry offense. It then applied a three-level downward\\nadjustment for acceptance of responsibility, resulting in an\\noffense level of 15. The PSR placed Cuevas-Lopez in\\ncriminal history category V based on the two burglaries\\ndiscussed above and other convictions between 2006 and\\n2008. Cuevas-Lopez’s criminal history category and offense\\nlevel resulted in a Guidelines range of 37 to 46 months in\\nprison. Without aggregating Cuevas-Lopez’s two burglary\\nsentences, § 2L1.2(b)(3)(B) would have called for an eight-\\nlevel enhancement based on a 3.5-year sentence, which\\nwould have resulted in an offense level of 13 after the\\ndownward adjustment and a Guidelines range of 30 to\\n37 months. See U.S.S.G. ch. 5, pt. A, Sentencing Table.\\n\\n At the sentencing hearing, the Government agreed with\\nthe PSR’s recommendation and argued, based on the\\n§ 2L1.2(b)(3)(A) Enhancement, that Cuevas-Lopez’s base\\noffense level of eight was “properly enhanced by ten levels\\nbecause the defendant received a seven-year consecutive\\nstate sentence . . . for two second-degree burglary counts\\ncharged in a single indictment.” The district court adopted\\nthe Guidelines calculation in the PSR, to which Cuevas-\\nLopez did not object. The court sentenced Cuevas-Lopez to\\n37 months in prison and three years of supervised release.\\n\\n8 UNITED STATES V. CUEVAS-LOPEZ\\n\\n Cuevas-Lopez timely appealed and now argues that the\\nsingle sentence rule does not apply to § 2L1.2(b)(3)\\nenhancements, and that the district court therefore should not\\nhave aggregated his two prior 3.5-year sentences for\\npurposes of determining his Guidelines range.\\n\\n II.\\n\\n A.\\n\\n Where, as here, a defendant makes an argument on\\nappeal that was not the basis for an objection in the district\\ncourt, we generally review for plain error. See United States\\nv. Gomez, 725 F.3d 1121, 1125 (9th Cir. 2013) (“If the\\ndefendant fails to object, we review for plain error.”); see\\nalso Fed. R. Crim P. 52(b) (“A plain error that affects\\nsubstantial rights may be considered even though it was not\\nbrought to the court’s attention.”). We have held, however,\\nthat “we are not limited to [plain error] review when we are\\npresented with a question that ‘is purely one of law’ and\\nwhere ‘the opposing party will suffer no prejudice as a result\\nof the failure to raise the issue in the trial court.’” United\\nStates v. Saavedra-Velazquez, 578 F.3d 1103, 1106 (9th Cir.\\n2009) (quoting United States v. Echavarria-Escobar,\\n270 F.3d 1265, 1267–68 (9th Cir. 2001)).\\n\\n Cuevas-Lopez urges us to apply de novo review, despite\\nhis failure to object in the district court, because his\\nargument about when the single sentence rule applies\\npresents a pure question of law. The Government responds\\nthat our court’s “pure question of law” exception to plain\\nerror review is inconsistent with the Supreme Court’s\\ninterpretation of Federal Rule of Criminal Procedure 52(b).\\nIn support, the Government cites Puckett v. United States,\\n556 U.S. 129 (2009), in which the Supreme Court held that\\n“[f]ailure to abide by [the] contemporaneous-objection rule\\n\\n UNITED STATES V. CUEVAS-LOPEZ 9\\n\\nordinarily precludes the raising on appeal of [an]\\nunpreserved claim of trial error,” with a “limited exception,”\\nid. at 135, for a “plain error that affects substantial rights,”\\nid. (quoting Fed. R. Crim. P. 52(b)). The Government also\\nrelies on Judge Graber’s concurrence in United States v.\\nZhou, 838 F.3d 1007 (9th Cir. 2016), which opined that\\n“[o]ur ‘pure question of law’ exception contradicts Rule\\n52(b) and the Supreme Court’s case law.” Id. at 1016\\n(Graber, J., concurring). We need not resolve this dispute or\\notherwise decide which standard of review applies here,\\nbecause we would affirm under either de novo or plain error\\nreview. For the reasons explained below, we conclude that\\nthe district court correctly interpreted the Guidelines Manual\\nin determining that the single sentence rule applies to\\n§ 2L1.2(b) enhancements.\\n\\n B.\\n\\n 1.\\n\\n Although the Guidelines are advisory only, a “district\\ncourt must correctly calculate the recommended Guidelines\\nsentence and use that recommendation as the ‘starting point\\nand the initial benchmark.’” United States v. Munoz-\\nCamarena, 631 F.3d 1028, 1030 (9th Cir. 2011) (quoting\\nKimbrough v. United States, 552 U.S. 85, 108 (2007)). The\\ncourt must keep the Guidelines range “in mind throughout\\nthe process,” id. (quoting United States v. Carty, 520 F.3d\\n984, 991 (9th Cir. 2008) (en banc)), and “justify the extent\\nof [any] departure from the Guidelines,” id. “Failure to\\ncalculate the correct Guidelines range constitutes procedural\\nerror.” Peugh v. United States, 569 U.S. 530, 537 (2013).\\n\\n As the Supreme Court has explained, the Guidelines\\nManual “contains text of three varieties,” each of which is\\nwritten by the Sentencing Commission. Stinson v. United\\n\\n10 UNITED STATES V. CUEVAS-LOPEZ\\n\\nStates, 508 U.S. 36, 41 (1993). The “[f]irst is a guideline\\nprovision itself,” which “provide[s] direction as to the\\nappropriate type [and extent] of punishment.” Id. A “second\\nvariety of text in the [Guidelines] Manual is a policy\\nstatement . . . regarding application of the guidelines or other\\naspects of sentencing that would further the purposes of the\\n[Sentencing Reform] Act.” Id. (quotation marks omitted).\\nThe third type of text in the Guidelines Manual is\\ncommentary, which accompanies both guidelines and policy\\nstatements. Id. Commentary may serve three functions: to\\n“interpret a guideline or explain how it is to be applied”; to\\n“suggest circumstances which may warrant departure from\\nthe guidelines”; and to “provide background information,\\nincluding factors considered in promulgating the guideline\\nor reasons underlying promulgation of the guideline.” Id.\\n(alterations and citation omitted). Any modifications or\\namendments to the guidelines provisions themselves (the\\nfirst category of text) must be accompanied by a “statement\\nof the reasons therefor” authored by the Sentencing\\nCommission, and take effect on a date set by the\\nCommission, within certain statutory parameters and subject\\nto any changes made by Congress. See 28 U.S.C. § 994(p).\\n\\n “We interpret the Sentencing Guidelines using the\\nordinary tools of statutory interpretation.” United States v.\\nMartinez, 870 F.3d 1163, 1166 (9th Cir. 2017). Our\\ninterpretation “will most often begin and end with the text\\nand structure of the [g]uidelines” provisions themselves. Id.\\n(quoting United States v. Joey, 845 F.3d 1291, 1297 n.8 (9th\\nCir. 2017)). We also consider “the Commission’s\\ncommentary interpreting or explaining the text” of those\\nguidelines provisions. Id. The commentary “is authoritative\\nunless it violates the Constitution or a federal statute, or is\\ninconsistent with, or a plainly erroneous reading of, that\\nguideline.” Stinson, 508 U.S. at 38; see also United States\\n\\n UNITED STATES V. CUEVAS-LOPEZ 11\\n\\nv. Prien-Pinto, 917 F.3d 1155, 1157–58 (9th Cir. 2019)\\n(following Stinson). Policy statements, likewise, are binding\\n“[t]o the extent that they interpret substantive guidelines and\\ndo not conflict with them or any statutory directives.”\\nUnited States v. Chea, 231 F.3d 531, 536 n.1 (9th Cir. 2000)\\n(citation omitted); see also Stinson, 508 U.S. at 42 (“The\\nprinciple that the Guidelines Manual is binding on federal\\ncourts applies as well to policy statements.”). “We may also\\nlook to [a] provision’s history and purpose, such as by\\nconsulting the Commission’s statements of reason for a\\nparticular amendment.” Martinez, 870 F.3d at 1166 (citation\\nand quotation marks omitted); see also United States v.\\nOrnelas, 825 F.3d 548, 554 (9th Cir. 2016) (noting that we\\n“look to the Commission’s statements of reason for\\nguidance” in interpreting individual guidelines).\\n\\n 2.\\n\\n Chapter Two of the Guidelines Manual governs offense\\nconduct. The offense guideline within Chapter Two\\napplicable to a conviction for attempted illegal reentry is\\n§ 2L1.2. 3 Section 2L1.2(a) provides a base offense level of\\neight. Sections 2L1.2(b)(1), (b)(2), and (b)(3), which\\ncontain the specific offense characteristics, apply\\nenhancements of between two and ten levels based on a\\ndefendant’s prior convictions. Some of these enhancements\\ndepend on the types of convictions sustained and others\\n\\n 3\\n The guideline applicable to a substantive offense generally also\\napplies to inchoate offenses such as attempt. See U.S.S.G. § 2X1.1\\n(2018). The parties do not dispute that § 2L1.2, which is titled\\n“Unlawfully Entering or Remaining in the United States,” applies to\\nCuevas-Lopez’s attempted illegal reentry. See United States v. Rosales-\\nAguilar, 818 F.3d 965, 972 (9th Cir. 2016) (applying § 2L1.2 to an\\nattempted illegal reentry case).\\n\\n12 UNITED STATES V. CUEVAS-LOPEZ\\n\\ndepend on the length of sentences received. In\\n§ 2L1.2(b)(1), § 2L1.2(b)(2)(E), and § 2L1.2(b)(3)(E), the\\nsize of the enhancement turns on the type of offense. 4 The\\nother subsections in § 2L1.2(b)(2) and § 2L1.2(b)(3),\\nincluding the Enhancement, add varying levels of\\nenhancement based on the length of a defendant’s sentence\\nimposed for prior convictions.\\n\\n Subsection (b)(3), which the district court applied to\\nenhance Cuevas-Lopez’s base offense level in this case,\\nprovides in full:\\n\\n (3) (Apply the Greatest) If, at any time after\\n the defendant was ordered deported or\\n ordered removed from the United States for\\n the first time, the defendant engaged in\\n criminal conduct resulting in—\\n\\n (A) a conviction for a felony offense (other\\n than an illegal reentry offense) for which\\n the sentence imposed was five years or\\n more, increase by 10 levels;\\n\\n (B) a conviction for a felony offense (other\\n than an illegal reentry offense) for which\\n\\n\\n 4\\n In § 2L1.2(b)(1), the level of enhancement depends on whether the\\ndefendant, before committing the offense for which he is being\\nsentenced, sustained “a conviction for a felony that is an illegal reentry\\noffense” or sustained “two or more convictions for [improper entry]\\nmisdemeanors under 8 U.S.C. § 1325(a).” U.S.S.G. § 2L1.2(b)(1). Both\\n(b)(2)(E) and (b)(3)(E) apply a two-level enhancement where a\\ndefendant has “three or more convictions for misdemeanors that are\\ncrimes of violence or drug trafficking offenses.”\\nU.S.S.G. § 2L1.2(b)(2)(E), (b)(3)(E).\\n\\n UNITED STATES V. CUEVAS-LOPEZ 13\\n\\n the sentence imposed was two years or\\n more, increase by 8 levels;\\n\\n (C) a conviction for a felony offense (other\\n than an illegal reentry offense) for which\\n the sentence imposed exceeded one year\\n and one month, increase by 6 levels;\\n\\n (D) a conviction for any other felony offense\\n (other than an illegal reentry offense),\\n increase by 4 levels; or\\n\\n (E) three or more convictions for\\n misdemeanors that are crimes of violence\\n or drug trafficking offenses, increase by\\n 2 levels.\\n\\nU.S.S.G. § 2L1.2(b)(3) (underlining added). Subsection\\n2L1.2(b)(2) mirrors subsection (b)(3), but relates to a\\ndefendant’s convictions, if any, sustained before being\\nordered deported or removed. See § 2L1.2(b)(2). 5\\n\\n\\n 5\\n Although we apply the 2016 Guidelines Manual here, we note that\\n§ 2L1.2 was amended in 2018 to “establish[] that the application of the\\n§ 2L1.2(b)(2) enhancement depends on the timing of the underlying\\n‘criminal conduct,’ and not on the timing of the resulting conviction.”\\nU.S.S.G. Supp. to app. C, amend. 809 at 188 (Nov. 1, 2018). The\\namended § 2L1.2(b)(2) applies when “before the defendant was ordered\\ndeported or ordered removed . . . for the first time, the defendant engaged\\nin criminal conduct that, at any time, resulted in” the various convictions\\nand sentences enumerated in that section. U.S.S.G. § 2L1.2(b)(2)\\n(2018). The Sentencing Commission simultaneously updated subsection\\n(b)(3) to mirror the language in subsection (b)(2), see § 2L1.2(b)(3)\\n(2018), though subsection (b)(3), unlike (b)(2), had already (before the\\n2018 amendment) depended upon the timing of the underlying conduct,\\nrather than the underlying conviction, see § 2L1.2(b)(3) (2016).\\n\\n14 UNITED STATES V. CUEVAS-LOPEZ\\n\\n Cuevas-Lopez contends that the words “a conviction” in\\nthe Enhancement (underlined above) preclude a district\\ncourt from aggregating consecutive sentences for purposes\\nof applying the Enhancement. Although this contention\\nfinds some support in the language of § 2L1.2(b)(3), we are\\npersuaded in light of the application notes in the commentary\\nto § 2L1.2, as well as the Sentencing Commission’s\\nstatement of reasons for the 2016 amendment to § 2L1.2,\\nthat the district court properly aggregated Cuevas-Lopez’s\\ntwo 3.5-year sentences here.\\n\\n In aggregating Cuevas-Lopez’s two sentences, the\\ndistrict court relied on the single sentence rule. The single\\nsentence rule falls within Chapter Four of the Guidelines\\nManual, which contains guidelines for categorizing a\\ndefendant’s criminal history. Section 4A1.2(a) of Chapter\\nFour defines “prior sentence” as “any sentence previously\\nimposed upon adjudication of guilt, whether by guilty plea,\\ntrial, or plea of nolo contendere, for conduct not part of the\\ninstant offense.” U.S.S.G. § 4A1.2(a)(1). The guideline\\nthen instructs that whether to treat multiple prior sentences\\nas a single sentence depends on whether they were separated\\nby an intervening arrest, charged in the same instrument, or\\nimposed on the same day. Specifically, § 4A1.2(a)(2) states:\\n\\n If the defendant has multiple prior sentences,\\n determine whether those sentences are\\n counted separately or treated as a single\\n sentence. Prior sentences always are counted\\n separately if the sentences were imposed for\\n offenses that were separated by an\\n intervening arrest (i.e., the defendant is\\n arrested for the first offense prior to\\n committing the second offense). If there is\\n no intervening arrest, prior sentences are\\n\\n UNITED STATES V. CUEVAS-LOPEZ 15\\n\\n counted separately unless (A) the sentences\\n resulted from offenses contained in the same\\n charging instrument; or (B) the sentences\\n were imposed on the same day. Treat any\\n prior sentence covered by (A) or (B) as a\\n single sentence. See also § 4A1.1(e).\\n\\nU.S.S.G. § 4A1.2(a)(2).\\n\\n Section 4A1.2(a)(2) goes on to provide that “if prior\\nsentences are treated as a single sentence,” a court should\\n“use the longest sentence of imprisonment if concurrent\\nsentences were imposed” and “use the aggregate sentence of\\nimprisonment” where “consecutive sentences were\\nimposed.” U.S.S.G. § 4A1.2(a)(2). Here, the district court\\napplied a ten-level enhancement based on Cuevas-Lopez’s\\nconsecutive 3.5-year sentences, rather than an eight-level\\nenhancement, which would have applied had Cuevas-\\nLopez’s two 3.5-year sentences been ordered to run\\nconcurrently (or if they had been ordered to run\\nconsecutively but the single sentence rule’s aggregation\\nprovisions did not apply). Our task, therefore, is to\\ndetermine whether Chapter Four’s single sentence rule\\nproperly applies to § 2L1.2.\\n\\n In support of his argument that the single sentence rule’s\\naggregation provisions should not apply here, Cuevas-Lopez\\nrelies on § 2L1.2’s commentary—specifically Application\\nNote 2. Application Note 2 to § 2L1.2 gives “sentence\\nimposed” “the meaning given the term ‘sentence of\\nimprisonment’ in Application Note 2 and subsection (b) of\\n§ 4A1.2.” U.S.S.G. § 2L1.2 cmt. n.2. As Cuevas-Lopez\\npoints out, this application note to § 2L1.2 does not reference\\nsubsection (a) of § 4A1.2, which defines “prior sentence”\\nand contains the single sentence rule. Cuevas-Lopez argues\\n\\n16 UNITED STATES V. CUEVAS-LOPEZ\\n\\nthat because Application Note 2 specifically cross-\\nreferences a part of § 4A1.2 and yet does not reference the\\npart containing the single sentence rule, the Commission\\nmeant to exclude the single sentence rule from applying to\\nthe § 2L1.2(b) enhancements.\\n\\n In our view, the absence of a mention of subsection (a)\\nof § 4A1.2 in Application Note 2 to § 2L1.2 does not hold\\nthe significance Cuevas-Lopez wishes. 6 Rather, we read\\n§ 4A1.2(b) and Application Note 2 to that section—the two\\nGuidelines Manual components specifically referenced in\\n§ 2L1.2’s commentary for purposes of defining “sentence\\nimposed”—as simply being silent on the question whether\\nseparate sentences that are imposed on the same day and\\nordered to run consecutively should be aggregated for\\npurposes of applying the Enhancement. Other application\\nnotes to the offense guideline in which the Enhancement\\nappears, however, speak more to the issue and evince the\\nSentencing Commission’s intent that the single sentence rule\\napply to § 2L1.2(b).\\n\\n One example is Application Note 3 to the § 2L1.2\\noffense guideline. Note 3 instructs a court that when\\n“applying subsections (b)(1), (b)(2), and (b)(3)” of\\n§ 2L1.2—i.e., the § 2L1.2(b) enhancements—a court should\\n\\n 6\\n The Guidelines Manual does state, as Cuevas-Lopez argues, that\\n“[a]n instruction to use a particular subsection or table from another\\noffense guideline refers only to the particular subsection or table\\nreferenced, and not to the entire offense guideline.” U.S.S.G.\\n§ 1B1.5(b)(2). Section 4A1.2 is not, however, an offense guideline—\\nrather, it provides “Definitions and Instructions for Computing Criminal\\nHistory.” And, in any event, even if the reference to subsection (b) does\\nnot itself amount to an instruction to apply subsection (a), it still leaves\\nus with silence on whether or not to apply subsection (a)’s single\\nsentence rule.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 17\\n\\n“use only those convictions that receive criminal history\\npoints under § 4A1.1(a), (b), or (c).” U.S.S.G. § 2L1.2 cmt.\\nn.3. Criminal history points are added under § 4A1.1(a) and\\n§ 4A1.1(b) based on “prior sentence[s] of imprisonment” of\\nover 13 months (for which three points are added) and\\nbetween 60 days and 13 months (for which two points are\\nadded), respectively. Under § 4A1.1(c), one additional point\\nis added for “each prior sentence” not counted in either (a)\\nor (b). Subsection (e) to § 4A1.1—which is not referenced\\nin § 2L1.2’s Application Note 3—instructs courts to “[a]dd\\n1 point for each prior sentence resulting from a conviction of\\na crime of violence that did not receive any points under (a),\\n(b), or (c) . . . because such sentence was treated as a single\\nsentence.” U.S.S.G. § 4A1.1(e) (emphasis added). The\\nupshot is that certain “prior sentence[s] of imprisonment” in\\n§ 4A1.1(a)–(c) will actually sometimes consist of\\naggregated sentences from multiple separate convictions\\npursuant to the single sentence rule. Because Application\\nNote 3 to § 2L1.2 instructs courts, for purposes of applying\\nthe § 2L1.2(b) enhancements, to use only convictions that\\nreceive criminal history points under subsections (a)–(c) of\\n§ 4A1.1, it suggests that whether and how a conviction\\nshould be counted for purposes of a § 2L1.2(b) enhancement\\ndepends on the operation of the single sentence rule.\\n\\n This understanding is reinforced by a second directive in\\nApplication Note 3 to § 2L1.2—that “for purposes of\\nsubsections (b)(1)(B), (b)(2)(E), and (b)(3)(E)” 7 of that\\nsection, courts “use only those convictions that are counted\\nseparately under [the single sentence rule in] § 4A1.2(a)(2).”\\nU.S.S.G. § 2L1.2 cmt. n.3. This directive further suggests\\n\\n 7\\n These subsections retained a categorical approach after the 2016\\namendment to § 2L1.2, discussed below, and deal with prior\\nmisdemeanor convictions. See infra n.9.\\n\\n18 UNITED STATES V. CUEVAS-LOPEZ\\n\\nthat the Sentencing Commission anticipated that in some\\ninstances two or more convictions could be counted—and\\nreferred to—as one under the single sentence rule, and that a\\ncourt should do so for purposes of other subsections of\\n§ 2L1.2. Were this not the baseline assumption, there would\\nbe no need to explicitly make the single sentence rule\\ninapplicable to these subsections. Application Note 3\\ntherefore weighs in favor of applying the single sentence rule\\nto determine which § 2L1.2(b)(3) enhancement applies.\\n\\n Application Note 4 to § 2L1.2, which deals with cases in\\nwhich a sentence for an illegal reentry offense was imposed\\nat the same time as one for another felony offense, lends\\nsome additional support for applying the single sentence\\nrule. It provides:\\n\\n There may be cases in which the sentences\\n for an illegal reentry offense and another\\n felony offense were imposed at the same time\\n and treated as a single sentence for purposes\\n of calculating the criminal history score\\n under § 4A1.1(a), (b), and (c). In such a case,\\n use the illegal reentry offense in determining\\n the appropriate enhancement under\\n subsection (b)(1), if it independently would\\n have received criminal history points. In\\n addition, use the prior sentence for the other\\n felony offense in determining the appropriate\\n enhancement under subsection (b)(3), if it\\n independently would have received criminal\\n history points.\\n\\nU.S.S.G. § 2L1.2 cmt. n.4 (emphasis added). Like Note 3,\\nthis application note appears to assume that § 4A1.2(a)’s\\n\\n UNITED STATES V. CUEVAS-LOPEZ 19\\n\\nsingle sentence rule would normally apply to § 2L1.2(b)\\nenhancements.\\n\\n 3.\\n\\n The Sentencing Commission’s statement of reasons for\\nthe amendment to § 2L1.2 that created the version applicable\\nhere also supports the district court’s application of the\\nsingle sentence rule. See Ornelas, 825 F.3d at 554 (looking\\n“to the Commission’s statements of reason for guidance” in\\ninterpreting a guideline). Before that 2016 amendment,\\n§ 2L1.2(b) enhancements were based on “the nature of a\\ndefendant’s most serious conviction,” as determined by the\\n“categorical approach to the penal statute underlying the\\nprior conviction.” U.S.S.G. Supp. to app. C, amend. 802\\nat155 (Nov. 1, 2016) (“Amendment 802”). Under the\\ncategorical approach, courts “compare the elements of the\\nstatute forming the basis of the defendant’s conviction with\\nthe elements of the ‘generic’ crime—i.e., the offense as\\ncommonly understood.” Descamps v. United States,\\n570 U.S. 254, 257 (2013). 8 Due to a concern that this\\nmethod of determining a level of enhancement was “overly\\ncomplex,” the Sentencing Commission adopted a simpler,\\nsentence-based model whereby, with a few exceptions, 9\\n\\n 8\\n We note, however, that even before 2016, the § 2L1.2(b)(1)(A) and\\n(B) enhancements required a court to use both a categorical approach\\nand a sentence-imposed approach, as subsection (A) applied where a\\ndefendant sustained “a conviction for a felony offense that [was] . . . a\\ndrug trafficking offense for which the sentence imposed exceeded 13\\nmonths,” and subsection (B) applied where a defendant sustained a\\nfelony drug trafficking conviction “for which the sentence imposed was\\n13 months or less.” U.S.S.G. § 2L1.2(b)(1) (2015).\\n 9\\n The guideline retained the categorical approach for\\n§ 2L1.2(b)(1)(B), (b)(2)(E), and (b)(3)(E), relating to predicate\\n\\n20 UNITED STATES V. CUEVAS-LOPEZ\\n\\n“[t]he level of the sentencing enhancement for a prior\\nconviction generally [would] be determined by the length of\\nthe sentence imposed for the prior offense.” 10 Amendment\\n802 at 155.\\n\\n In a section of the statement of reasons titled\\n“Accounting for Other Prior Convictions,” the Sentencing\\nCommission analogized the sentence-imposed approach in\\nsubsections (b)(2) and (b)(3) “to how Chapter Four of the\\nGuidelines Manual determines a defendant’s criminal\\nhistory score based on his or her prior convictions”—notably\\nreferencing Chapter Four (in which the single sentence rule\\nappears) in its entirety. Amendment 802 at 156. “The\\n[Sentencing] Commission concluded that the length of\\nsentence imposed by a sentencing court is a strong indicator\\nof the court’s assessment of the seriousness of the predicate\\noffense at the time, . . . consistent with how criminal history\\nis generally scored in . . . Chapter Four of the Guidelines\\nManual.” Amendment 802 at 157. The Commission also\\nwrote, in a section titled “Illegal Reentry”: “The definition\\nof ‘sentence imposed’ [in the amended § 2L1.2] is the same\\ndefinition that appears in Chapter Four of the Guidelines\\nManual.” 11 Amendment 802 at 155. The Sentencing\\n\\n\\nmisdemeanor convictions, as well as § 2L1.2(b)(1)(A), relating to\\npredicate illegal reentry felonies.\\n 10\\n Additionally, before 2016, § 2L1.2(b) included only an\\nenhancement for convictions that occurred before deportation. See\\nU.S.S.G. § 2L1.2 (2015). Amendment 802 sought to address a concern\\nthat, because of this, § 2L1.2 did not sufficiently account for all types of\\ncriminal conduct committed by persons charged with illegal reentry.\\nAmendment 802 at 155–56.\\n 11\\n We note that although Chapter Four uses the term “sentence\\nimposed,” it does not specifically define it.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 21\\n\\nCommission did not limit this explanation to any particular\\nprovision of Chapter Four, thereby seeming to incorporate\\nChapter Four’s single sentence rule.\\n\\n Similarly, the Commission wrote, in a section of the\\nstatement of reasons titled “Excluding Stale Convictions,”\\nthat in the context of § 2L1.2’s specific offense\\ncharacteristics, “it is . . . appropriate to employ the criminal\\nhistory rules.” Amendment 802 at 159. Because the single\\nsentence rule is contained in the part of Chapter Four that\\nprescribes the criminal history rules, this statement also\\nsuggests that the Commission envisioned that the rule would\\napply to the § 2L1.2(b) enhancements.\\n\\n Perhaps the strongest evidence of the Commission’s\\nintent appears in another passage of the statement of reasons\\nunder the heading “Application of the ‘Single Sentence\\nRule,’” which discusses the 2016 addition of what became\\nApplication Note 4 to § 2L1.2. See Amendment 802 at 159.\\nAs we have mentioned, Application Note 4 deals with cases\\nin which a defendant is sentenced for an illegal reentry\\noffense at the same time as another federal felony offense.\\nIn such cases, “the illegal reentry offense counts towards\\nsubsection (b)(1), while the other felony offense counts\\ntowards subsection (b)(3).” Amendment 802 at 159. The\\nCommission explained in the statement of reasons that it\\nintended “to make a distinction between illegal reentry\\noffenses and other types of offenses,” and “concluded that it\\nwas appropriate to ensure that such convictions are\\nseparately accounted for under the applicable specific\\noffense characteristics, even if they might otherwise\\nconstitute a ‘single sentence’ under § 4A1.2(a)(2).”\\nAmendment 802 at 159. Thus, both the text of Application\\nNote 4 and the Commission’s stated reason for adding it are\\n\\n22 UNITED STATES V. CUEVAS-LOPEZ\\n\\nbased on an assumption that the single sentence rule would\\napply to § 2L1.2.\\n\\n Applying the single sentence rule to § 2L1.2 also makes\\nsense in light of the Sentencing Commission’s purpose in\\ndrafting Amendment 802. The amended enhancements’\\n“sentence imposed” approach replaced the earlier\\ncategorical method and was intended to capture—in a\\nsimpler way—the seriousness of a defendant’s prior offense.\\nThe single sentence rule requires aggregation only when\\nsentences are ordered to run consecutively. The imposition\\nof consecutive, rather than concurrent, sentences generally\\nreflects a decision by either a sentencing judge or, in some\\ncases when sentences are statutorily required to run\\nconsecutively, by a legislature, that a consecutive term of\\nimprisonment would better reflect the seriousness of a\\ndefendant’s conduct as well as the need for deterrence,\\neducation or treatment of the defendant, and protection of\\nthe public. See 18 U.S.C. § 3584 (stating that “[m]ultiple\\nterms of imprisonment imposed at the same time run\\nconcurrently unless the court orders or the statute mandates\\nthat the terms are to run consecutively,” and directing judges\\nto consider the sentencing factors set forth in 18 U.S.C.\\n§ 3553 “in determining whether the terms imposed are to be\\nordered to run concurrently or consecutively”); 18 U.S.C.\\n§ 3553 (sentencing factors); Setser v. United States, 566 U.S.\\n231, 236 (2012) (“Judges have long been understood to have\\ndiscretion to select whether the sentences they impose will\\nrun concurrently or consecutively with respect to other\\nsentences that they impose.”). Applying the single sentence\\nrule to the § 2L1.2(b)(2) and (b)(3) enhancements therefore\\ncaptures the seriousness of a defendant’s prior convictions\\nand achieves the same goal that the previous categorical\\napproach sought to achieve.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 23\\n\\n 4.\\n\\n Even if there were some ambiguity in how broadly the\\nSentencing Commission intended the single sentence rule to\\napply, as the dissent’s arguments for a contrary\\ninterpretation suggest there may be, the goal of avoiding a\\ncircuit split would lead us to hold that the single sentence\\nrule applies here. “[A]bsent a strong reason to do so, we will\\nnot create a direct conflict with other circuits.” United States\\nv. Chavez-Vernaza, 844 F.2d 1368, 1374 (9th Cir. 1987); see\\nalso Global Linguist Sols., LLC v. Abdelmeged, 913 F.3d\\n921, 923 (9th Cir. 2019) (“[W]e so hold to avoid an\\nunnecessary circuit split.”). The Fifth Circuit recently\\nbecame the first circuit court to answer whether, under the\\n2016 Guidelines Manual, the single sentence rule applies to\\n§ 2L1.2(b). In United States v. Garcia-Sanchez, that court\\nheld that it does. 916 F.3d 522, 526 (5th Cir. 2019). 12 The\\nGarcia-Sanchez decision therefore weighs heavily in favor\\nof affirming the district court’s interpretation here.\\n\\n The Fifth Circuit found support for its conclusion in two\\ncomponents of Amendment 802 discussed above. It first\\nlooked to the section on “Accounting for Other Prior\\nConvictions” in the Sentencing Commission’s statement of\\nreasons for Amendment 802, which described the length of\\na sentence imposed as “a strong indicator of the court’s\\nassessment of the seriousness of the predicate offense,” and\\nas “consistent with the Chapter Four criminal history rules.”\\nGarcia-Sanchez, 916 F.3d at 527 (quoting Amendment 802\\nat 157–58). The Fifth Circuit further relied on the section\\n\\n 12\\n Prior to Garcia-Sanchez, the Fifth Circuit had held in United\\nStates v. Ponce-Flores, 900 F.3d 215 (5th Cir. 2018), that because any\\nerror could not have been obvious, a district court did not plainly err in\\napplying the single sentence rule to a § 2L1.2 enhancement. Id. at 219.\\n\\n24 UNITED STATES V. CUEVAS-LOPEZ\\n\\ntitled “Excluding Stale Convictions,” which explains that, in\\nthe context of § 2L1.2(b)’s specific offense characteristics,\\n“it is . . . appropriate to employ the criminal history rules,”\\nwhich contain the single sentence rule. Id. (quoting\\nAmendment 802 at 159).\\n\\n The court in Garcia-Sanchez also found persuasive the\\nFourth Circuit’s reasoning in United States v. Martinez-\\nVarela, 531 F.3d 298 (4th Cir. 2008). In that case, the\\ndistrict court aggregated sentences for purposes of applying\\na pre-2016 version of § 2L1.2(b)(1), which called for an\\nenhancement if a defendant was “previously convicted of a\\ndrug trafficking offense for which the ‘sentence imposed’\\nwas greater than thirteen months.” Martinez-Varela,\\n531 F.3d at 299. That version of § 2L1.2, like the 2016\\nversion, contained an application note that referenced\\nApplication Note 2 and subsection (b) of § 4A1.2 for\\npurposes of defining “sentence imposed.” Id. at 300. The\\nFourth Circuit affirmed, reasoning that although the\\nprovisions cross-referenced in application notes to § 2L1.2\\ndid not specifically address the issue of aggregating\\nsentences, commentary to § 4A1.1 instructed that “§§ 4A1.1\\nand 4A1.2 must be read together,” providing “strong\\nevidence that these two provisions should be read together\\nin determining [the defendant’s] criminal history points,”\\nand thus that the single sentence rule should apply to the\\n§ 2L1.2(b)(1) enhancement at issue. Id. at 301–02.\\n\\n Although our court had no previous occasion to reach the\\nsentencing question presented here, the holding in Garcia-\\nSanchez and the reasoning in Martinez-Varela are consistent\\nwith our case law. We previously recognized that\\n“[s]ections 2L1.2(b) and 4A1.1-2 serve the same underlying\\nfunction” of “determin[ing] the extent to which prior\\nconvictions affect a defendant’s sentence for the current\\n\\n UNITED STATES V. CUEVAS-LOPEZ 25\\n\\noffense,” and that it therefore “makes sense to treat prior\\nsentences in the same manner when they are used to\\ndetermine the offense level under section 2L1.2 as when\\nthey are used to determine the criminal history category.”\\nUnited States v. Ortiz-Gutierrez, 36 F.3d 80, 82 (9th Cir.\\n1994); 13 see also United States v. Moreno-Cisneros,\\n319 F.3d 456, 458–59 (9th Cir. 2003) (noting that\\n“Guideline § 4A1.2 is analogous to § 2L1.2(b)(1),” and that\\n§ 4A1.2 is “broadly applicable”); United States v. Frias,\\n338 F.3d 206, 210 (3d Cir. 2003) (describing Chapter 4 and\\nan earlier version of § 2L1.2 as “clearly aimed at the same\\nthing, which is varying the punishment based on the criminal\\nrecord,” and concluding that it therefore made sense to refer\\nto Chapter Four’s definitions in applying § 2L1.2 (quotation\\nmarks omitted)); United States v. Galicia-Delgado, 130 F.3d\\n518, 521 (2d Cir. 1997) (“[T]he definitions found in § 4A1.2\\nhave often been borrowed to interpret terms in § 2L1.2.”).\\n\\n We thus now join the Fifth Circuit in holding that the\\nsingle sentence rule applies to the enhancements in\\n§ 2L1.2(b)(2) and (b)(3). 14\\n\\n\\n 13\\n In Ortiz-Gutierrez, we applied the single sentence rule to another\\npre-2016 enhancement under § 2L1.2 that turned on whether a\\ndefendant’s predicate offense was an aggravated felony, which at the\\ntime “include[d] a crime of violence for which the defendant was\\nsentenced to at least five years imprisonment.” 36 F.3d at 82. Section\\n2L1.2 did not cross-reference either § 4A1.1 or § 4A1.2. Id.\\n 14\\n We reject Cuevas-Lopez’s argument that we should apply the rule\\nof lenity to hold in his favor. “[T]he rule of lenity applies to the\\nSentencing Guidelines,” but “only . . . where there is grievous ambiguity\\nor uncertainty in the guidelines.” United States v. D.M., 869 F.3d 1133,\\n1144 (9th Cir. 2017). We find no such “grievous ambiguity or\\nuncertainty” in the question of Guidelines Manual interpretation\\npresented here.\\n\\n26 UNITED STATES V. CUEVAS-LOPEZ\\n\\n III.\\n\\n For the foregoing reasons, we reject Cuevas-Lopez’s\\nargument that the district court was wrong to apply the single\\nsentence rule in calculating his sentence. We therefore\\nAFFIRM.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426679/\", \"author_raw\": \"FRIEDLAND, Circuit Judge:\"}, {\"author\": \"IKUTA, Circuit Judge, dissenting\", \"type\": \"dissent\", \"text\": \"IKUTA, Circuit Judge, dissenting:\\n\\n Under the plain language of the Guidelines, the district\\ncourt should have calculated a Guidelines range of 30 to 37\\nmonths for Cuevas-Lopez. Instead, the district court applied\\nan unrelated section of the Guidelines to miscalculate a\\nGuidelines range of 37 to 46 months. Because the district\\ncourt sentenced Cuevas-Lopez to the bottom of the\\nGuidelines range, Cuevas-Lopez was effectively deprived of\\nthe chance to get a 30-month sentence, instead of a 37-month\\nsentence. On de novo review, I would apply the Guidelines\\nas written, and therefore I dissent.\\n\\n I\\n\\n The factual background is simple. Vicente Cuevas-\\nLopez, a citizen of Mexico, was deported from the United\\nStates in 2004. In 2007, while in the United States illegally,\\nCuevas-Lopez was convicted in Arizona state court of two\\ncounts of second-degree burglary, one for burglarizing a\\nresidence and the second for taking power tools from a\\nvictim’s back yard. On March 10, 2008, he was sentenced\\nto three and a half years on each count.\\n\\n In 2017, Cuevas-Lopez attempted to reenter the United\\nStates illegally in Nogales, Arizona, but was stopped by\\n\\n UNITED STATES V. CUEVAS-LOPEZ 27\\n\\nBorder Patrol agents. He later pleaded guilty to attempted\\nillegal reentry after deportation, 8 U.S.C. § 1326.\\n\\n In calculating the Guidelines range, the district court\\napplied a ten-level sentencing enhancement. Such an\\nenhancement applies only to a defendant who engaged in\\nconduct that resulted in a conviction for a felony offense for\\nwhich the sentence imposed was five years of imprisonment\\nor more. U.S.S.G. § 2L1.2(b)(3)(A). 1 Although Cuevas-\\nLopez had not been convicted of such an offense—the\\nlongest sentence imposed on him was three and a half years\\nof imprisonment—the district court added his two sentences\\nfor second-degree burglary together, and concluded the ten-\\nlevel enhancement was applicable. In light of this\\nenhancement, the court calculated a Guidelines range of 37\\nto 46 months in prison. Under the correct calculation, the\\nGuidelines range would have been 30 to 37 months in prison.\\nSee id. § 2L1.2(b)(3)(B). Because the court imposed a\\nsentence at the bottom of the range, there is a chance that\\nCuevas-Lopez would have been sentenced differently had\\nthe district court made the correct calculation. See United\\nStates v. Munoz-Camarena, 631 F.3d 1028, 1031 (9th Cir.\\n2011) (holding that an error in calculating the Guidelines\\nrange was not harmless because “had the district court\\nstarted with the correct Guidelines range of 24 to 30 months,\\nrather than 33 to 41 months, it may have arrived at a different\\nsentence”).\\n\\n Today, the majority affirms the district court’s\\ncalculations under de novo review. But a straightforward\\n\\n\\n 1\\n Because Cuevas-Lopez was sentenced in 2016, all references to\\nthe Guidelines refer to the 2016 version of the Guidelines, unless\\notherwise stated. See U.S.S.G. § 1B1.11.\\n\\n28 UNITED STATES V. CUEVAS-LOPEZ\\n\\nreading of the text of the Guidelines shows that the majority\\nis wrong.\\n\\n II\\n\\n A district court must “begin all sentencing proceedings\\nby correctly calculating the applicable Guidelines range.”\\nGall v. United States, 552 U.S. 38, 49 (2007). Failure to\\ncorrectly calculate “the recommended Guidelines sentencing\\nrange is a significant procedural error that requires us to\\nremand for resentencing.” Munoz-Camarena, 631 F.3d at\\n1030.\\n\\n “We interpret the Sentencing Guidelines using the\\nordinary tools of statutory interpretation.” United States v.\\nMartinez, 870 F.3d 1163, 1166 (9th Cir. 2017).\\n“[C]ommentary in the Guidelines Manual that interprets or\\nexplains a guideline is authoritative unless it violates the\\nConstitution or a federal statute, or is inconsistent with, or a\\nplainly erroneous reading of, that guideline.” Stinson v.\\nUnited States, 508 U.S. 36, 38 (1993). “As with the\\ninterpretation of legal texts generally, our search for the\\nSentencing Commission’s intent will most often begin and\\nend with the text and structure of the Guidelines.” United\\nStates v. Joey, 845 F.3d 1291, 1297 n.8 (9th Cir. 2017)\\n(cleaned up).\\n\\n The offense guideline section applicable to Cuevas-\\nLopez’s offense is § 2L1.2 (the “Unlawful Reentry Offense”\\nguideline). To calculate Cuevas-Lopez’s offense level, the\\ncourt must first identify the “Base Offense Level” under the\\nUnlawful Reentry Offense guideline. See U.S.S.G.\\n§ 1B1.1(a)(1). In this case, the Unlawful Reentry Offense\\nguideline provides a single Base Offense Level of eight\\npoints for violations of 18 U.S.C. § 1326. Id. § 2L1.2(a).\\n\\n UNITED STATES V. CUEVAS-LOPEZ 29\\n\\n The court should then turn to the “Specific Offense\\nCharacteristics” for the Unlawful Reentry Offense,\\n§ 2L1.2(b), which sets out the enhancements to the Base\\nOffense Level, id. § 1B1.1(a)(2). Under subsection (3)(A)\\nof the Special Offense Characteristics, the district court must\\nimpose a ten-level enhancement if the defendant engaged in\\ncriminal conduct that resulted in “a conviction for a felony\\noffense (other than an illegal reentry offense) for which the\\nsentence imposed was five years or more.” Id.\\n 2\\n§ 2L1.2(b)(3)(A). Subsection (3)(B) of the Special Offense\\nCharacteristics mirrors subsection (3)(A) and requires the\\ndistrict court to impose an eight-level enhancement if the\\ndefendant’s criminal conduct resulted in “a conviction for a\\nfelony offense (other than an illegal reentry offense) for\\nwhich the sentence imposed was two years or more.” Id.\\n§ 2L1.2(b)(3)(B). 3\\n\\n\\n\\n 2\\n Section 2L1.2(b)(3)(A) provides:\\n\\n If, at any time after the defendant was ordered\\n deported or ordered removed from the United States\\n for the first time, the defendant engaged in criminal\\n conduct resulting in . . . a conviction for a felony\\n offense (other than an illegal reentry offense) for\\n which the sentence imposed was five years or more,\\n increase by 10 levels.\\n 3\\n Section 2L1.2(b)(3)(B) provides:\\n\\n If, at any time after the defendant was ordered\\n deported or ordered removed from the United States\\n for the first time, the defendant engaged in criminal\\n conduct resulting in . . . a conviction for a felony\\n offense (other than an illegal reentry offense) for\\n which the sentence imposed was two years or more,\\n increase by 8 levels.\\n\\n30 UNITED STATES V. CUEVAS-LOPEZ\\n\\n In this case, Cuevas-Lopez’s criminal conduct resulted\\nin “a conviction” for second degree burglary for which the\\nsentence imposed was three and a half years. Although\\nCuevas-Lopez’s conduct also resulted in a second conviction\\nfor a second degree burglary, the plain text of subsection\\n(3)(A) to the Special Offense Characteristics does not permit\\na court to aggregate the sentences of multiple convictions.\\nSee id. § 2L1.2(b)(3)(A). Rather, subsection (3)(A)\\ndistinctly refers in the singular to “a conviction” and “the\\nsentence.” Id. Therefore, subsection (3)(B) of the Special\\nOffense Characteristics, which applies to “a conviction for a\\nfelony offense . . . for which the sentence imposed was two\\nyears or more,” is the applicable enhancement, and the court\\nshould have added only eight levels to Cuevas-Lopez’s\\noffense level. Id. § 2L1.2(b)(3)(B).\\n\\n The commentary to the Unlawful Reentry Offense\\nguideline, § 2L1.2, is consistent with the conclusion that the\\n“sentenced imposed” in this case was a three and a half year\\nsentence, and that the court could not aggregate sentences\\nfrom multiple convictions. The term “sentence imposed” is\\ndefined in Comment 2 to the Unlawful Reentry Offense\\nguideline 4 as having “the same meaning given the term\\n‘sentence of imprisonment’” in “Definition and Instructions\\n\\n\\n 4\\n Application Note 2 to § 2L1.2 provides:\\n\\n “Sentence imposed” has the meaning given the term\\n “sentence of imprisonment” in Application Note 2 and\\n subsection (b) of §4A1.2 (Definitions and Instructions\\n for Computing Criminal History). The length of the\\n sentence imposed includes any term of imprisonment\\n given upon revocation of probation, parole, or\\n supervised release.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 31\\n\\nfor Computing Criminal History,” § 4A1.2(b) 5, and in\\nApplication Note 2 6 to that section, see id. § 2L1.2 cmt. 2.\\nBecause the cross-referenced provisions are in Chapter 4,\\ntheir purpose is to instruct the court on how to calculate the\\n\\n 5\\n Section 4A1.2(b) defines “sentence of imprisonment.” It provides:\\n\\n (1) The term “sentence of imprisonment” means a\\n sentence of incarceration and refers to the maximum\\n sentence imposed.\\n\\n (2) If part of a sentence of imprisonment was\\n suspended, “sentence of imprisonment” refers only to\\n the portion that was not suspended.\\n 6\\n Application Note 2 to § 4A1.2 provides:\\n\\n Sentence of Imprisonment.—To qualify as a sentence\\n of imprisonment, the defendant must have actually\\n served a period of imprisonment on such sentence (or,\\n if the defendant escaped, would have served time). See\\n §4A1.2(a)(3) and (b)(2). For the purposes of applying\\n §4A1.1(a), (b), or (c), the length of a sentence of\\n imprisonment is the stated maximum (e.g., in the case\\n of a determinate sentence of five years, the stated\\n maximum is five years; in the case of an indeterminate\\n sentence of one to five years, the stated maximum is\\n five years; in the case of an indeterminate sentence for\\n a term not to exceed five years, the stated maximum is\\n five years; in the case of an indeterminate sentence for\\n a term not to exceed the defendant’s twenty-first\\n birthday, the stated maximum is the amount of time in\\n pre-trial detention plus the amount of time between the\\n date of sentence and the defendant’s twenty-first\\n birthday). That is, criminal history points are based on\\n the sentence pronounced, not the length of time\\n actually served. See § 4A1.2(b)(1) and (2). A sentence\\n of probation is to be treated as a sentence under\\n §4A1.1(c) unless a condition of probation requiring\\n imprisonment of at least sixty days was imposed.\\n\\n32 UNITED STATES V. CUEVAS-LOPEZ\\n\\ncriminal history points associated with each sentence. Id.\\n§ 1B1.1(a)(6).\\n\\n The cross-referenced section and Application Note make\\nclear that the term “sentence of imprisonment” refers to the\\nsentence pronounced by the court on a single conviction.\\nThe section in “Definition and Instructions for Computing\\nCriminal History,” § 4A1.2(b), states that a sentence of\\nimprisonment “refers to the maximum sentence imposed”\\nand does not include any portion of the sentence that was\\nsuspended. Application Note 2 to that section explains that\\n“to qualify as a sentence of imprisonment, the defendant\\nmust have actually served a period of imprisonment on such\\nsentence.” Id. § 4A1.2 cmt. 2. Moreover, in calculating the\\ncriminal history points associated with a “sentence of\\nimprisonment,” the commentary makes clear that “criminal\\nhistory points are based on the sentence pronounced, not the\\nlength of time actually served.” Id. In short, each reference\\nto “sentence of imprisonment” refers to a single sentence.\\nThe term “sentence imposed” has “the same meaning given\\nthe term ‘sentence of imprisonment,’” and it therefore also\\nrefers to the sentence pronounced on a single conviction.\\nSee id. § 2L1.2 cmt. 2.\\n\\n Because the language of the Guidelines “is plain and\\nadmits of no more than one meaning” our “sole function” is\\nto enforce the terms of the Guidelines pursuant to their plain\\nmeaning. Carson Harbor Vill., Ltd. v. Unocal Corp., 270\\nF.3d 863, 878 (9th Cir. 2001) (en banc) (quoting Caminetti\\nv. United States, 242 U.S. 470, 485 (1917)). Here,\\nsubsection (3)(A)’s ten-level enhancement does not apply to\\nCuevas-Lopez because he does not have “a conviction for a\\nfelony offense . . . for which the sentence imposed was five\\nyears or more.” U.S.S.G. § 2L1.2(b)(3)(A). Although\\nCuevas-Lopez’s conduct resulted in two sentences for three\\n\\n UNITED STATES V. CUEVAS-LOPEZ 33\\n\\nand a half years each, neither sentence is a sentence of “five\\nyears or more.” Id. Therefore, the enhancement is\\ninapplicable.\\n\\n III\\n\\n Instead of applying the language of the Guidelines, the\\nmajority relies on complex and roundabout extrapolations to\\nassert that an unrelated section in the Criminal History\\nchapter (Chapter 4) applies to the Unlawful Reentry Offense\\nguideline, § 2L1.2, and requires the district court to add\\ntogether Cuevas-Lopez’s two separate second-degree\\nburglary offenses.\\n\\n A\\n\\n The unrelated section on which the majority puts so\\nmuch weight is referred to as the “Single Sentence Rule.”\\nId. § 4A1.2(a)(2). 7 In calculating a defendant’s criminal\\n\\n\\n 7\\n Section 4A1.2(a)(2) provides:\\n\\n If the defendant has multiple prior sentences,\\n determine whether those sentences are counted\\n separately or treated as a single sentence. Prior\\n sentences always are counted separately if the\\n sentences were imposed for offenses that were\\n separated by an intervening arrest (i.e., the defendant\\n is arrested for the first offense prior to committing the\\n second offense). If there is no intervening arrest, prior\\n sentences are counted separately unless (A) the\\n sentences resulted from offenses contained in the same\\n charging instrument; or (B) the sentences were\\n imposed on the same day. Treat any prior sentence\\n covered by (A) or (B) as a single sentence. See also\\n § 4A1.1(e).\\n\\n34 UNITED STATES V. CUEVAS-LOPEZ\\n\\nhistory, the court must add a specified number of points for\\neach of the defendant’s prior sentences. Id. § 1B1.1(a)(2).\\nChapter 4 defines the term “prior sentence” to mean “any\\nsentence previously imposed upon adjudication of guilt.” Id.\\n§ 4A1.2(a)(1). The Single Sentence Rule explains that if the\\ndefendant “has multiple prior sentences,” the court counts\\nthem separately if the sentences are separated by an\\nintervening arrest, as when the defendant was arrested for\\nthe first offense before committing the second offense. Id.\\n§ 4A1.2(a)(2). Prior sentences are counted as a single\\nsentence if there was no intervening arrest and they “resulted\\nfrom offenses contained in the same charging instrument” or\\n“the sentences were imposed on the same day.” Id.\\n\\n The majority concludes that a court must read the term\\n“sentenced imposed” in subsection (3)(A) of the Unlawful\\nReentry Offense guidelines to mean the same as the term\\n“prior sentence” in the Single Sentence Rule. Maj. Op. 16.\\nTherefore, according to the majority, under\\nsubsection (3)(A) of Unlawful Reentry Offense Guidelines,\\nCuevas-Lopez’s two convictions for second-degree burglary\\nshould be counted as a single conviction, and the two\\nseparate three and a half year sentences imposed for each\\nconvicted should be counted as a single sentence for “five\\nyears or more.” Maj. Op. 16.\\n\\n Notably lacking from the majority’s analysis is a single\\nword, phrase, or cross-reference in the Guidelines that would\\nauthorize a court to apply the Single Sentence Rule in the\\n\\n For purposes of applying §4A1.1(a), (b), and (c), if\\n prior sentences are treated as a single sentence, use the\\n longest sentence of imprisonment if concurrent\\n sentences were imposed. If consecutive sentences\\n were imposed, use the aggregate sentence of\\n imprisonment.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 35\\n\\ncontext of the Unlawful Reentry Offense guidelines. Nor is\\nthere any provision in the Guidelines which suggests the\\nterm “sentence imposed” has the same meaning as “prior\\nsentence.” See Burgess v. United States, 553 U.S. 124, 130\\n(2008) (“As a rule, [a] definition which declares what a term\\n‘means’ . . . excludes any meaning that is not stated.”\\n(quoting Colautti v. Franklin, 439 U.S. 379, 392–393 n.10\\n(1979)). Therefore, to reach this conclusion, the majority\\nlets loose an entire quiver of strained arguments, none of\\nwhich hits the mark.\\n\\n First, the majority brushes aside Cuevas-Lopez’s textual\\nargument (i.e., that nothing in subsection (3)(A) of the\\nUnlawful Reentry Offense guidelines incorporates the\\nSingle Sentence Rule’s definition of “prior sentence”) on the\\nspecious ground that subsection (3)(A) does not expressly\\nsay that the Single Sentence Rule is not applicable. See Maj.\\nOp. 16 (noting that subsection (3)(A) is “simply being silent”\\non that question). This is meritless. If a Guidelines section\\nuses a defined term (here, “sentenced imposed”) that does\\nnot cross-reference a different defined term in a different\\nsection (here, “prior sentence”), the natural inference is that\\nthe definition of the unrelated term is not applicable. See\\nBurgess, 553 U.S. at 130; Lamie v. U.S. Trustee, 540 U.S.\\n526, 539 (2004) (holding that courts should not add an\\n“absent word” to a statute).\\n\\n But the majority has more arguments up its sleeve.\\nAccording to the majority, Application Note 3 to the\\nUnlawful Reentry Offense guidelines 8, “evince[s] the\\n\\n 8\\n Application Note 3 to § 2L1.2 provides:\\n\\n Criminal History Points.—For purposes of applying\\n subsections (b)(1), (b)(2), and (b)(3), use only those\\n\\n36 UNITED STATES V. CUEVAS-LOPEZ\\n\\nSentencing Commission’s intent that the single sentence rule\\napply” to subsection (3)(A). Maj. Op. 16. The majority’s\\nreasoning is difficult to follow. The majority notes that\\nApplication Note 3 to the Unlawful Reentry Offense\\nguideline explains which convictions qualify as “a\\nconviction for a felony offense . . . for which the sentence\\nimposed” was a specified length of time. U.S.S.G.\\n§ 2L1.2(b)(3)(A); Maj. Op. 16–17. According to\\nApplication Note 3, the court should count only a conviction\\nthat received criminal history points pursuant to the\\n“Criminal History Category” guideline, §§ 4A1.1(a), (b) and\\n(c). Id. § 2L1.2 cmt. 3. These “Criminal History Category”\\nguideline subsections tell a court to add a specified number\\nof criminal history points for each “prior sentence of\\nimprisonment,” depending on the sentence’s length. Id.\\n§§ 4A1.1(a), (b), & (c). The majority then jumps to a\\ndifferent subsection of the “Criminal History Category”\\nguideline, § 4A1.1(e), Maj. Op. 17, which is not mentioned\\nin Application Note 3, id. § 2L1.2 cmt. 3. This\\n(unmentioned) subsection—§ 4A1.1(e)—directs a court to\\nadd a point for any prior sentence resulting from a conviction\\nof a crime of violence that did not receive a point under the\\n“Criminal History Category” guideline, §§ 4A1.1(a), (b) and\\n(c), “because such a sentence was treated as a single\\nsentence,” id. § 4A1.1(e). According to the majority, the\\n\\n convictions that receive criminal history points under\\n §4A1.1(a), (b), or (c). In addition, for purposes of\\n subsections (b)(1)(B), (b)(2)(E), and (b)(3)(E), use\\n only those convictions that are counted separately\\n under §4A1.2(a)(2).\\n\\nA conviction taken into account under subsection (b)(1), (b)(2), or (b)(3)\\nis not excluded from consideration of whether that conviction receives\\ncriminal history points pursuant to Chapter Four, Part A (Criminal\\nHistory).\\n\\n UNITED STATES V. CUEVAS-LOPEZ 37\\n\\nlanguage in this subsection of the Criminal History Category\\nguidelines, § 4A1.1(e) (although not mentioned in\\nApplication Note 3), shows that the Single Sentence Rule is\\nincorporated into determining whether a defendant’s\\ncriminal conduct resulted in “a conviction” for an offense\\n“for which the sentence imposed” was five years or more\\nunder subsection (3)(A) of the Unlawful Reentry Offense\\nguidelines. Maj. Op. 17.\\n\\n This attenuated reasoning for discerning the Sentencing\\nCommission’s intent is groundless. “[T]he primary\\ntouchstone for discovering that intent is the text of the\\nGuidelines manual” itself. Joey, 845 F.3d at 1297. On its\\nface, the instruction under the subsection of the “Criminal\\nHistory Category” guideline, § 4A1.1(e) (i.e., that a court\\nshould add a point for any prior sentence resulting from a\\nconviction of a crime of violence that did not receive a point\\nbecause it was treated as a single sentence), is completely\\nirrelevant to the question before us. See U.S.S.G.\\n§ 4A1.1(e). At a minimum, this instruction about treatment\\nof aggregated sentences for crimes of violence does not\\ndirect a court to consider Cuevas-Lopez’s two convictions\\nand sentences for second degree burglary as “a conviction”\\nfor which a single, seven-year sentence was imposed. See\\nid. There is nothing in this unrelated guideline section that\\ncould override the plain text of subsection 3(A), the\\napplicable Unlawful Reentry Offense guideline.\\n\\n One more example of the majority’s selective tour\\nthrough the Guidelines will suffice. Application Note 3 to\\nthe Unlawful Reentry Offense guidelines also explains that\\nif the defendant has prior convictions for multiple\\nmisdemeanors, the court should count only those\\nconvictions that would be counted separately under the\\nSingle Sentence Rule. Maj. Op. 17–18. The majority argues\\n\\n38 UNITED STATES V. CUEVAS-LOPEZ\\n\\nthat this portion of Application Note 3 “suggests” that a court\\nshould apply the Single Sentence Rule if a defendant has\\nprior convictions for multiple felonies under subsection\\n(3)(A). Maj. Op. 17–18. But of course, the language from\\nApplication Note 3 raises exactly the opposite inference: the\\nSentencing Commission knew how to instruct a court to\\nconsider the Single Sentence Rule for multiple\\nmisdemeanors, and did not do so for felony counts described\\nin subsection (3)(A). See, e.g., Whitfield v. United States,\\n543 U.S. 209, 216 (2005) (“Congress has included an\\nexpress overt-act requirement in at least 22 other current\\nconspiracy statutes, clearly demonstrating that it knows how\\nto impose such a requirement when it wishes to do so.”).\\n\\n The majority’s remaining catalog of textual arguments\\nbased on irrelevant Guidelines sections are equally\\nunsupported, and merit no further discussion here. 9\\n\\n B\\n\\n The majority’s arguments based on the Amendment to\\nSection 2L1.2 fare no better. Maj. Op. 19–22.\\n\\n Prior to the 2016 amendments, § 2L1.2(b) defined the\\nSpecial Offense Characteristics enhancements by reference\\nto federal generic offenses. See U.S.S.G. § 2L1.2 (2015).\\nFor instance, the Specific Offense subsection required the\\nimposition of an enhancement for defendants who\\npreviously had been convicted of a specified drug trafficking\\noffense, crime of violence, firearms offense and the like. Id.\\n\\n 9\\n The majority replicates the errors of the Fourth and the Fifth\\nCircuits in United States v. Martinez-Varela, 531 F.3d 298 (4th Cir.\\n2008) and United States v. Garcia-Sanchez, 916 F.3d 522, 526 (5th Cir.\\n2019). Maj. Op. 23–25. Because those cases are also contrary to the\\nplain text of the Guidelines, we should decline to follow them.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 39\\n\\n§ 2L1.2(b)(1)(A) (2015). In order to determine whether a\\nprior conviction qualified as one of these categories, the\\ncourt was required to use the categorical approach set forth\\nin Taylor v. United States, 495 U.S. 575 (1990), and its\\nprogeny. See U.S.S.G. § 2L1.2 (2015). Courts complained\\nthat applying this categorical approach was “overly complex\\nand resource-intensive.” U.S.S.G. Supp. to 5 app. C, amend.\\n802, at 155 (Nov. 1, 2016). Consequently, in 2016, the\\nSentencing Commission decided to apply “a much simpler\\nsentence-imposed model” in which “[t]he level of the\\nsentencing enhancement for a prior conviction generally will\\nbe determined by the length of the sentence imposed for the\\nprior offense.” Id. It also noted that the “length of sentence\\nimposed by a sentencing court is a strong indicator of the\\ncourt’s assessment of the seriousness of the predicate offense\\nat the time.” Id. at 157.\\n\\n The majority argues that because the Amendment states\\nthat the new approach for accounting for prior convictions\\nwas “similar to how Chapter Four of the Guidelines Manual\\ndetermines a defendant’s criminal history score based on his\\nor her prior convictions,” id. at 155–56, the Sentencing\\nCommission must have intended to incorporate Chapter 4’s\\nSingle Sentence Rule into the Unlawful Reentry Offense\\nguideline, Maj. Op. at 20–21. Like the majority’s other\\narguments, this one too is meritless. First, the Amendment’s\\nbrief references to Chapter Four of the Guidelines cannot\\noverride the plain language of the Guidelines text. Stinson,\\n508 U.S. at 38 (holding that Guidelines commentary is not\\nauthoritative if it is inconsistent with the guideline itself).\\nBut here, there is no inconsistency between the Amendment\\nand the Guidelines, because the Amendment does not even\\nreference the Single Sentence Rule, and there is thus no basis\\nfor concluding the rule is incorporated in the Unlawful\\n\\n40 UNITED STATES V. CUEVAS-LOPEZ\\n\\nReentry Offense guideline. See U.S.S.G. Supp. to 5 app. C,\\namend. 802, at 155–57 (Nov. 1, 2016).\\n\\n Moreover, even if it were proper to stray from the text of\\nthe Guidelines, the reasoning expressed in the Amendment\\nis more consistent with the conclusion that courts should not\\napply the Single Sentence Rule when determining the\\nSpecial Offense Characteristics under the Unlawful Reentry\\nOffense guidelines. Id. The Commission explained that it\\nused the “the length of sentence imposed” because the length\\nof a sentence “is a strong indicator of the court’s assessment\\nof the seriousness of the predicate offense.” Id. at 157. In\\nother words, the length of the sentence serves as a proxy for\\nthe seriousness of the crime. But if that was the\\nCommission’s intent, then it is illogical to aggregate crimes.\\nPlainly, two second-degree burglaries which each receive\\nthree-year sentences are not as serious as, for example, an\\narmed robbery which results in a six-year sentence. Thus, to\\nthe extent that the Amendments are relevant, they support\\nthe same outcome. 10\\n\\n\\n\\n\\n 10\\n The majority asserts that 18 U.S.C. § 3584 (which requires a\\nsentencing court to consider various factors when determining whether\\nprison terms imposed for each offense are to run consecutively or\\nconcurrently) supports its argument that the Single Sentence Rule\\ncaptures the seriousness of a defendant’s prior offense, and therefore is\\na “baseline assumption” underlying the Guidelines. Maj. Op. 18, 22. Of\\ncourse, § 3584 provides no guidance on how to interpret the Unlawful\\nReentry Offense guidelines. Moreover, the court’s duty to determine the\\nappropriate total sentence for the defendant under § 3584, see Dean v.\\nUnited States, 137 S. Ct. 1170, 1175–76 (2017), is distinct from the\\nquestion before us here: whether a single offense is sufficiently serious\\nto warrant an enhancement.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 41\\n\\n ***\\n\\n Under the plain language of the Guidelines, Cuevas-\\nLopez does not have “a conviction for a felony offense . . .\\nfor which the sentence imposed was five years or more.”\\nU.S.S.G. § 2L1.2. This “language is plain and admits of no\\nmore than one meaning,” and therefore, our “sole function”\\nis to enforce the terms of the Guidelines pursuant to their\\nplain meaning. Carson Harbor Vill., Ltd., 270 F.3d at 878.\\nHere the majority concedes there is no “grievous ambiguity\\nor uncertainty” in the Guidelines here, Maj. Op. 25 n.14, but\\nnevertheless declines to give Cuevas-Lopez the benefit of\\nthe Guidelines’ plain language. Because we should not\\nincrease the applicable Guidelines range based solely on\\ninferences regarding the Sentencing Commission’s\\nunspoken intent, Cuevas-Lopez is entitled to a Guidelines\\nrange of 30 to 37 months, rather than the 37 to 46 months\\nrange erroneously calculated by the district court. I dissent.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426679/\", \"author_raw\": \"IKUTA, Circuit Judge, dissenting\"}]}"},"judge1":{"kind":"string","value":"RICHARD R CLIFTON"},"judge2":{"kind":"string","value":"SANDRA S IKUTA"},"judge3":{"kind":"string","value":"MICHELLE T FRIEDLAND"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649426/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Vicente Cuevas-Lopez\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-19\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-10438\n Plaintiff-Appellee,\n D.C. No.\n v. 4:17-cr-00306-\n JGZ-DTF-1\nVICENTE CUEVAS-LOPEZ, AKA\nVicente Cuevas Lopez,\n Defendant-Appellant. OPINION\n\n Appeal from the United States District Court\n for the District of Arizona\n Jennifer G. Zipps, District Judge, Presiding\n\n Argued and Submitted March 6, 2019\n Phoenix, Arizona\n\n Filed August 19, 2019\n\n Before: Richard R. Clifton, Sandra S. Ikuta,\n and Michelle T. Friedland, Circuit Judges.\n\n Opinion by Judge Friedland;\n Dissent by Judge Ikuta\n\n2 UNITED STATES V. CUEVAS-LOPEZ\n\n SUMMARY *\n\n\n Criminal Law\n\n Affirming a sentence for attempted illegal reentry after\ndeportation in violation of 8 U.S.C. § 1326, the panel held\nthat the “single sentence rule” in U.S.S.G. § 4A1.2(a)(2)\napplies to the enhancements in U.S.S.G. § 2L1.2(b)(2) and\n(b)(3).\n\n The single sentence rule instructs that whether to treat\nmultiple prior sentences as a single sentence depends on\nwhether they were separated by an intervening arrest,\ncharged in the same instrument, or imposed on the same day;\nand provides that if prior sentences are treated as a single\nsentence, a court should use the longest sentence of\nimprisonment if concurrent sentences were imposed and use\nthe aggregate sentence of imprisonment if consecutive\nsentences were imposed. A state court had previously\nsentenced the defendant to two consecutive 3.5-year terms\nimposed on the same day for two second-degree burglary\nconvictions.\n\n Because the single sentence rule applies to § 2L1.2, the\npanel concluded that the district court properly relied on the\nrule to aggregate the defendant’s two consecutive 3.5-year\nsentences in applying a ten-level enhancement pursuant to\nU.S.S.G. § 2L1.2(b)(3)(A), which applies to a defendant\ncharged under § 1326 who was previously ordered deported\nor removed and who subsequently committed a felony\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n UNITED STATES V. CUEVAS-LOPEZ 3\n\noffense for which the sentence imposed was five years or\nmore.\n\n Dissenting, Judge Ikuta wrote that under the plain\nlanguage of the Sentencing Guidelines, the defendant does\nnot have “a conviction for a felony offense . . . for which the\nsentence imposed was five years or more,” U.S.S.G.\n§ 2L1.2; and that the applicable Guidelines range should not\nbe increased based solely on inferences regarding the\nSentencing Commission’s unspoken intent.\n\n\n COUNSEL\n\nJeffrey G. Buchella (argued), Tucson, Arizona, for\nDefendant-Appellant.\n\nCorey J. Mantei (argued), Assistant United States Attorney;\nElizabeth A. Strange, First Assistant United States Attorney;\nRobert L. Miskell, Appellate Chief; United States Attorney’s\nOffice, Tucson, Arizona; for Plaintiff-Appellee.\n\n\n OPINION\n\nFRIEDLAND, Circuit Judge:\n\n Defendant-Appellant Vicente Cuevas-Lopez pleaded\nguilty to attempted illegal reentry after deportation in\nviolation of 8 U.S.C. § 1326. In determining Cuevas-\nLopez’s sentence, the district court applied a ten-level\nenhancement to his base offense level pursuant to United\nStates Sentencing Guidelines Manual (“U.S.S.G.” or\n“Guidelines Manual”) § 2L1.2(b)(3)(A) (“the\nEnhancement”), as had been recommended by the Pre-\n\n4 UNITED STATES V. CUEVAS-LOPEZ\n\nSentence Report. 1 The Enhancement applies to a defendant\ncharged under 8 U.S.C. § 1326 who was previously ordered\ndeported or removed and who subsequently committed “a\nfelony offense . . . for which the sentence imposed was five\nyears or more.” U.S.S.G. § 2L1.2(b)(3)(A).\n\n A state court had previously sentenced Cuevas-Lopez to\ntwo consecutive 3.5-year terms imposed on the same day for\ntwo second-degree burglary convictions. The district court\naggregated Cuevas-Lopez’s two 3.5-year sentences to\nproduce a seven-year sentence for purposes of applying the\nEnhancement, relying on § 4A1.2(a)(2) of the Guidelines\nManual, which is known as the “single sentence rule.”\nCuevas-Lopez, who did not object at sentencing, now argues\nthat the district court erred in adhering to the single sentence\nrule and thus in aggregating the two sentences when\nconsidering what level of enhancement to apply. We affirm,\njoining the Fifth Circuit in holding that the single sentence\nrule in § 4A1.2(a)(2) governs the determination whether an\nenhancement applies under § 2L1.2(b).\n\n I.\n\n A.\n\n The Sentencing Reform Act of 1984 established “factors\nto guide [federal] district courts in exercising their traditional\n\n 1\n We review Cuevas-Lopez’s sentence based on the 2016 Guidelines\nManual, which was in effect at the time of Cuevas-Lopez’s sentencing.\nU.S.S.G. § 1B1.11 (2018) (“The court shall use the Guidelines Manual\nin effect on the date that the defendant is sentenced,” unless doing so\n“would violate the ex post facto clause of the United States\nConstitution.”); see also United States v. Thomsen, 830 F.3d 1049, 1071\n(9th Cir. 2016). All section references, and all citations to the Guidelines\nManual, are to the 2016 version unless otherwise specified.\n\n UNITED STATES V. CUEVAS-LOPEZ 5\n\nsentencing discretion.” Beckles v. United States, 137 S. Ct.\n886, 893 (2017). Congress simultaneously “created the\nUnited States Sentencing Commission and charged it with\nestablishing guidelines to be used for sentencing.” Id.\nAlthough “[t]he Guidelines were initially binding on district\ncourts,” the Supreme Court in United States v. Booker,\n543 U.S. 220 (2005), “rendered them ‘effectively\nadvisory.’” Beckles, 137 S. Ct. at 894 (quoting Booker,\n543 U.S. at 245). The Guidelines Manual is nonetheless\n‘“the starting point and the initial benchmark’ for\nsentencing.” Id. (quoting Gall v. United States, 552 U.S. 38,\n49 (2007)).\n\n The Guidelines Manual provides sentencing ranges\ndetermined by a combination of “the seriousness of a\ndefendant’s offense . . . and his [or her] criminal history.”\nMolina-Martinez v. United States, 136 S. Ct. 1338, 1342\n(2016). The offense seriousness is reflected in an “offense\nlevel” comprised of a base offense level, which is assigned\nby the Guidelines Manual to each type of conviction;\nspecific offense characteristics, which can increase or\ndecrease the offense level for each offense; and upward and\ndownward adjustments, which can be applied to any offense.\nSee U.S.S.G. § 1B1.1(a)(1)–(5). A defendant is assigned\ncriminal history points based on his or her past criminal\nconduct, which then places the defendant in a criminal\nhistory category between I and VI. See U.S.S.G. § 4A1.1;\nU.S.S.G. ch. 5, pt. A, Sentencing Table. The Guidelines\nManual combines this criminal history category with a\ndefendant’s calculated offense level to produce a\nrecommended sentencing range. U.S.S.G. ch. 5, pt. A,\nSentencing Table.\n\n6 UNITED STATES V. CUEVAS-LOPEZ\n\n B.\n\n Cuevas-Lopez, who is a citizen of Mexico, was ordered\ndeported from the United States in 2004, and was deported\nseveral times between 2004 and 2015. He unsuccessfully\nattempted to reenter the country in 2017. He was then\ncharged with attempted illegal reentry after deportation, in\nviolation of 8 U.S.C. § 1326(a), as enhanced by 8 U.S.C.\n§ 1326(b)(1). Cuevas-Lopez pled guilty to the charge, and\nthe district court held a sentencing hearing in October 2017.\n\n In advance of the hearing, the U.S. Probation Office\nprepared a Pre-Sentence Report (“PSR”). Applying\n§ 2L1.2(a), the relevant offense guideline for illegal reentry\noffenses, the PSR recommended a base offense level of eight\nfor a violation of 8 U.S.C. § 1326. With regard to specific\noffense characteristics, the PSR added a ten-level\nenhancement pursuant to § 2L1.2(b)(3)(A) based on the fact\nthat “[a]fter [Cuevas-Lopez] was first ordered deported or\nremoved from the United States, he sustained a felony\nconviction for which the sentences imposed in a two-count\nindictment were ordered to run consecutive,” and which\n“resulted in a cumulative sentence that was five years or\nmore.” 2\n\n The ten-level enhancement arose out of Cuevas-Lopez’s\nNovember 3, 2007 arrest for two felony burglaries. The first\nburglary was reported on November 1, 2007, and the second\nwas committed on November 3, 2007. Cuevas-Lopez was\n\n 2\n Although the PSR referred to Cuevas-Lopez as having “sustained\na felony conviction” after first being deported, the relevant question\nunder § 2L1.2(b)(3) is whether a defendant “engaged in criminal\nconduct” after deportation that later resulted in a felony conviction. This\ndifference is immaterial here because Cuevas-Lopez’s relevant conduct\nand resulting convictions all occurred after he was first ordered deported.\n\n UNITED STATES V. CUEVAS-LOPEZ 7\n\nconvicted of both burglaries, and on March 10, 2008 an\nArizona state court sentenced him to 3.5 years in prison for\neach offense, ordered to run consecutively.\n\n The PSR applied the single sentence rule to treat Cuevas-\nLopez’s two consecutive 3.5-year sentences as a single\nseven-year sentence. This aggregated seven-year sentence\ntriggered the ten-level enhancement under § 2L1.2(b)(3)(A)\n(again, the “Enhancement”), so the PSR assigned Cuevas-\nLopez an adjusted offense level of 18 for the instant illegal\nreentry offense. It then applied a three-level downward\nadjustment for acceptance of responsibility, resulting in an\noffense level of 15. The PSR placed Cuevas-Lopez in\ncriminal history category V based on the two burglaries\ndiscussed above and other convictions between 2006 and\n2008. Cuevas-Lopez’s criminal history category and offense\nlevel resulted in a Guidelines range of 37 to 46 months in\nprison. Without aggregating Cuevas-Lopez’s two burglary\nsentences, § 2L1.2(b)(3)(B) would have called for an eight-\nlevel enhancement based on a 3.5-year sentence, which\nwould have resulted in an offense level of 13 after the\ndownward adjustment and a Guidelines range of 30 to\n37 months. See U.S.S.G. ch. 5, pt. A, Sentencing Table.\n\n At the sentencing hearing, the Government agreed with\nthe PSR’s recommendation and argued, based on the\n§ 2L1.2(b)(3)(A) Enhancement, that Cuevas-Lopez’s base\noffense level of eight was “properly enhanced by ten levels\nbecause the defendant received a seven-year consecutive\nstate sentence . . . for two second-degree burglary counts\ncharged in a single indictment.” The district court adopted\nthe Guidelines calculation in the PSR, to which Cuevas-\nLopez did not object. The court sentenced Cuevas-Lopez to\n37 months in prison and three years of supervised release.\n\n8 UNITED STATES V. CUEVAS-LOPEZ\n\n Cuevas-Lopez timely appealed and now argues that the\nsingle sentence rule does not apply to § 2L1.2(b)(3)\nenhancements, and that the district court therefore should not\nhave aggregated his two prior 3.5-year sentences for\npurposes of determining his Guidelines range.\n\n II.\n\n A.\n\n Where, as here, a defendant makes an argument on\nappeal that was not the basis for an objection in the district\ncourt, we generally review for plain error. See United States\nv. Gomez, 725 F.3d 1121, 1125 (9th Cir. 2013) (“If the\ndefendant fails to object, we review for plain error.”); see\nalso Fed. R. Crim P. 52(b) (“A plain error that affects\nsubstantial rights may be considered even though it was not\nbrought to the court’s attention.”). We have held, however,\nthat “we are not limited to [plain error] review when we are\npresented with a question that ‘is purely one of law’ and\nwhere ‘the opposing party will suffer no prejudice as a result\nof the failure to raise the issue in the trial court.’” United\nStates v. Saavedra-Velazquez, 578 F.3d 1103, 1106 (9th Cir.\n2009) (quoting United States v. Echavarria-Escobar,\n270 F.3d 1265, 1267–68 (9th Cir. 2001)).\n\n Cuevas-Lopez urges us to apply de novo review, despite\nhis failure to object in the district court, because his\nargument about when the single sentence rule applies\npresents a pure question of law. The Government responds\nthat our court’s “pure question of law” exception to plain\nerror review is inconsistent with the Supreme Court’s\ninterpretation of Federal Rule of Criminal Procedure 52(b).\nIn support, the Government cites Puckett v. United States,\n556 U.S. 129 (2009), in which the Supreme Court held that\n“[f]ailure to abide by [the] contemporaneous-objection rule\n\n UNITED STATES V. CUEVAS-LOPEZ 9\n\nordinarily precludes the raising on appeal of [an]\nunpreserved claim of trial error,” with a “limited exception,”\nid. at 135, for a “plain error that affects substantial rights,”\nid. (quoting Fed. R. Crim. P. 52(b)). The Government also\nrelies on Judge Graber’s concurrence in United States v.\nZhou, 838 F.3d 1007 (9th Cir. 2016), which opined that\n“[o]ur ‘pure question of law’ exception contradicts Rule\n52(b) and the Supreme Court’s case law.” Id. at 1016\n(Graber, J., concurring). We need not resolve this dispute or\notherwise decide which standard of review applies here,\nbecause we would affirm under either de novo or plain error\nreview. For the reasons explained below, we conclude that\nthe district court correctly interpreted the Guidelines Manual\nin determining that the single sentence rule applies to\n§ 2L1.2(b) enhancements.\n\n B.\n\n 1.\n\n Although the Guidelines are advisory only, a “district\ncourt must correctly calculate the recommended Guidelines\nsentence and use that recommendation as the ‘starting point\nand the initial benchmark.’” United States v. Munoz-\nCamarena, 631 F.3d 1028, 1030 (9th Cir. 2011) (quoting\nKimbrough v. United States, 552 U.S. 85, 108 (2007)). The\ncourt must keep the Guidelines range “in mind throughout\nthe process,” id. (quoting United States v. Carty, 520 F.3d\n984, 991 (9th Cir. 2008) (en banc)), and “justify the extent\nof [any] departure from the Guidelines,” id. “Failure to\ncalculate the correct Guidelines range constitutes procedural\nerror.” Peugh v. United States, 569 U.S. 530, 537 (2013).\n\n As the Supreme Court has explained, the Guidelines\nManual “contains text of three varieties,” each of which is\nwritten by the Sentencing Commission. Stinson v. United\n\n10 UNITED STATES V. CUEVAS-LOPEZ\n\nStates, 508 U.S. 36, 41 (1993). The “[f]irst is a guideline\nprovision itself,” which “provide[s] direction as to the\nappropriate type [and extent] of punishment.” Id. A “second\nvariety of text in the [Guidelines] Manual is a policy\nstatement . . . regarding application of the guidelines or other\naspects of sentencing that would further the purposes of the\n[Sentencing Reform] Act.” Id. (quotation marks omitted).\nThe third type of text in the Guidelines Manual is\ncommentary, which accompanies both guidelines and policy\nstatements. Id. Commentary may serve three functions: to\n“interpret a guideline or explain how it is to be applied”; to\n“suggest circumstances which may warrant departure from\nthe guidelines”; and to “provide background information,\nincluding factors considered in promulgating the guideline\nor reasons underlying promulgation of the guideline.” Id.\n(alterations and citation omitted). Any modifications or\namendments to the guidelines provisions themselves (the\nfirst category of text) must be accompanied by a “statement\nof the reasons therefor” authored by the Sentencing\nCommission, and take effect on a date set by the\nCommission, within certain statutory parameters and subject\nto any changes made by Congress. See 28 U.S.C. § 994(p).\n\n “We interpret the Sentencing Guidelines using the\nordinary tools of statutory interpretation.” United States v.\nMartinez, 870 F.3d 1163, 1166 (9th Cir. 2017). Our\ninterpretation “will most often begin and end with the text\nand structure of the [g]uidelines” provisions themselves. Id.\n(quoting United States v. Joey, 845 F.3d 1291, 1297 n.8 (9th\nCir. 2017)). We also consider “the Commission’s\ncommentary interpreting or explaining the text” of those\nguidelines provisions. Id. The commentary “is authoritative\nunless it violates the Constitution or a federal statute, or is\ninconsistent with, or a plainly erroneous reading of, that\nguideline.” Stinson, 508 U.S. at 38; see also United States\n\n UNITED STATES V. CUEVAS-LOPEZ 11\n\nv. Prien-Pinto, 917 F.3d 1155, 1157–58 (9th Cir. 2019)\n(following Stinson). Policy statements, likewise, are binding\n“[t]o the extent that they interpret substantive guidelines and\ndo not conflict with them or any statutory directives.”\nUnited States v. Chea, 231 F.3d 531, 536 n.1 (9th Cir. 2000)\n(citation omitted); see also Stinson, 508 U.S. at 42 (“The\nprinciple that the Guidelines Manual is binding on federal\ncourts applies as well to policy statements.”). “We may also\nlook to [a] provision’s history and purpose, such as by\nconsulting the Commission’s statements of reason for a\nparticular amendment.” Martinez, 870 F.3d at 1166 (citation\nand quotation marks omitted); see also United States v.\nOrnelas, 825 F.3d 548, 554 (9th Cir. 2016) (noting that we\n“look to the Commission’s statements of reason for\nguidance” in interpreting individual guidelines).\n\n 2.\n\n Chapter Two of the Guidelines Manual governs offense\nconduct. The offense guideline within Chapter Two\napplicable to a conviction for attempted illegal reentry is\n§ 2L1.2. 3 Section 2L1.2(a) provides a base offense level of\neight. Sections 2L1.2(b)(1), (b)(2), and (b)(3), which\ncontain the specific offense characteristics, apply\nenhancements of between two and ten levels based on a\ndefendant’s prior convictions. Some of these enhancements\ndepend on the types of convictions sustained and others\n\n 3\n The guideline applicable to a substantive offense generally also\napplies to inchoate offenses such as attempt. See U.S.S.G. § 2X1.1\n(2018). The parties do not dispute that § 2L1.2, which is titled\n“Unlawfully Entering or Remaining in the United States,” applies to\nCuevas-Lopez’s attempted illegal reentry. See United States v. Rosales-\nAguilar, 818 F.3d 965, 972 (9th Cir. 2016) (applying § 2L1.2 to an\nattempted illegal reentry case).\n\n12 UNITED STATES V. CUEVAS-LOPEZ\n\ndepend on the length of sentences received. In\n§ 2L1.2(b)(1), § 2L1.2(b)(2)(E), and § 2L1.2(b)(3)(E), the\nsize of the enhancement turns on the type of offense. 4 The\nother subsections in § 2L1.2(b)(2) and § 2L1.2(b)(3),\nincluding the Enhancement, add varying levels of\nenhancement based on the length of a defendant’s sentence\nimposed for prior convictions.\n\n Subsection (b)(3), which the district court applied to\nenhance Cuevas-Lopez’s base offense level in this case,\nprovides in full:\n\n (3) (Apply the Greatest) If, at any time after\n the defendant was ordered deported or\n ordered removed from the United States for\n the first time, the defendant engaged in\n criminal conduct resulting in—\n\n (A) a conviction for a felony offense (other\n than an illegal reentry offense) for which\n the sentence imposed was five years or\n more, increase by 10 levels;\n\n (B) a conviction for a felony offense (other\n than an illegal reentry offense) for which\n\n\n 4\n In § 2L1.2(b)(1), the level of enhancement depends on whether the\ndefendant, before committing the offense for which he is being\nsentenced, sustained “a conviction for a felony that is an illegal reentry\noffense” or sustained “two or more convictions for [improper entry]\nmisdemeanors under 8 U.S.C. § 1325(a).” U.S.S.G. § 2L1.2(b)(1). Both\n(b)(2)(E) and (b)(3)(E) apply a two-level enhancement where a\ndefendant has “three or more convictions for misdemeanors that are\ncrimes of violence or drug trafficking offenses.”\nU.S.S.G. § 2L1.2(b)(2)(E), (b)(3)(E).\n\n UNITED STATES V. CUEVAS-LOPEZ 13\n\n the sentence imposed was two years or\n more, increase by 8 levels;\n\n (C) a conviction for a felony offense (other\n than an illegal reentry offense) for which\n the sentence imposed exceeded one year\n and one month, increase by 6 levels;\n\n (D) a conviction for any other felony offense\n (other than an illegal reentry offense),\n increase by 4 levels; or\n\n (E) three or more convictions for\n misdemeanors that are crimes of violence\n or drug trafficking offenses, increase by\n 2 levels.\n\nU.S.S.G. § 2L1.2(b)(3) (underlining added). Subsection\n2L1.2(b)(2) mirrors subsection (b)(3), but relates to a\ndefendant’s convictions, if any, sustained before being\nordered deported or removed. See § 2L1.2(b)(2). 5\n\n\n 5\n Although we apply the 2016 Guidelines Manual here, we note that\n§ 2L1.2 was amended in 2018 to “establish[] that the application of the\n§ 2L1.2(b)(2) enhancement depends on the timing of the underlying\n‘criminal conduct,’ and not on the timing of the resulting conviction.”\nU.S.S.G. Supp. to app. C, amend. 809 at 188 (Nov. 1, 2018). The\namended § 2L1.2(b)(2) applies when “before the defendant was ordered\ndeported or ordered removed . . . for the first time, the defendant engaged\nin criminal conduct that, at any time, resulted in” the various convictions\nand sentences enumerated in that section. U.S.S.G. § 2L1.2(b)(2)\n(2018). The Sentencing Commission simultaneously updated subsection\n(b)(3) to mirror the language in subsection (b)(2), see § 2L1.2(b)(3)\n(2018), though subsection (b)(3), unlike (b)(2), had already (before the\n2018 amendment) depended upon the timing of the underlying conduct,\nrather than the underlying conviction, see § 2L1.2(b)(3) (2016).\n\n14 UNITED STATES V. CUEVAS-LOPEZ\n\n Cuevas-Lopez contends that the words “a conviction” in\nthe Enhancement (underlined above) preclude a district\ncourt from aggregating consecutive sentences for purposes\nof applying the Enhancement. Although this contention\nfinds some support in the language of § 2L1.2(b)(3), we are\npersuaded in light of the application notes in the commentary\nto § 2L1.2, as well as the Sentencing Commission’s\nstatement of reasons for the 2016 amendment to § 2L1.2,\nthat the district court properly aggregated Cuevas-Lopez’s\ntwo 3.5-year sentences here.\n\n In aggregating Cuevas-Lopez’s two sentences, the\ndistrict court relied on the single sentence rule. The single\nsentence rule falls within Chapter Four of the Guidelines\nManual, which contains guidelines for categorizing a\ndefendant’s criminal history. Section 4A1.2(a) of Chapter\nFour defines “prior sentence” as “any sentence previously\nimposed upon adjudication of guilt, whether by guilty plea,\ntrial, or plea of nolo contendere, for conduct not part of the\ninstant offense.” U.S.S.G. § 4A1.2(a)(1). The guideline\nthen instructs that whether to treat multiple prior sentences\nas a single sentence depends on whether they were separated\nby an intervening arrest, charged in the same instrument, or\nimposed on the same day. Specifically, § 4A1.2(a)(2) states:\n\n If the defendant has multiple prior sentences,\n determine whether those sentences are\n counted separately or treated as a single\n sentence. Prior sentences always are counted\n separately if the sentences were imposed for\n offenses that were separated by an\n intervening arrest (i.e., the defendant is\n arrested for the first offense prior to\n committing the second offense). If there is\n no intervening arrest, prior sentences are\n\n UNITED STATES V. CUEVAS-LOPEZ 15\n\n counted separately unless (A) the sentences\n resulted from offenses contained in the same\n charging instrument; or (B) the sentences\n were imposed on the same day. Treat any\n prior sentence covered by (A) or (B) as a\n single sentence. See also § 4A1.1(e).\n\nU.S.S.G. § 4A1.2(a)(2).\n\n Section 4A1.2(a)(2) goes on to provide that “if prior\nsentences are treated as a single sentence,” a court should\n“use the longest sentence of imprisonment if concurrent\nsentences were imposed” and “use the aggregate sentence of\nimprisonment” where “consecutive sentences were\nimposed.” U.S.S.G. § 4A1.2(a)(2). Here, the district court\napplied a ten-level enhancement based on Cuevas-Lopez’s\nconsecutive 3.5-year sentences, rather than an eight-level\nenhancement, which would have applied had Cuevas-\nLopez’s two 3.5-year sentences been ordered to run\nconcurrently (or if they had been ordered to run\nconsecutively but the single sentence rule’s aggregation\nprovisions did not apply). Our task, therefore, is to\ndetermine whether Chapter Four’s single sentence rule\nproperly applies to § 2L1.2.\n\n In support of his argument that the single sentence rule’s\naggregation provisions should not apply here, Cuevas-Lopez\nrelies on § 2L1.2’s commentary—specifically Application\nNote 2. Application Note 2 to § 2L1.2 gives “sentence\nimposed” “the meaning given the term ‘sentence of\nimprisonment’ in Application Note 2 and subsection (b) of\n§ 4A1.2.” U.S.S.G. § 2L1.2 cmt. n.2. As Cuevas-Lopez\npoints out, this application note to § 2L1.2 does not reference\nsubsection (a) of § 4A1.2, which defines “prior sentence”\nand contains the single sentence rule. Cuevas-Lopez argues\n\n16 UNITED STATES V. CUEVAS-LOPEZ\n\nthat because Application Note 2 specifically cross-\nreferences a part of § 4A1.2 and yet does not reference the\npart containing the single sentence rule, the Commission\nmeant to exclude the single sentence rule from applying to\nthe § 2L1.2(b) enhancements.\n\n In our view, the absence of a mention of subsection (a)\nof § 4A1.2 in Application Note 2 to § 2L1.2 does not hold\nthe significance Cuevas-Lopez wishes. 6 Rather, we read\n§ 4A1.2(b) and Application Note 2 to that section—the two\nGuidelines Manual components specifically referenced in\n§ 2L1.2’s commentary for purposes of defining “sentence\nimposed”—as simply being silent on the question whether\nseparate sentences that are imposed on the same day and\nordered to run consecutively should be aggregated for\npurposes of applying the Enhancement. Other application\nnotes to the offense guideline in which the Enhancement\nappears, however, speak more to the issue and evince the\nSentencing Commission’s intent that the single sentence rule\napply to § 2L1.2(b).\n\n One example is Application Note 3 to the § 2L1.2\noffense guideline. Note 3 instructs a court that when\n“applying subsections (b)(1), (b)(2), and (b)(3)” of\n§ 2L1.2—i.e., the § 2L1.2(b) enhancements—a court should\n\n 6\n The Guidelines Manual does state, as Cuevas-Lopez argues, that\n“[a]n instruction to use a particular subsection or table from another\noffense guideline refers only to the particular subsection or table\nreferenced, and not to the entire offense guideline.” U.S.S.G.\n§ 1B1.5(b)(2). Section 4A1.2 is not, however, an offense guideline—\nrather, it provides “Definitions and Instructions for Computing Criminal\nHistory.” And, in any event, even if the reference to subsection (b) does\nnot itself amount to an instruction to apply subsection (a), it still leaves\nus with silence on whether or not to apply subsection (a)’s single\nsentence rule.\n\n UNITED STATES V. CUEVAS-LOPEZ 17\n\n“use only those convictions that receive criminal history\npoints under § 4A1.1(a), (b), or (c).” U.S.S.G. § 2L1.2 cmt.\nn.3. Criminal history points are added under § 4A1.1(a) and\n§ 4A1.1(b) based on “prior sentence[s] of imprisonment” of\nover 13 months (for which three points are added) and\nbetween 60 days and 13 months (for which two points are\nadded), respectively. Under § 4A1.1(c), one additional point\nis added for “each prior sentence” not counted in either (a)\nor (b). Subsection (e) to § 4A1.1—which is not referenced\nin § 2L1.2’s Application Note 3—instructs courts to “[a]dd\n1 point for each prior sentence resulting from a conviction of\na crime of violence that did not receive any points under (a),\n(b), or (c) . . . because such sentence was treated as a single\nsentence.” U.S.S.G. § 4A1.1(e) (emphasis added). The\nupshot is that certain “prior sentence[s] of imprisonment” in\n§ 4A1.1(a)–(c) will actually sometimes consist of\naggregated sentences from multiple separate convictions\npursuant to the single sentence rule. Because Application\nNote 3 to § 2L1.2 instructs courts, for purposes of applying\nthe § 2L1.2(b) enhancements, to use only convictions that\nreceive criminal history points under subsections (a)–(c) of\n§ 4A1.1, it suggests that whether and how a conviction\nshould be counted for purposes of a § 2L1.2(b) enhancement\ndepends on the operation of the single sentence rule.\n\n This understanding is reinforced by a second directive in\nApplication Note 3 to § 2L1.2—that “for purposes of\nsubsections (b)(1)(B), (b)(2)(E), and (b)(3)(E)” 7 of that\nsection, courts “use only those convictions that are counted\nseparately under [the single sentence rule in] § 4A1.2(a)(2).”\nU.S.S.G. § 2L1.2 cmt. n.3. This directive further suggests\n\n 7\n These subsections retained a categorical approach after the 2016\namendment to § 2L1.2, discussed below, and deal with prior\nmisdemeanor convictions. See infra n.9.\n\n18 UNITED STATES V. CUEVAS-LOPEZ\n\nthat the Sentencing Commission anticipated that in some\ninstances two or more convictions could be counted—and\nreferred to—as one under the single sentence rule, and that a\ncourt should do so for purposes of other subsections of\n§ 2L1.2. Were this not the baseline assumption, there would\nbe no need to explicitly make the single sentence rule\ninapplicable to these subsections. Application Note 3\ntherefore weighs in favor of applying the single sentence rule\nto determine which § 2L1.2(b)(3) enhancement applies.\n\n Application Note 4 to § 2L1.2, which deals with cases in\nwhich a sentence for an illegal reentry offense was imposed\nat the same time as one for another felony offense, lends\nsome additional support for applying the single sentence\nrule. It provides:\n\n There may be cases in which the sentences\n for an illegal reentry offense and another\n felony offense were imposed at the same time\n and treated as a single sentence for purposes\n of calculating the criminal history score\n under § 4A1.1(a), (b), and (c). In such a case,\n use the illegal reentry offense in determining\n the appropriate enhancement under\n subsection (b)(1), if it independently would\n have received criminal history points. In\n addition, use the prior sentence for the other\n felony offense in determining the appropriate\n enhancement under subsection (b)(3), if it\n independently would have received criminal\n history points.\n\nU.S.S.G. § 2L1.2 cmt. n.4 (emphasis added). Like Note 3,\nthis application note appears to assume that § 4A1.2(a)’s\n\n UNITED STATES V. CUEVAS-LOPEZ 19\n\nsingle sentence rule would normally apply to § 2L1.2(b)\nenhancements.\n\n 3.\n\n The Sentencing Commission’s statement of reasons for\nthe amendment to § 2L1.2 that created the version applicable\nhere also supports the district court’s application of the\nsingle sentence rule. See Ornelas, 825 F.3d at 554 (looking\n“to the Commission’s statements of reason for guidance” in\ninterpreting a guideline). Before that 2016 amendment,\n§ 2L1.2(b) enhancements were based on “the nature of a\ndefendant’s most serious conviction,” as determined by the\n“categorical approach to the penal statute underlying the\nprior conviction.” U.S.S.G. Supp. to app. C, amend. 802\nat155 (Nov. 1, 2016) (“Amendment 802”). Under the\ncategorical approach, courts “compare the elements of the\nstatute forming the basis of the defendant’s conviction with\nthe elements of the ‘generic’ crime—i.e., the offense as\ncommonly understood.” Descamps v. United States,\n570 U.S. 254, 257 (2013). 8 Due to a concern that this\nmethod of determining a level of enhancement was “overly\ncomplex,” the Sentencing Commission adopted a simpler,\nsentence-based model whereby, with a few exceptions, 9\n\n 8\n We note, however, that even before 2016, the § 2L1.2(b)(1)(A) and\n(B) enhancements required a court to use both a categorical approach\nand a sentence-imposed approach, as subsection (A) applied where a\ndefendant sustained “a conviction for a felony offense that [was] . . . a\ndrug trafficking offense for which the sentence imposed exceeded 13\nmonths,” and subsection (B) applied where a defendant sustained a\nfelony drug trafficking conviction “for which the sentence imposed was\n13 months or less.” U.S.S.G. § 2L1.2(b)(1) (2015).\n 9\n The guideline retained the categorical approach for\n§ 2L1.2(b)(1)(B), (b)(2)(E), and (b)(3)(E), relating to predicate\n\n20 UNITED STATES V. CUEVAS-LOPEZ\n\n“[t]he level of the sentencing enhancement for a prior\nconviction generally [would] be determined by the length of\nthe sentence imposed for the prior offense.” 10 Amendment\n802 at 155.\n\n In a section of the statement of reasons titled\n“Accounting for Other Prior Convictions,” the Sentencing\nCommission analogized the sentence-imposed approach in\nsubsections (b)(2) and (b)(3) “to how Chapter Four of the\nGuidelines Manual determines a defendant’s criminal\nhistory score based on his or her prior convictions”—notably\nreferencing Chapter Four (in which the single sentence rule\nappears) in its entirety. Amendment 802 at 156. “The\n[Sentencing] Commission concluded that the length of\nsentence imposed by a sentencing court is a strong indicator\nof the court’s assessment of the seriousness of the predicate\noffense at the time, . . . consistent with how criminal history\nis generally scored in . . . Chapter Four of the Guidelines\nManual.” Amendment 802 at 157. The Commission also\nwrote, in a section titled “Illegal Reentry”: “The definition\nof ‘sentence imposed’ [in the amended § 2L1.2] is the same\ndefinition that appears in Chapter Four of the Guidelines\nManual.” 11 Amendment 802 at 155. The Sentencing\n\n\nmisdemeanor convictions, as well as § 2L1.2(b)(1)(A), relating to\npredicate illegal reentry felonies.\n 10\n Additionally, before 2016, § 2L1.2(b) included only an\nenhancement for convictions that occurred before deportation. See\nU.S.S.G. § 2L1.2 (2015). Amendment 802 sought to address a concern\nthat, because of this, § 2L1.2 did not sufficiently account for all types of\ncriminal conduct committed by persons charged with illegal reentry.\nAmendment 802 at 155–56.\n 11\n We note that although Chapter Four uses the term “sentence\nimposed,” it does not specifically define it.\n\n UNITED STATES V. CUEVAS-LOPEZ 21\n\nCommission did not limit this explanation to any particular\nprovision of Chapter Four, thereby seeming to incorporate\nChapter Four’s single sentence rule.\n\n Similarly, the Commission wrote, in a section of the\nstatement of reasons titled “Excluding Stale Convictions,”\nthat in the context of § 2L1.2’s specific offense\ncharacteristics, “it is . . . appropriate to employ the criminal\nhistory rules.” Amendment 802 at 159. Because the single\nsentence rule is contained in the part of Chapter Four that\nprescribes the criminal history rules, this statement also\nsuggests that the Commission envisioned that the rule would\napply to the § 2L1.2(b) enhancements.\n\n Perhaps the strongest evidence of the Commission’s\nintent appears in another passage of the statement of reasons\nunder the heading “Application of the ‘Single Sentence\nRule,’” which discusses the 2016 addition of what became\nApplication Note 4 to § 2L1.2. See Amendment 802 at 159.\nAs we have mentioned, Application Note 4 deals with cases\nin which a defendant is sentenced for an illegal reentry\noffense at the same time as another federal felony offense.\nIn such cases, “the illegal reentry offense counts towards\nsubsection (b)(1), while the other felony offense counts\ntowards subsection (b)(3).” Amendment 802 at 159. The\nCommission explained in the statement of reasons that it\nintended “to make a distinction between illegal reentry\noffenses and other types of offenses,” and “concluded that it\nwas appropriate to ensure that such convictions are\nseparately accounted for under the applicable specific\noffense characteristics, even if they might otherwise\nconstitute a ‘single sentence’ under § 4A1.2(a)(2).”\nAmendment 802 at 159. Thus, both the text of Application\nNote 4 and the Commission’s stated reason for adding it are\n\n22 UNITED STATES V. CUEVAS-LOPEZ\n\nbased on an assumption that the single sentence rule would\napply to § 2L1.2.\n\n Applying the single sentence rule to § 2L1.2 also makes\nsense in light of the Sentencing Commission’s purpose in\ndrafting Amendment 802. The amended enhancements’\n“sentence imposed” approach replaced the earlier\ncategorical method and was intended to capture—in a\nsimpler way—the seriousness of a defendant’s prior offense.\nThe single sentence rule requires aggregation only when\nsentences are ordered to run consecutively. The imposition\nof consecutive, rather than concurrent, sentences generally\nreflects a decision by either a sentencing judge or, in some\ncases when sentences are statutorily required to run\nconsecutively, by a legislature, that a consecutive term of\nimprisonment would better reflect the seriousness of a\ndefendant’s conduct as well as the need for deterrence,\neducation or treatment of the defendant, and protection of\nthe public. See 18 U.S.C. § 3584 (stating that “[m]ultiple\nterms of imprisonment imposed at the same time run\nconcurrently unless the court orders or the statute mandates\nthat the terms are to run consecutively,” and directing judges\nto consider the sentencing factors set forth in 18 U.S.C.\n§ 3553 “in determining whether the terms imposed are to be\nordered to run concurrently or consecutively”); 18 U.S.C.\n§ 3553 (sentencing factors); Setser v. United States, 566 U.S.\n231, 236 (2012) (“Judges have long been understood to have\ndiscretion to select whether the sentences they impose will\nrun concurrently or consecutively with respect to other\nsentences that they impose.”). Applying the single sentence\nrule to the § 2L1.2(b)(2) and (b)(3) enhancements therefore\ncaptures the seriousness of a defendant’s prior convictions\nand achieves the same goal that the previous categorical\napproach sought to achieve.\n\n UNITED STATES V. CUEVAS-LOPEZ 23\n\n 4.\n\n Even if there were some ambiguity in how broadly the\nSentencing Commission intended the single sentence rule to\napply, as the dissent’s arguments for a contrary\ninterpretation suggest there may be, the goal of avoiding a\ncircuit split would lead us to hold that the single sentence\nrule applies here. “[A]bsent a strong reason to do so, we will\nnot create a direct conflict with other circuits.” United States\nv. Chavez-Vernaza, 844 F.2d 1368, 1374 (9th Cir. 1987); see\nalso Global Linguist Sols., LLC v. Abdelmeged, 913 F.3d\n921, 923 (9th Cir. 2019) (“[W]e so hold to avoid an\nunnecessary circuit split.”). The Fifth Circuit recently\nbecame the first circuit court to answer whether, under the\n2016 Guidelines Manual, the single sentence rule applies to\n§ 2L1.2(b). In United States v. Garcia-Sanchez, that court\nheld that it does. 916 F.3d 522, 526 (5th Cir. 2019). 12 The\nGarcia-Sanchez decision therefore weighs heavily in favor\nof affirming the district court’s interpretation here.\n\n The Fifth Circuit found support for its conclusion in two\ncomponents of Amendment 802 discussed above. It first\nlooked to the section on “Accounting for Other Prior\nConvictions” in the Sentencing Commission’s statement of\nreasons for Amendment 802, which described the length of\na sentence imposed as “a strong indicator of the court’s\nassessment of the seriousness of the predicate offense,” and\nas “consistent with the Chapter Four criminal history rules.”\nGarcia-Sanchez, 916 F.3d at 527 (quoting Amendment 802\nat 157–58). The Fifth Circuit further relied on the section\n\n 12\n Prior to Garcia-Sanchez, the Fifth Circuit had held in United\nStates v. Ponce-Flores, 900 F.3d 215 (5th Cir. 2018), that because any\nerror could not have been obvious, a district court did not plainly err in\napplying the single sentence rule to a § 2L1.2 enhancement. Id. at 219.\n\n24 UNITED STATES V. CUEVAS-LOPEZ\n\ntitled “Excluding Stale Convictions,” which explains that, in\nthe context of § 2L1.2(b)’s specific offense characteristics,\n“it is . . . appropriate to employ the criminal history rules,”\nwhich contain the single sentence rule. Id. (quoting\nAmendment 802 at 159).\n\n The court in Garcia-Sanchez also found persuasive the\nFourth Circuit’s reasoning in United States v. Martinez-\nVarela, 531 F.3d 298 (4th Cir. 2008). In that case, the\ndistrict court aggregated sentences for purposes of applying\na pre-2016 version of § 2L1.2(b)(1), which called for an\nenhancement if a defendant was “previously convicted of a\ndrug trafficking offense for which the ‘sentence imposed’\nwas greater than thirteen months.” Martinez-Varela,\n531 F.3d at 299. That version of § 2L1.2, like the 2016\nversion, contained an application note that referenced\nApplication Note 2 and subsection (b) of § 4A1.2 for\npurposes of defining “sentence imposed.” Id. at 300. The\nFourth Circuit affirmed, reasoning that although the\nprovisions cross-referenced in application notes to § 2L1.2\ndid not specifically address the issue of aggregating\nsentences, commentary to § 4A1.1 instructed that “§§ 4A1.1\nand 4A1.2 must be read together,” providing “strong\nevidence that these two provisions should be read together\nin determining [the defendant’s] criminal history points,”\nand thus that the single sentence rule should apply to the\n§ 2L1.2(b)(1) enhancement at issue. Id. at 301–02.\n\n Although our court had no previous occasion to reach the\nsentencing question presented here, the holding in Garcia-\nSanchez and the reasoning in Martinez-Varela are consistent\nwith our case law. We previously recognized that\n“[s]ections 2L1.2(b) and 4A1.1-2 serve the same underlying\nfunction” of “determin[ing] the extent to which prior\nconvictions affect a defendant’s sentence for the current\n\n UNITED STATES V. CUEVAS-LOPEZ 25\n\noffense,” and that it therefore “makes sense to treat prior\nsentences in the same manner when they are used to\ndetermine the offense level under section 2L1.2 as when\nthey are used to determine the criminal history category.”\nUnited States v. Ortiz-Gutierrez, 36 F.3d 80, 82 (9th Cir.\n1994); 13 see also United States v. Moreno-Cisneros,\n319 F.3d 456, 458–59 (9th Cir. 2003) (noting that\n“Guideline § 4A1.2 is analogous to § 2L1.2(b)(1),” and that\n§ 4A1.2 is “broadly applicable”); United States v. Frias,\n338 F.3d 206, 210 (3d Cir. 2003) (describing Chapter 4 and\nan earlier version of § 2L1.2 as “clearly aimed at the same\nthing, which is varying the punishment based on the criminal\nrecord,” and concluding that it therefore made sense to refer\nto Chapter Four’s definitions in applying § 2L1.2 (quotation\nmarks omitted)); United States v. Galicia-Delgado, 130 F.3d\n518, 521 (2d Cir. 1997) (“[T]he definitions found in § 4A1.2\nhave often been borrowed to interpret terms in § 2L1.2.”).\n\n We thus now join the Fifth Circuit in holding that the\nsingle sentence rule applies to the enhancements in\n§ 2L1.2(b)(2) and (b)(3). 14\n\n\n 13\n In Ortiz-Gutierrez, we applied the single sentence rule to another\npre-2016 enhancement under § 2L1.2 that turned on whether a\ndefendant’s predicate offense was an aggravated felony, which at the\ntime “include[d] a crime of violence for which the defendant was\nsentenced to at least five years imprisonment.” 36 F.3d at 82. Section\n2L1.2 did not cross-reference either § 4A1.1 or § 4A1.2. Id.\n 14\n We reject Cuevas-Lopez’s argument that we should apply the rule\nof lenity to hold in his favor. “[T]he rule of lenity applies to the\nSentencing Guidelines,” but “only . . . where there is grievous ambiguity\nor uncertainty in the guidelines.” United States v. D.M., 869 F.3d 1133,\n1144 (9th Cir. 2017). We find no such “grievous ambiguity or\nuncertainty” in the question of Guidelines Manual interpretation\npresented here.\n\n26 UNITED STATES V. CUEVAS-LOPEZ\n\n III.\n\n For the foregoing reasons, we reject Cuevas-Lopez’s\nargument that the district court was wrong to apply the single\nsentence rule in calculating his sentence. We therefore\nAFFIRM.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Vicente Cuevas-Lopez\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-19\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10438\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 4:17-cr-00306-\\n JGZ-DTF-1\\nVICENTE CUEVAS-LOPEZ, AKA\\nVicente Cuevas Lopez,\\n Defendant-Appellant. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Jennifer G. Zipps, District Judge, Presiding\\n\\n Argued and Submitted March 6, 2019\\n Phoenix, Arizona\\n\\n Filed August 19, 2019\\n\\n Before: Richard R. Clifton, Sandra S. Ikuta,\\n and Michelle T. Friedland, Circuit Judges.\\n\\n Opinion by Judge Friedland;\\n Dissent by Judge Ikuta\\n\\n2 UNITED STATES V. CUEVAS-LOPEZ\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n Affirming a sentence for attempted illegal reentry after\\ndeportation in violation of 8 U.S.C. § 1326, the panel held\\nthat the “single sentence rule” in U.S.S.G. § 4A1.2(a)(2)\\napplies to the enhancements in U.S.S.G. § 2L1.2(b)(2) and\\n(b)(3).\\n\\n The single sentence rule instructs that whether to treat\\nmultiple prior sentences as a single sentence depends on\\nwhether they were separated by an intervening arrest,\\ncharged in the same instrument, or imposed on the same day;\\nand provides that if prior sentences are treated as a single\\nsentence, a court should use the longest sentence of\\nimprisonment if concurrent sentences were imposed and use\\nthe aggregate sentence of imprisonment if consecutive\\nsentences were imposed. A state court had previously\\nsentenced the defendant to two consecutive 3.5-year terms\\nimposed on the same day for two second-degree burglary\\nconvictions.\\n\\n Because the single sentence rule applies to § 2L1.2, the\\npanel concluded that the district court properly relied on the\\nrule to aggregate the defendant’s two consecutive 3.5-year\\nsentences in applying a ten-level enhancement pursuant to\\nU.S.S.G. § 2L1.2(b)(3)(A), which applies to a defendant\\ncharged under § 1326 who was previously ordered deported\\nor removed and who subsequently committed a felony\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 3\\n\\noffense for which the sentence imposed was five years or\\nmore.\\n\\n Dissenting, Judge Ikuta wrote that under the plain\\nlanguage of the Sentencing Guidelines, the defendant does\\nnot have “a conviction for a felony offense . . . for which the\\nsentence imposed was five years or more,” U.S.S.G.\\n§ 2L1.2; and that the applicable Guidelines range should not\\nbe increased based solely on inferences regarding the\\nSentencing Commission’s unspoken intent.\\n\\n\\n COUNSEL\\n\\nJeffrey G. Buchella (argued), Tucson, Arizona, for\\nDefendant-Appellant.\\n\\nCorey J. Mantei (argued), Assistant United States Attorney;\\nElizabeth A. Strange, First Assistant United States Attorney;\\nRobert L. Miskell, Appellate Chief; United States Attorney’s\\nOffice, Tucson, Arizona; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nFRIEDLAND, Circuit Judge:\\n\\n Defendant-Appellant Vicente Cuevas-Lopez pleaded\\nguilty to attempted illegal reentry after deportation in\\nviolation of 8 U.S.C. § 1326. In determining Cuevas-\\nLopez’s sentence, the district court applied a ten-level\\nenhancement to his base offense level pursuant to United\\nStates Sentencing Guidelines Manual (“U.S.S.G.” or\\n“Guidelines Manual”) § 2L1.2(b)(3)(A) (“the\\nEnhancement”), as had been recommended by the Pre-\\n\\n4 UNITED STATES V. CUEVAS-LOPEZ\\n\\nSentence Report. 1 The Enhancement applies to a defendant\\ncharged under 8 U.S.C. § 1326 who was previously ordered\\ndeported or removed and who subsequently committed “a\\nfelony offense . . . for which the sentence imposed was five\\nyears or more.” U.S.S.G. § 2L1.2(b)(3)(A).\\n\\n A state court had previously sentenced Cuevas-Lopez to\\ntwo consecutive 3.5-year terms imposed on the same day for\\ntwo second-degree burglary convictions. The district court\\naggregated Cuevas-Lopez’s two 3.5-year sentences to\\nproduce a seven-year sentence for purposes of applying the\\nEnhancement, relying on § 4A1.2(a)(2) of the Guidelines\\nManual, which is known as the “single sentence rule.”\\nCuevas-Lopez, who did not object at sentencing, now argues\\nthat the district court erred in adhering to the single sentence\\nrule and thus in aggregating the two sentences when\\nconsidering what level of enhancement to apply. We affirm,\\njoining the Fifth Circuit in holding that the single sentence\\nrule in § 4A1.2(a)(2) governs the determination whether an\\nenhancement applies under § 2L1.2(b).\\n\\n I.\\n\\n A.\\n\\n The Sentencing Reform Act of 1984 established “factors\\nto guide [federal] district courts in exercising their traditional\\n\\n 1\\n We review Cuevas-Lopez’s sentence based on the 2016 Guidelines\\nManual, which was in effect at the time of Cuevas-Lopez’s sentencing.\\nU.S.S.G. § 1B1.11 (2018) (“The court shall use the Guidelines Manual\\nin effect on the date that the defendant is sentenced,” unless doing so\\n“would violate the ex post facto clause of the United States\\nConstitution.”); see also United States v. Thomsen, 830 F.3d 1049, 1071\\n(9th Cir. 2016). All section references, and all citations to the Guidelines\\nManual, are to the 2016 version unless otherwise specified.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 5\\n\\nsentencing discretion.” Beckles v. United States, 137 S. Ct.\\n886, 893 (2017). Congress simultaneously “created the\\nUnited States Sentencing Commission and charged it with\\nestablishing guidelines to be used for sentencing.” Id.\\nAlthough “[t]he Guidelines were initially binding on district\\ncourts,” the Supreme Court in United States v. Booker,\\n543 U.S. 220 (2005), “rendered them ‘effectively\\nadvisory.’” Beckles, 137 S. Ct. at 894 (quoting Booker,\\n543 U.S. at 245). The Guidelines Manual is nonetheless\\n‘“the starting point and the initial benchmark’ for\\nsentencing.” Id. (quoting Gall v. United States, 552 U.S. 38,\\n49 (2007)).\\n\\n The Guidelines Manual provides sentencing ranges\\ndetermined by a combination of “the seriousness of a\\ndefendant’s offense . . . and his [or her] criminal history.”\\nMolina-Martinez v. United States, 136 S. Ct. 1338, 1342\\n(2016). The offense seriousness is reflected in an “offense\\nlevel” comprised of a base offense level, which is assigned\\nby the Guidelines Manual to each type of conviction;\\nspecific offense characteristics, which can increase or\\ndecrease the offense level for each offense; and upward and\\ndownward adjustments, which can be applied to any offense.\\nSee U.S.S.G. § 1B1.1(a)(1)–(5). A defendant is assigned\\ncriminal history points based on his or her past criminal\\nconduct, which then places the defendant in a criminal\\nhistory category between I and VI. See U.S.S.G. § 4A1.1;\\nU.S.S.G. ch. 5, pt. A, Sentencing Table. The Guidelines\\nManual combines this criminal history category with a\\ndefendant’s calculated offense level to produce a\\nrecommended sentencing range. U.S.S.G. ch. 5, pt. A,\\nSentencing Table.\\n\\n6 UNITED STATES V. CUEVAS-LOPEZ\\n\\n B.\\n\\n Cuevas-Lopez, who is a citizen of Mexico, was ordered\\ndeported from the United States in 2004, and was deported\\nseveral times between 2004 and 2015. He unsuccessfully\\nattempted to reenter the country in 2017. He was then\\ncharged with attempted illegal reentry after deportation, in\\nviolation of 8 U.S.C. § 1326(a), as enhanced by 8 U.S.C.\\n§ 1326(b)(1). Cuevas-Lopez pled guilty to the charge, and\\nthe district court held a sentencing hearing in October 2017.\\n\\n In advance of the hearing, the U.S. Probation Office\\nprepared a Pre-Sentence Report (“PSR”). Applying\\n§ 2L1.2(a), the relevant offense guideline for illegal reentry\\noffenses, the PSR recommended a base offense level of eight\\nfor a violation of 8 U.S.C. § 1326. With regard to specific\\noffense characteristics, the PSR added a ten-level\\nenhancement pursuant to § 2L1.2(b)(3)(A) based on the fact\\nthat “[a]fter [Cuevas-Lopez] was first ordered deported or\\nremoved from the United States, he sustained a felony\\nconviction for which the sentences imposed in a two-count\\nindictment were ordered to run consecutive,” and which\\n“resulted in a cumulative sentence that was five years or\\nmore.” 2\\n\\n The ten-level enhancement arose out of Cuevas-Lopez’s\\nNovember 3, 2007 arrest for two felony burglaries. The first\\nburglary was reported on November 1, 2007, and the second\\nwas committed on November 3, 2007. Cuevas-Lopez was\\n\\n 2\\n Although the PSR referred to Cuevas-Lopez as having “sustained\\na felony conviction” after first being deported, the relevant question\\nunder § 2L1.2(b)(3) is whether a defendant “engaged in criminal\\nconduct” after deportation that later resulted in a felony conviction. This\\ndifference is immaterial here because Cuevas-Lopez’s relevant conduct\\nand resulting convictions all occurred after he was first ordered deported.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 7\\n\\nconvicted of both burglaries, and on March 10, 2008 an\\nArizona state court sentenced him to 3.5 years in prison for\\neach offense, ordered to run consecutively.\\n\\n The PSR applied the single sentence rule to treat Cuevas-\\nLopez’s two consecutive 3.5-year sentences as a single\\nseven-year sentence. This aggregated seven-year sentence\\ntriggered the ten-level enhancement under § 2L1.2(b)(3)(A)\\n(again, the “Enhancement”), so the PSR assigned Cuevas-\\nLopez an adjusted offense level of 18 for the instant illegal\\nreentry offense. It then applied a three-level downward\\nadjustment for acceptance of responsibility, resulting in an\\noffense level of 15. The PSR placed Cuevas-Lopez in\\ncriminal history category V based on the two burglaries\\ndiscussed above and other convictions between 2006 and\\n2008. Cuevas-Lopez’s criminal history category and offense\\nlevel resulted in a Guidelines range of 37 to 46 months in\\nprison. Without aggregating Cuevas-Lopez’s two burglary\\nsentences, § 2L1.2(b)(3)(B) would have called for an eight-\\nlevel enhancement based on a 3.5-year sentence, which\\nwould have resulted in an offense level of 13 after the\\ndownward adjustment and a Guidelines range of 30 to\\n37 months. See U.S.S.G. ch. 5, pt. A, Sentencing Table.\\n\\n At the sentencing hearing, the Government agreed with\\nthe PSR’s recommendation and argued, based on the\\n§ 2L1.2(b)(3)(A) Enhancement, that Cuevas-Lopez’s base\\noffense level of eight was “properly enhanced by ten levels\\nbecause the defendant received a seven-year consecutive\\nstate sentence . . . for two second-degree burglary counts\\ncharged in a single indictment.” The district court adopted\\nthe Guidelines calculation in the PSR, to which Cuevas-\\nLopez did not object. The court sentenced Cuevas-Lopez to\\n37 months in prison and three years of supervised release.\\n\\n8 UNITED STATES V. CUEVAS-LOPEZ\\n\\n Cuevas-Lopez timely appealed and now argues that the\\nsingle sentence rule does not apply to § 2L1.2(b)(3)\\nenhancements, and that the district court therefore should not\\nhave aggregated his two prior 3.5-year sentences for\\npurposes of determining his Guidelines range.\\n\\n II.\\n\\n A.\\n\\n Where, as here, a defendant makes an argument on\\nappeal that was not the basis for an objection in the district\\ncourt, we generally review for plain error. See United States\\nv. Gomez, 725 F.3d 1121, 1125 (9th Cir. 2013) (“If the\\ndefendant fails to object, we review for plain error.”); see\\nalso Fed. R. Crim P. 52(b) (“A plain error that affects\\nsubstantial rights may be considered even though it was not\\nbrought to the court’s attention.”). We have held, however,\\nthat “we are not limited to [plain error] review when we are\\npresented with a question that ‘is purely one of law’ and\\nwhere ‘the opposing party will suffer no prejudice as a result\\nof the failure to raise the issue in the trial court.’” United\\nStates v. Saavedra-Velazquez, 578 F.3d 1103, 1106 (9th Cir.\\n2009) (quoting United States v. Echavarria-Escobar,\\n270 F.3d 1265, 1267–68 (9th Cir. 2001)).\\n\\n Cuevas-Lopez urges us to apply de novo review, despite\\nhis failure to object in the district court, because his\\nargument about when the single sentence rule applies\\npresents a pure question of law. The Government responds\\nthat our court’s “pure question of law” exception to plain\\nerror review is inconsistent with the Supreme Court’s\\ninterpretation of Federal Rule of Criminal Procedure 52(b).\\nIn support, the Government cites Puckett v. United States,\\n556 U.S. 129 (2009), in which the Supreme Court held that\\n“[f]ailure to abide by [the] contemporaneous-objection rule\\n\\n UNITED STATES V. CUEVAS-LOPEZ 9\\n\\nordinarily precludes the raising on appeal of [an]\\nunpreserved claim of trial error,” with a “limited exception,”\\nid. at 135, for a “plain error that affects substantial rights,”\\nid. (quoting Fed. R. Crim. P. 52(b)). The Government also\\nrelies on Judge Graber’s concurrence in United States v.\\nZhou, 838 F.3d 1007 (9th Cir. 2016), which opined that\\n“[o]ur ‘pure question of law’ exception contradicts Rule\\n52(b) and the Supreme Court’s case law.” Id. at 1016\\n(Graber, J., concurring). We need not resolve this dispute or\\notherwise decide which standard of review applies here,\\nbecause we would affirm under either de novo or plain error\\nreview. For the reasons explained below, we conclude that\\nthe district court correctly interpreted the Guidelines Manual\\nin determining that the single sentence rule applies to\\n§ 2L1.2(b) enhancements.\\n\\n B.\\n\\n 1.\\n\\n Although the Guidelines are advisory only, a “district\\ncourt must correctly calculate the recommended Guidelines\\nsentence and use that recommendation as the ‘starting point\\nand the initial benchmark.’” United States v. Munoz-\\nCamarena, 631 F.3d 1028, 1030 (9th Cir. 2011) (quoting\\nKimbrough v. United States, 552 U.S. 85, 108 (2007)). The\\ncourt must keep the Guidelines range “in mind throughout\\nthe process,” id. (quoting United States v. Carty, 520 F.3d\\n984, 991 (9th Cir. 2008) (en banc)), and “justify the extent\\nof [any] departure from the Guidelines,” id. “Failure to\\ncalculate the correct Guidelines range constitutes procedural\\nerror.” Peugh v. United States, 569 U.S. 530, 537 (2013).\\n\\n As the Supreme Court has explained, the Guidelines\\nManual “contains text of three varieties,” each of which is\\nwritten by the Sentencing Commission. Stinson v. United\\n\\n10 UNITED STATES V. CUEVAS-LOPEZ\\n\\nStates, 508 U.S. 36, 41 (1993). The “[f]irst is a guideline\\nprovision itself,” which “provide[s] direction as to the\\nappropriate type [and extent] of punishment.” Id. A “second\\nvariety of text in the [Guidelines] Manual is a policy\\nstatement . . . regarding application of the guidelines or other\\naspects of sentencing that would further the purposes of the\\n[Sentencing Reform] Act.” Id. (quotation marks omitted).\\nThe third type of text in the Guidelines Manual is\\ncommentary, which accompanies both guidelines and policy\\nstatements. Id. Commentary may serve three functions: to\\n“interpret a guideline or explain how it is to be applied”; to\\n“suggest circumstances which may warrant departure from\\nthe guidelines”; and to “provide background information,\\nincluding factors considered in promulgating the guideline\\nor reasons underlying promulgation of the guideline.” Id.\\n(alterations and citation omitted). Any modifications or\\namendments to the guidelines provisions themselves (the\\nfirst category of text) must be accompanied by a “statement\\nof the reasons therefor” authored by the Sentencing\\nCommission, and take effect on a date set by the\\nCommission, within certain statutory parameters and subject\\nto any changes made by Congress. See 28 U.S.C. § 994(p).\\n\\n “We interpret the Sentencing Guidelines using the\\nordinary tools of statutory interpretation.” United States v.\\nMartinez, 870 F.3d 1163, 1166 (9th Cir. 2017). Our\\ninterpretation “will most often begin and end with the text\\nand structure of the [g]uidelines” provisions themselves. Id.\\n(quoting United States v. Joey, 845 F.3d 1291, 1297 n.8 (9th\\nCir. 2017)). We also consider “the Commission’s\\ncommentary interpreting or explaining the text” of those\\nguidelines provisions. Id. The commentary “is authoritative\\nunless it violates the Constitution or a federal statute, or is\\ninconsistent with, or a plainly erroneous reading of, that\\nguideline.” Stinson, 508 U.S. at 38; see also United States\\n\\n UNITED STATES V. CUEVAS-LOPEZ 11\\n\\nv. Prien-Pinto, 917 F.3d 1155, 1157–58 (9th Cir. 2019)\\n(following Stinson). Policy statements, likewise, are binding\\n“[t]o the extent that they interpret substantive guidelines and\\ndo not conflict with them or any statutory directives.”\\nUnited States v. Chea, 231 F.3d 531, 536 n.1 (9th Cir. 2000)\\n(citation omitted); see also Stinson, 508 U.S. at 42 (“The\\nprinciple that the Guidelines Manual is binding on federal\\ncourts applies as well to policy statements.”). “We may also\\nlook to [a] provision’s history and purpose, such as by\\nconsulting the Commission’s statements of reason for a\\nparticular amendment.” Martinez, 870 F.3d at 1166 (citation\\nand quotation marks omitted); see also United States v.\\nOrnelas, 825 F.3d 548, 554 (9th Cir. 2016) (noting that we\\n“look to the Commission’s statements of reason for\\nguidance” in interpreting individual guidelines).\\n\\n 2.\\n\\n Chapter Two of the Guidelines Manual governs offense\\nconduct. The offense guideline within Chapter Two\\napplicable to a conviction for attempted illegal reentry is\\n§ 2L1.2. 3 Section 2L1.2(a) provides a base offense level of\\neight. Sections 2L1.2(b)(1), (b)(2), and (b)(3), which\\ncontain the specific offense characteristics, apply\\nenhancements of between two and ten levels based on a\\ndefendant’s prior convictions. Some of these enhancements\\ndepend on the types of convictions sustained and others\\n\\n 3\\n The guideline applicable to a substantive offense generally also\\napplies to inchoate offenses such as attempt. See U.S.S.G. § 2X1.1\\n(2018). The parties do not dispute that § 2L1.2, which is titled\\n“Unlawfully Entering or Remaining in the United States,” applies to\\nCuevas-Lopez’s attempted illegal reentry. See United States v. Rosales-\\nAguilar, 818 F.3d 965, 972 (9th Cir. 2016) (applying § 2L1.2 to an\\nattempted illegal reentry case).\\n\\n12 UNITED STATES V. CUEVAS-LOPEZ\\n\\ndepend on the length of sentences received. In\\n§ 2L1.2(b)(1), § 2L1.2(b)(2)(E), and § 2L1.2(b)(3)(E), the\\nsize of the enhancement turns on the type of offense. 4 The\\nother subsections in § 2L1.2(b)(2) and § 2L1.2(b)(3),\\nincluding the Enhancement, add varying levels of\\nenhancement based on the length of a defendant’s sentence\\nimposed for prior convictions.\\n\\n Subsection (b)(3), which the district court applied to\\nenhance Cuevas-Lopez’s base offense level in this case,\\nprovides in full:\\n\\n (3) (Apply the Greatest) If, at any time after\\n the defendant was ordered deported or\\n ordered removed from the United States for\\n the first time, the defendant engaged in\\n criminal conduct resulting in—\\n\\n (A) a conviction for a felony offense (other\\n than an illegal reentry offense) for which\\n the sentence imposed was five years or\\n more, increase by 10 levels;\\n\\n (B) a conviction for a felony offense (other\\n than an illegal reentry offense) for which\\n\\n\\n 4\\n In § 2L1.2(b)(1), the level of enhancement depends on whether the\\ndefendant, before committing the offense for which he is being\\nsentenced, sustained “a conviction for a felony that is an illegal reentry\\noffense” or sustained “two or more convictions for [improper entry]\\nmisdemeanors under 8 U.S.C. § 1325(a).” U.S.S.G. § 2L1.2(b)(1). Both\\n(b)(2)(E) and (b)(3)(E) apply a two-level enhancement where a\\ndefendant has “three or more convictions for misdemeanors that are\\ncrimes of violence or drug trafficking offenses.”\\nU.S.S.G. § 2L1.2(b)(2)(E), (b)(3)(E).\\n\\n UNITED STATES V. CUEVAS-LOPEZ 13\\n\\n the sentence imposed was two years or\\n more, increase by 8 levels;\\n\\n (C) a conviction for a felony offense (other\\n than an illegal reentry offense) for which\\n the sentence imposed exceeded one year\\n and one month, increase by 6 levels;\\n\\n (D) a conviction for any other felony offense\\n (other than an illegal reentry offense),\\n increase by 4 levels; or\\n\\n (E) three or more convictions for\\n misdemeanors that are crimes of violence\\n or drug trafficking offenses, increase by\\n 2 levels.\\n\\nU.S.S.G. § 2L1.2(b)(3) (underlining added). Subsection\\n2L1.2(b)(2) mirrors subsection (b)(3), but relates to a\\ndefendant’s convictions, if any, sustained before being\\nordered deported or removed. See § 2L1.2(b)(2). 5\\n\\n\\n 5\\n Although we apply the 2016 Guidelines Manual here, we note that\\n§ 2L1.2 was amended in 2018 to “establish[] that the application of the\\n§ 2L1.2(b)(2) enhancement depends on the timing of the underlying\\n‘criminal conduct,’ and not on the timing of the resulting conviction.”\\nU.S.S.G. Supp. to app. C, amend. 809 at 188 (Nov. 1, 2018). The\\namended § 2L1.2(b)(2) applies when “before the defendant was ordered\\ndeported or ordered removed . . . for the first time, the defendant engaged\\nin criminal conduct that, at any time, resulted in” the various convictions\\nand sentences enumerated in that section. U.S.S.G. § 2L1.2(b)(2)\\n(2018). The Sentencing Commission simultaneously updated subsection\\n(b)(3) to mirror the language in subsection (b)(2), see § 2L1.2(b)(3)\\n(2018), though subsection (b)(3), unlike (b)(2), had already (before the\\n2018 amendment) depended upon the timing of the underlying conduct,\\nrather than the underlying conviction, see § 2L1.2(b)(3) (2016).\\n\\n14 UNITED STATES V. CUEVAS-LOPEZ\\n\\n Cuevas-Lopez contends that the words “a conviction” in\\nthe Enhancement (underlined above) preclude a district\\ncourt from aggregating consecutive sentences for purposes\\nof applying the Enhancement. Although this contention\\nfinds some support in the language of § 2L1.2(b)(3), we are\\npersuaded in light of the application notes in the commentary\\nto § 2L1.2, as well as the Sentencing Commission’s\\nstatement of reasons for the 2016 amendment to § 2L1.2,\\nthat the district court properly aggregated Cuevas-Lopez’s\\ntwo 3.5-year sentences here.\\n\\n In aggregating Cuevas-Lopez’s two sentences, the\\ndistrict court relied on the single sentence rule. The single\\nsentence rule falls within Chapter Four of the Guidelines\\nManual, which contains guidelines for categorizing a\\ndefendant’s criminal history. Section 4A1.2(a) of Chapter\\nFour defines “prior sentence” as “any sentence previously\\nimposed upon adjudication of guilt, whether by guilty plea,\\ntrial, or plea of nolo contendere, for conduct not part of the\\ninstant offense.” U.S.S.G. § 4A1.2(a)(1). The guideline\\nthen instructs that whether to treat multiple prior sentences\\nas a single sentence depends on whether they were separated\\nby an intervening arrest, charged in the same instrument, or\\nimposed on the same day. Specifically, § 4A1.2(a)(2) states:\\n\\n If the defendant has multiple prior sentences,\\n determine whether those sentences are\\n counted separately or treated as a single\\n sentence. Prior sentences always are counted\\n separately if the sentences were imposed for\\n offenses that were separated by an\\n intervening arrest (i.e., the defendant is\\n arrested for the first offense prior to\\n committing the second offense). If there is\\n no intervening arrest, prior sentences are\\n\\n UNITED STATES V. CUEVAS-LOPEZ 15\\n\\n counted separately unless (A) the sentences\\n resulted from offenses contained in the same\\n charging instrument; or (B) the sentences\\n were imposed on the same day. Treat any\\n prior sentence covered by (A) or (B) as a\\n single sentence. See also § 4A1.1(e).\\n\\nU.S.S.G. § 4A1.2(a)(2).\\n\\n Section 4A1.2(a)(2) goes on to provide that “if prior\\nsentences are treated as a single sentence,” a court should\\n“use the longest sentence of imprisonment if concurrent\\nsentences were imposed” and “use the aggregate sentence of\\nimprisonment” where “consecutive sentences were\\nimposed.” U.S.S.G. § 4A1.2(a)(2). Here, the district court\\napplied a ten-level enhancement based on Cuevas-Lopez’s\\nconsecutive 3.5-year sentences, rather than an eight-level\\nenhancement, which would have applied had Cuevas-\\nLopez’s two 3.5-year sentences been ordered to run\\nconcurrently (or if they had been ordered to run\\nconsecutively but the single sentence rule’s aggregation\\nprovisions did not apply). Our task, therefore, is to\\ndetermine whether Chapter Four’s single sentence rule\\nproperly applies to § 2L1.2.\\n\\n In support of his argument that the single sentence rule’s\\naggregation provisions should not apply here, Cuevas-Lopez\\nrelies on § 2L1.2’s commentary—specifically Application\\nNote 2. Application Note 2 to § 2L1.2 gives “sentence\\nimposed” “the meaning given the term ‘sentence of\\nimprisonment’ in Application Note 2 and subsection (b) of\\n§ 4A1.2.” U.S.S.G. § 2L1.2 cmt. n.2. As Cuevas-Lopez\\npoints out, this application note to § 2L1.2 does not reference\\nsubsection (a) of § 4A1.2, which defines “prior sentence”\\nand contains the single sentence rule. Cuevas-Lopez argues\\n\\n16 UNITED STATES V. CUEVAS-LOPEZ\\n\\nthat because Application Note 2 specifically cross-\\nreferences a part of § 4A1.2 and yet does not reference the\\npart containing the single sentence rule, the Commission\\nmeant to exclude the single sentence rule from applying to\\nthe § 2L1.2(b) enhancements.\\n\\n In our view, the absence of a mention of subsection (a)\\nof § 4A1.2 in Application Note 2 to § 2L1.2 does not hold\\nthe significance Cuevas-Lopez wishes. 6 Rather, we read\\n§ 4A1.2(b) and Application Note 2 to that section—the two\\nGuidelines Manual components specifically referenced in\\n§ 2L1.2’s commentary for purposes of defining “sentence\\nimposed”—as simply being silent on the question whether\\nseparate sentences that are imposed on the same day and\\nordered to run consecutively should be aggregated for\\npurposes of applying the Enhancement. Other application\\nnotes to the offense guideline in which the Enhancement\\nappears, however, speak more to the issue and evince the\\nSentencing Commission’s intent that the single sentence rule\\napply to § 2L1.2(b).\\n\\n One example is Application Note 3 to the § 2L1.2\\noffense guideline. Note 3 instructs a court that when\\n“applying subsections (b)(1), (b)(2), and (b)(3)” of\\n§ 2L1.2—i.e., the § 2L1.2(b) enhancements—a court should\\n\\n 6\\n The Guidelines Manual does state, as Cuevas-Lopez argues, that\\n“[a]n instruction to use a particular subsection or table from another\\noffense guideline refers only to the particular subsection or table\\nreferenced, and not to the entire offense guideline.” U.S.S.G.\\n§ 1B1.5(b)(2). Section 4A1.2 is not, however, an offense guideline—\\nrather, it provides “Definitions and Instructions for Computing Criminal\\nHistory.” And, in any event, even if the reference to subsection (b) does\\nnot itself amount to an instruction to apply subsection (a), it still leaves\\nus with silence on whether or not to apply subsection (a)’s single\\nsentence rule.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 17\\n\\n“use only those convictions that receive criminal history\\npoints under § 4A1.1(a), (b), or (c).” U.S.S.G. § 2L1.2 cmt.\\nn.3. Criminal history points are added under § 4A1.1(a) and\\n§ 4A1.1(b) based on “prior sentence[s] of imprisonment” of\\nover 13 months (for which three points are added) and\\nbetween 60 days and 13 months (for which two points are\\nadded), respectively. Under § 4A1.1(c), one additional point\\nis added for “each prior sentence” not counted in either (a)\\nor (b). Subsection (e) to § 4A1.1—which is not referenced\\nin § 2L1.2’s Application Note 3—instructs courts to “[a]dd\\n1 point for each prior sentence resulting from a conviction of\\na crime of violence that did not receive any points under (a),\\n(b), or (c) . . . because such sentence was treated as a single\\nsentence.” U.S.S.G. § 4A1.1(e) (emphasis added). The\\nupshot is that certain “prior sentence[s] of imprisonment” in\\n§ 4A1.1(a)–(c) will actually sometimes consist of\\naggregated sentences from multiple separate convictions\\npursuant to the single sentence rule. Because Application\\nNote 3 to § 2L1.2 instructs courts, for purposes of applying\\nthe § 2L1.2(b) enhancements, to use only convictions that\\nreceive criminal history points under subsections (a)–(c) of\\n§ 4A1.1, it suggests that whether and how a conviction\\nshould be counted for purposes of a § 2L1.2(b) enhancement\\ndepends on the operation of the single sentence rule.\\n\\n This understanding is reinforced by a second directive in\\nApplication Note 3 to § 2L1.2—that “for purposes of\\nsubsections (b)(1)(B), (b)(2)(E), and (b)(3)(E)” 7 of that\\nsection, courts “use only those convictions that are counted\\nseparately under [the single sentence rule in] § 4A1.2(a)(2).”\\nU.S.S.G. § 2L1.2 cmt. n.3. This directive further suggests\\n\\n 7\\n These subsections retained a categorical approach after the 2016\\namendment to § 2L1.2, discussed below, and deal with prior\\nmisdemeanor convictions. See infra n.9.\\n\\n18 UNITED STATES V. CUEVAS-LOPEZ\\n\\nthat the Sentencing Commission anticipated that in some\\ninstances two or more convictions could be counted—and\\nreferred to—as one under the single sentence rule, and that a\\ncourt should do so for purposes of other subsections of\\n§ 2L1.2. Were this not the baseline assumption, there would\\nbe no need to explicitly make the single sentence rule\\ninapplicable to these subsections. Application Note 3\\ntherefore weighs in favor of applying the single sentence rule\\nto determine which § 2L1.2(b)(3) enhancement applies.\\n\\n Application Note 4 to § 2L1.2, which deals with cases in\\nwhich a sentence for an illegal reentry offense was imposed\\nat the same time as one for another felony offense, lends\\nsome additional support for applying the single sentence\\nrule. It provides:\\n\\n There may be cases in which the sentences\\n for an illegal reentry offense and another\\n felony offense were imposed at the same time\\n and treated as a single sentence for purposes\\n of calculating the criminal history score\\n under § 4A1.1(a), (b), and (c). In such a case,\\n use the illegal reentry offense in determining\\n the appropriate enhancement under\\n subsection (b)(1), if it independently would\\n have received criminal history points. In\\n addition, use the prior sentence for the other\\n felony offense in determining the appropriate\\n enhancement under subsection (b)(3), if it\\n independently would have received criminal\\n history points.\\n\\nU.S.S.G. § 2L1.2 cmt. n.4 (emphasis added). Like Note 3,\\nthis application note appears to assume that § 4A1.2(a)’s\\n\\n UNITED STATES V. CUEVAS-LOPEZ 19\\n\\nsingle sentence rule would normally apply to § 2L1.2(b)\\nenhancements.\\n\\n 3.\\n\\n The Sentencing Commission’s statement of reasons for\\nthe amendment to § 2L1.2 that created the version applicable\\nhere also supports the district court’s application of the\\nsingle sentence rule. See Ornelas, 825 F.3d at 554 (looking\\n“to the Commission’s statements of reason for guidance” in\\ninterpreting a guideline). Before that 2016 amendment,\\n§ 2L1.2(b) enhancements were based on “the nature of a\\ndefendant’s most serious conviction,” as determined by the\\n“categorical approach to the penal statute underlying the\\nprior conviction.” U.S.S.G. Supp. to app. C, amend. 802\\nat155 (Nov. 1, 2016) (“Amendment 802”). Under the\\ncategorical approach, courts “compare the elements of the\\nstatute forming the basis of the defendant’s conviction with\\nthe elements of the ‘generic’ crime—i.e., the offense as\\ncommonly understood.” Descamps v. United States,\\n570 U.S. 254, 257 (2013). 8 Due to a concern that this\\nmethod of determining a level of enhancement was “overly\\ncomplex,” the Sentencing Commission adopted a simpler,\\nsentence-based model whereby, with a few exceptions, 9\\n\\n 8\\n We note, however, that even before 2016, the § 2L1.2(b)(1)(A) and\\n(B) enhancements required a court to use both a categorical approach\\nand a sentence-imposed approach, as subsection (A) applied where a\\ndefendant sustained “a conviction for a felony offense that [was] . . . a\\ndrug trafficking offense for which the sentence imposed exceeded 13\\nmonths,” and subsection (B) applied where a defendant sustained a\\nfelony drug trafficking conviction “for which the sentence imposed was\\n13 months or less.” U.S.S.G. § 2L1.2(b)(1) (2015).\\n 9\\n The guideline retained the categorical approach for\\n§ 2L1.2(b)(1)(B), (b)(2)(E), and (b)(3)(E), relating to predicate\\n\\n20 UNITED STATES V. CUEVAS-LOPEZ\\n\\n“[t]he level of the sentencing enhancement for a prior\\nconviction generally [would] be determined by the length of\\nthe sentence imposed for the prior offense.” 10 Amendment\\n802 at 155.\\n\\n In a section of the statement of reasons titled\\n“Accounting for Other Prior Convictions,” the Sentencing\\nCommission analogized the sentence-imposed approach in\\nsubsections (b)(2) and (b)(3) “to how Chapter Four of the\\nGuidelines Manual determines a defendant’s criminal\\nhistory score based on his or her prior convictions”—notably\\nreferencing Chapter Four (in which the single sentence rule\\nappears) in its entirety. Amendment 802 at 156. “The\\n[Sentencing] Commission concluded that the length of\\nsentence imposed by a sentencing court is a strong indicator\\nof the court’s assessment of the seriousness of the predicate\\noffense at the time, . . . consistent with how criminal history\\nis generally scored in . . . Chapter Four of the Guidelines\\nManual.” Amendment 802 at 157. The Commission also\\nwrote, in a section titled “Illegal Reentry”: “The definition\\nof ‘sentence imposed’ [in the amended § 2L1.2] is the same\\ndefinition that appears in Chapter Four of the Guidelines\\nManual.” 11 Amendment 802 at 155. The Sentencing\\n\\n\\nmisdemeanor convictions, as well as § 2L1.2(b)(1)(A), relating to\\npredicate illegal reentry felonies.\\n 10\\n Additionally, before 2016, § 2L1.2(b) included only an\\nenhancement for convictions that occurred before deportation. See\\nU.S.S.G. § 2L1.2 (2015). Amendment 802 sought to address a concern\\nthat, because of this, § 2L1.2 did not sufficiently account for all types of\\ncriminal conduct committed by persons charged with illegal reentry.\\nAmendment 802 at 155–56.\\n 11\\n We note that although Chapter Four uses the term “sentence\\nimposed,” it does not specifically define it.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 21\\n\\nCommission did not limit this explanation to any particular\\nprovision of Chapter Four, thereby seeming to incorporate\\nChapter Four’s single sentence rule.\\n\\n Similarly, the Commission wrote, in a section of the\\nstatement of reasons titled “Excluding Stale Convictions,”\\nthat in the context of § 2L1.2’s specific offense\\ncharacteristics, “it is . . . appropriate to employ the criminal\\nhistory rules.” Amendment 802 at 159. Because the single\\nsentence rule is contained in the part of Chapter Four that\\nprescribes the criminal history rules, this statement also\\nsuggests that the Commission envisioned that the rule would\\napply to the § 2L1.2(b) enhancements.\\n\\n Perhaps the strongest evidence of the Commission’s\\nintent appears in another passage of the statement of reasons\\nunder the heading “Application of the ‘Single Sentence\\nRule,’” which discusses the 2016 addition of what became\\nApplication Note 4 to § 2L1.2. See Amendment 802 at 159.\\nAs we have mentioned, Application Note 4 deals with cases\\nin which a defendant is sentenced for an illegal reentry\\noffense at the same time as another federal felony offense.\\nIn such cases, “the illegal reentry offense counts towards\\nsubsection (b)(1), while the other felony offense counts\\ntowards subsection (b)(3).” Amendment 802 at 159. The\\nCommission explained in the statement of reasons that it\\nintended “to make a distinction between illegal reentry\\noffenses and other types of offenses,” and “concluded that it\\nwas appropriate to ensure that such convictions are\\nseparately accounted for under the applicable specific\\noffense characteristics, even if they might otherwise\\nconstitute a ‘single sentence’ under § 4A1.2(a)(2).”\\nAmendment 802 at 159. Thus, both the text of Application\\nNote 4 and the Commission’s stated reason for adding it are\\n\\n22 UNITED STATES V. CUEVAS-LOPEZ\\n\\nbased on an assumption that the single sentence rule would\\napply to § 2L1.2.\\n\\n Applying the single sentence rule to § 2L1.2 also makes\\nsense in light of the Sentencing Commission’s purpose in\\ndrafting Amendment 802. The amended enhancements’\\n“sentence imposed” approach replaced the earlier\\ncategorical method and was intended to capture—in a\\nsimpler way—the seriousness of a defendant’s prior offense.\\nThe single sentence rule requires aggregation only when\\nsentences are ordered to run consecutively. The imposition\\nof consecutive, rather than concurrent, sentences generally\\nreflects a decision by either a sentencing judge or, in some\\ncases when sentences are statutorily required to run\\nconsecutively, by a legislature, that a consecutive term of\\nimprisonment would better reflect the seriousness of a\\ndefendant’s conduct as well as the need for deterrence,\\neducation or treatment of the defendant, and protection of\\nthe public. See 18 U.S.C. § 3584 (stating that “[m]ultiple\\nterms of imprisonment imposed at the same time run\\nconcurrently unless the court orders or the statute mandates\\nthat the terms are to run consecutively,” and directing judges\\nto consider the sentencing factors set forth in 18 U.S.C.\\n§ 3553 “in determining whether the terms imposed are to be\\nordered to run concurrently or consecutively”); 18 U.S.C.\\n§ 3553 (sentencing factors); Setser v. United States, 566 U.S.\\n231, 236 (2012) (“Judges have long been understood to have\\ndiscretion to select whether the sentences they impose will\\nrun concurrently or consecutively with respect to other\\nsentences that they impose.”). Applying the single sentence\\nrule to the § 2L1.2(b)(2) and (b)(3) enhancements therefore\\ncaptures the seriousness of a defendant’s prior convictions\\nand achieves the same goal that the previous categorical\\napproach sought to achieve.\\n\\n UNITED STATES V. CUEVAS-LOPEZ 23\\n\\n 4.\\n\\n Even if there were some ambiguity in how broadly the\\nSentencing Commission intended the single sentence rule to\\napply, as the dissent’s arguments for a contrary\\ninterpretation suggest there may be, the goal of avoiding a\\ncircuit split would lead us to hold that the single sentence\\nrule applies here. “[A]bsent a strong reason to do so, we will\\nnot create a direct conflict with other circuits.” United States\\nv. Chavez-Vernaza, 844 F.2d 1368, 1374 (9th Cir. 1987); see\\nalso Global Linguist Sols., LLC v. Abdelmeged, 913 F.3d\\n921, 923 (9th Cir. 2019) (“[W]e so hold to avoid an\\nunnecessary circuit split.”). The Fifth Circuit recently\\nbecame the first circuit court to answer whether, under the\\n2016 Guidelines Manual, the single sentence rule applies to\\n§ 2L1.2(b). In United States v. Garcia-Sanchez, that court\\nheld that it does. 916 F.3d 522, 526 (5th Cir. 2019). 12 The\\nGarcia-Sanchez decision therefore weighs heavily in favor\\nof affirming the district court’s interpretation here.\\n\\n The Fifth Circuit found support for its conclusion in two\\ncomponents of Amendment 802 discussed above. It first\\nlooked to the section on “Accounting for Other Prior\\nConvictions” in the Sentencing Commission’s statement of\\nreasons for Amendment 802, which described the length of\\na sentence imposed as “a strong indicator of the court’s\\nassessment of the seriousness of the predicate offense,” and\\nas “consistent with the Chapter Four criminal history rules.”\\nGarcia-Sanchez, 916 F.3d at 527 (quoting Amendment 802\\nat 157–58). The Fifth Circuit further relied on the section\\n\\n 12\\n Prior to Garcia-Sanchez, the Fifth Circuit had held in United\\nStates v. Ponce-Flores, 900 F.3d 215 (5th Cir. 2018), that because any\\nerror could not have been obvious, a district court did not plainly err in\\napplying the single sentence rule to a § 2L1.2 enhancement. Id. at 219.\\n\\n24 UNITED STATES V. CUEVAS-LOPEZ\\n\\ntitled “Excluding Stale Convictions,” which explains that, in\\nthe context of § 2L1.2(b)’s specific offense characteristics,\\n“it is . . . appropriate to employ the criminal history rules,”\\nwhich contain the single sentence rule. Id. (quoting\\nAmendment 802 at 159).\\n\\n The court in Garcia-Sanchez also found persuasive the\\nFourth Circuit’s reasoning in United States v. Martinez-\\nVarela, 531 F.3d 298 (4th Cir. 2008). In that case, the\\ndistrict court aggregated sentences for purposes of applying\\na pre-2016 version of § 2L1.2(b)(1), which called for an\\nenhancement if a defendant was “previously convicted of a\\ndrug trafficking offense for which the ‘sentence imposed’\\nwas greater than thirteen months.” Martinez-Varela,\\n531 F.3d at 299. That version of § 2L1.2, like the 2016\\nversion, contained an application note that referenced\\nApplication Note 2 and subsection (b) of § 4A1.2 for\\npurposes of defining “sentence imposed.” Id. at 300. The\\nFourth Circuit affirmed, reasoning that although the\\nprovisions cross-referenced in application notes to § 2L1.2\\ndid not specifically address the issue of aggregating\\nsentences, commentary to § 4A1.1 instructed that “§§ 4A1.1\\nand 4A1.2 must be read together,” providing “strong\\nevidence that these two provisions should be read together\\nin determining [the defendant’s] criminal history points,”\\nand thus that the single sentence rule should apply to the\\n§ 2L1.2(b)(1) enhancement at issue. Id. at 301–02.\\n\\n Although our court had no previous occasion to reach the\\nsentencing question presented here, the holding in Garcia-\\nSanchez and the reasoning in Martinez-Varela are consistent\\nwith our case law. We previously recognized that\\n“[s]ections 2L1.2(b) and 4A1.1-2 serve the same underlying\\nfunction” of “determin[ing] the extent to which prior\\nconvictions affect a defendant’s sentence for the current\\n\\n UNITED STATES V. CUEVAS-LOPEZ 25\\n\\noffense,” and that it therefore “makes sense to treat prior\\nsentences in the same manner when they are used to\\ndetermine the offense level under section 2L1.2 as when\\nthey are used to determine the criminal history category.”\\nUnited States v. Ortiz-Gutierrez, 36 F.3d 80, 82 (9th Cir.\\n1994); 13 see also United States v. Moreno-Cisneros,\\n319 F.3d 456, 458–59 (9th Cir. 2003) (noting that\\n“Guideline § 4A1.2 is analogous to § 2L1.2(b)(1),” and that\\n§ 4A1.2 is “broadly applicable”); United States v. Frias,\\n338 F.3d 206, 210 (3d Cir. 2003) (describing Chapter 4 and\\nan earlier version of § 2L1.2 as “clearly aimed at the same\\nthing, which is varying the punishment based on the criminal\\nrecord,” and concluding that it therefore made sense to refer\\nto Chapter Four’s definitions in applying § 2L1.2 (quotation\\nmarks omitted)); United States v. Galicia-Delgado, 130 F.3d\\n518, 521 (2d Cir. 1997) (“[T]he definitions found in § 4A1.2\\nhave often been borrowed to interpret terms in § 2L1.2.”).\\n\\n We thus now join the Fifth Circuit in holding that the\\nsingle sentence rule applies to the enhancements in\\n§ 2L1.2(b)(2) and (b)(3). 14\\n\\n\\n 13\\n In Ortiz-Gutierrez, we applied the single sentence rule to another\\npre-2016 enhancement under § 2L1.2 that turned on whether a\\ndefendant’s predicate offense was an aggravated felony, which at the\\ntime “include[d] a crime of violence for which the defendant was\\nsentenced to at least five years imprisonment.” 36 F.3d at 82. Section\\n2L1.2 did not cross-reference either § 4A1.1 or § 4A1.2. Id.\\n 14\\n We reject Cuevas-Lopez’s argument that we should apply the rule\\nof lenity to hold in his favor. “[T]he rule of lenity applies to the\\nSentencing Guidelines,” but “only . . . where there is grievous ambiguity\\nor uncertainty in the guidelines.” United States v. D.M., 869 F.3d 1133,\\n1144 (9th Cir. 2017). We find no such “grievous ambiguity or\\nuncertainty” in the question of Guidelines Manual interpretation\\npresented here.\\n\\n26 UNITED STATES V. CUEVAS-LOPEZ\\n\\n III.\\n\\n For the foregoing reasons, we reject Cuevas-Lopez’s\\nargument that the district court was wrong to apply the single\\nsentence rule in calculating his sentence. We therefore\\nAFFIRM.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4079,"cells":{"id":{"kind":"number","value":4649427,"string":"4,649,427"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Randly Begay"},"decision_date":{"kind":"string","value":"2019-08-19"},"docket_number":{"kind":"string","value":"14-10080"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Dorothy W. Nelson, Richard R. Clifton, and N. Randy Smith, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"D.W. NELSON, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 14-10080\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:13-cr-08073-\\n NVW-1\\nRANDLY IRVIN BEGAY, AKA Randly\\nBegay,\\n Defendant-Appellant. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Neil V. Wake, District Judge, Presiding\\n\\n Argued and Submission Deferred October 20, 2015\\n Submitted May 26, 2016\\n Submission Vacated and Deferred March 29, 2017\\n Resubmitted August 19, 2019\\n San Francisco, California\\n\\n Filed August 19, 2019\\n\\n Before: Dorothy W. Nelson, Richard R. Clifton,\\n and N. Randy Smith, Circuit Judges.\\n\\n Opinion by Judge D.W. Nelson;\\n Dissent by Judge N.R. Smith\\n\\n2 UNITED STATES V. BEGAY\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel affirmed a conviction for second-degree\\nmurder (18 U.S.C. §§ 1111 and 1153), reversed a conviction\\nfor discharging a firearm during a “crime of violence”\\n(18 U.S.C. § 924(c)(1)(A)), reversed a mandatory restitution\\norder, and remanded for resentencing.\\n\\n Affirming the second-degree murder conviction, the\\npanel held that the district court did not plainly err in failing\\nto instruct the jury on absence of “heat of passion” as an\\nelement of second-degree murder.\\n\\n The panel held that because second-degree murder can\\nbe committed recklessly, it does not categorically constitute\\na “crime of violence” under the elements clause, 18 U.S.C.\\n§ 924(c)(3)(A). Because in light of United States v. Davis,\\n139 S. Ct. 2319 (2019), second-degree murder likewise\\ncannot constitute a crime of violence under the residual\\nclause, 18 U.S.C. § 924(c)(3)(B), the panel concluded that\\nthe defendant’s § 924(c) conviction cannot stand.\\n\\n The panel held that because second-degree murder is not\\ncategorically a crime of violence, the district court erred in\\nimposing mandatory restitution under 18 U.S.C. § 3663A.\\n\\n Dissenting from Parts II and III, Judge N.R. Smith would\\nhold that second-degree murder is a crime of violence under\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. BEGAY 3\\n\\n§ 924(c)(3)(A). He wrote that the majority goes too far in\\nextending the concept of recklessness to include malice\\naforethought, which, including “depraved heart” murder,\\nrequires conduct well beyond mere recklessness.\\n\\n\\n COUNSEL\\n\\nM. Edith Cunningham (argued), Assistant Federal Public\\nDefender; Jon M. Sands, Federal Public Defender; Office of\\nthe Federal Public Defender, Tucson, Arizona; for\\nDefendant-Appellant.\\n\\nKarla Hotis Delord (argued), Assistant United States\\nAttorney; Krissa M. Lanham, Deputy Appellate Chief;\\nRobert L. Miskell, Appellate Chief; United States Attorney’s\\nOffice, Phoenix, Arizona; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nD.W. NELSON, Circuit Judge:\\n\\n Randly Irvin Begay appeals his jury conviction and\\nsentence for second-degree murder under 18 U.S.C. §§ 1111\\nand 1153 and discharging a firearm during a “crime of\\nviolence” in violation of 18 U.S.C. § 924(c)(1)(A). We\\naffirm Begay’s conviction for second-degree murder. We\\nhold that second-degree murder is not a “crime of violence,”\\nreverse the § 924(c) count of conviction, reverse the\\nmandatory restitution order, and remand for resentencing.\\n\\n Begay’s conviction is AFFIRMED in part, REVERSED\\nin part, and REMANDED for resentencing.\\n\\n4 UNITED STATES V. BEGAY\\n\\n FACTS AND PROCEEDINGS\\n\\n On March 4, 2013, Randly Begay, a Native American,\\nshot Roderick Ben in the head with a handgun, killing Ben.\\nBegay was charged with second-degree murder and\\ndischarging a firearm during a “crime of violence.”\\n\\n After a few hours of drinking and smoking\\nmethamphetamine, Begay, Ben, Begay’s nephew Lionel\\nBegay, and Begay’s girlfriend Meghan Williams were\\nsitting inside a van parked in front of Begay’s parents’\\nresidence in Tuba City, Arizona, located within the Navajo\\nNation Indian Reservation. While sitting inside the van,\\nWilliams and Begay were arguing about Williams’ alleged\\ninfidelity, including Begay’s accusations that she had been\\ncheating on him with Ben. Begay stated he was tired of\\npeople thinking that he was a “bitch” because of Williams\\ncheating on him. During the argument, Begay pulled out a\\ngun and laid it on his leg. Ben saw the gun and told Begay\\nnot to shoot the windows of the van. Begay continued to\\nargue with Williams, stating that he was not going to be a\\n“bitch” anymore and that he was not scared to go to prison\\nfor life. Begay then shot Ben in the head once.\\n\\n At trial, Begay’s theory of defense was that someone else\\nin the car had shot Ben. In closing argument, Begay’s\\nattorney briefly argued that Begay lacked the requisite\\nmalice because he did not act deliberately. During closing\\nargument, the government argued that Begay intentionally\\nshot Ben because he was angry about Williams’ alleged\\ninfidelity with Ben. Begay and the government submitted\\njoint jury instructions, requesting an instruction on second-\\ndegree murder only. Begay’s attorney did not request a\\nvoluntary or involuntary manslaughter instruction, nor did\\nhe object to the instructions as presented.\\n\\n UNITED STATES V. BEGAY 5\\n\\n The jury, instructed to find second-degree murder to be\\na “crime of violence,” convicted Begay of second-degree\\nmurder (Count One) and discharging a firearm during a\\n“crime of violence” under 18 U.S.C. § 924(c)(1)(A)(iii)\\n(Count Two). The district court sentenced Begay to 204\\nmonths’ imprisonment on Count One and a consecutive term\\nof 120 months’ imprisonment on Count Two. The district\\ncourt imposed mandatory restitution under 18 U.S.C.\\n§ 3663A in the amount of $23,622. Begay appeals his\\nconviction on both counts as well as the restitution order.\\n\\n JURISDICTION\\n\\n An “Indian” who commits murder in “Indian country” is\\nsubject to applicable federal criminal laws. 18 U.S.C.\\n§ 1153(a). The location of the shooting here, the Navajo\\nNation Indian Reservation, is “Indian country” for the\\npurposes of § 1153. 18 U.S.C. § 1151. We have jurisdiction\\nunder 28 U.S.C. § 1291.\\n\\n DISCUSSION\\n\\n I\\n\\n “[T]he Due Process Clause requires the prosecution to\\nprove beyond a reasonable doubt the absence of the heat of\\npassion on sudden provocation when the issue is properly\\npresented in a homicide case.” Mullaney v. Wilbur, 421 U.S.\\n684, 704 (1975). Where, as here, a defendant does not object\\nto the jury instructions given by the district court, we review\\nfor plain error. United States v. Matsumaru, 244 F.3d 1092,\\n1102 (9th Cir. 2001).\\n\\n On appeal, Begay argues for the first time that the district\\ncourt erred by failing to instruct the jury that, in order to\\nestablish the element of malice, the government had to prove\\n\\n6 UNITED STATES V. BEGAY\\n\\nbeyond a reasonable doubt that Begay did not act in the heat\\nof passion or upon a sudden quarrel. At trial, Begay’s\\nattorney did not request a voluntary manslaughter\\ninstruction, object to the absence of such an instruction, or\\nintroduce evidence of sudden quarrel or heat of passion at\\ntrial. Begay’s theory of defense was that he did not kill the\\nvictim. Cf. United States v. Anderson, 201 F.3d 1145, 1153\\n(9th Cir. 2000) (“A defendant arguing voluntary\\nmanslaughter ‘attempts to negate the malice element by\\nclaiming, in essence, that she was not acting maliciously\\nbecause some extreme provocation . . . severely impaired her\\ncapacity for self-control in committing the killing.’”\\n(quoting United States v. Quintero, 21 F.3d 885, 890 (9th\\nCir. 1994))).\\n\\n During closing argument, Begay’s attorney briefly\\nargued that the government could not prove malice because\\nBegay did not act deliberately, but did not argue that Begay\\nacted in heat of passion or upon a sudden quarrel. The\\nargument Begay’s attorney made regarding malice centered\\non Begay’s lack of deliberation, arguing “to deliberate and\\nto intend to do something is not to simply be high and pull\\nthe trigger.” Begay’s attorney did not argue that Begay\\nlacked malice because he shot Ben in heat of passion or\\nbecause he was provoked. See Anderson, 201 F.3d at 1152\\n(“A failure to give a jury instruction, even if error, does not\\nseriously affect the fairness and integrity of judicial\\nproceedings if the defense at trial made no argument relevant\\nto the omitted instruction.”).\\n\\n Although the government introduced evidence at trial\\nthat Begay may have been angry at the time of the shooting\\nand discussed Begay’s anger during closing argument, the\\nevidence, arguments, and jury instructions taken as a whole\\ndid not “properly present” the issue of heat of passion or\\n\\n UNITED STATES V. BEGAY 7\\n\\nprovocation to the jury. See United States v. Roston,\\n986 F.2d 1287, 1290 (9th Cir. 1993) (holding that the issue\\nof heat of passion was not properly presented where there\\nwas insufficient evidence of provocation and the defendant’s\\ntheory of the case was that he did not kill the victim); cf.\\nUnited States v. Lesina, 833 F.2d 156, 160 (9th Cir. 1987)\\n(finding that the district court erred by failing to give\\ndefendant’s requested instructions regarding accidental\\nkilling and heat of passion where “that defense [was]\\nraised”). As such, the district court did not plainly err in\\nfailing to instruct the jury on absence of “heat of passion” as\\nan element of second-degree murder. We therefore affirm\\nBegay’s conviction under Count One for second-degree\\nmurder.\\n\\n II\\n\\n “We review de novo whether a criminal conviction is a\\n‘crime of violence’ [under section 924(c)(3)] and whether a\\njury instruction misstated the elements of an offense.”\\nUnited States v. Benally, 843 F.3d 350, 353 (9th Cir. 2016).\\nWhere, as here, a party raises an argument for the first time\\non appeal, we generally review for plain error; however, we\\nare not limited to plain error review “when we are presented\\nwith a question that ‘is purely one of law’ and where ‘the\\nopposing party will suffer no prejudice as a result of the\\nfailure to raise the issue in the trial court.’” United States v.\\nSaavedra-Velazquez, 578 F.3d 1103, 1106 (9th Cir. 2009)\\n(quoting United States v. Echavarria-Escobar, 270 F.3d\\n1265, 1267–68 (9th Cir. 2001)). Here, whether second-\\ndegree murder is a crime of violence is a pure question of\\nlaw, and the government, which has fully briefed the issue,\\nsuffers no prejudice. We therefore apply de novo review. See\\nid. at 1106 (reviewing de novo whether California attempted\\n\\n8 UNITED STATES V. BEGAY\\n\\nrobbery is a crime of violence despite the petitioner’s failure\\nto raise the issue before the district court).\\n\\n Begay was convicted of discharging a firearm during a\\n“crime of violence” under 18 U.S.C. § 924(c). On appeal,\\nBegay argues that second-degree murder does not qualify as\\na “crime of violence.” To determine whether second-degree\\nmurder is a “crime of violence” we apply the “categorical\\napproach” laid out in Taylor v. United States, 495 U.S. 575\\n(1990). Benally, 843 F.3d at 352. Based on the facts of this\\ncase, it may be hard to understand how the shooting of Ben\\nby Begay might not be a “crime of violence.” Under the\\ncategorical approach, however, we do not look to the facts\\nunderlying the conviction, but “compare the elements of the\\nstatute forming the basis of the defendant’s conviction with\\nthe elements of” a “crime of violence.” See Descamps v.\\nUnited States, 570 U.S. 254, 257 (2013). The defendant’s\\ncrime cannot be a categorical “crime of violence” if the\\nconduct proscribed by the statute of conviction is broader\\nthan the conduct encompassed by the statutory definition of\\na “crime of violence.” See id.\\n\\n If the statute of conviction is overbroad, we may ask\\nwhether the statute is divisible, and, if it is, apply the\\n“modified categorical approach” to determine which\\n“statutory phrase” formed the basis of the defendant’s\\nconviction. Descamps, 570 U.S. at 263 (quoting Nijhawan v.\\nHolder, 557 U.S. 29, 41 (2009)). Where, as here, the\\ngovernment has not argued that the statute of conviction is\\ndivisible, we need not conduct a modified categorical\\nanalysis. See United States v. Walton, 881 F.3d 768, 774–75\\n(9th Cir. 2018) (declining to conduct a modified categorical\\nanalysis because the government failed to argue that the\\nstatute of conviction was divisible).\\n\\n UNITED STATES V. BEGAY 9\\n\\n Accordingly, we compare the elements of 18 U.S.C.\\n§ 1111, the second-degree murder statute, to the definition\\nof a “crime of violence” found in § 924(c)(3). A “crime of\\nviolence” is defined as a felony that\\n\\n (A) has as an element the use, attempted use,\\n or threatened use of physical force against the\\n person or property of another, or\\n\\n (B) that by its nature, involves a substantial\\n risk that physical force against the person or\\n property of another may be used in the course\\n of committing the offense.\\n\\n18 U.S.C. § 924(c)(3). Subsection (A) is commonly known\\nas the “elements clause” and subsection (B) is commonly\\nknown as the “residual clause.” Because the Supreme Court\\ndeclared 18 U.S.C. § 924 (c)(3)(B) unconstitutionally vague\\nin United States v. Davis, 139 S.Ct. 2319, 2336 (2019), we\\nneed only determine whether second-degree murder\\nconstitutes a crime of violence under the “elements clause”\\nin subsection (A).\\n\\n A\\n\\n Second-degree murder does not constitute a crime of\\nviolence under the elements clause—18 U.S.C.\\n§ 924(c)(3)(A)—because it can be committed recklessly.\\n\\n “[B]ecause the wording of [18 U.S.C. § 924(c)(3) and\\n18 U.S.C. § 16] is virtually identical, we interpret their plain\\nlanguage in the same manner.” Benally, 843 F.3d at 354\\n(analyzing the required mental state for § 924(c)(3) by\\nlooking to case law interpreting § 16); see also Davis,\\n139 S.Ct. at 2326 (“Like § 924(c)(3), § 16 contains an\\nelements clause and a residual clause. The only difference is\\n\\n10 UNITED STATES V. BEGAY\\n\\nthat § 16’s elements clause, unlike § 924(c)(3)’s elements\\nclause, isn’t limited to felonies . . . .”). 18 U.S.C. § 16\\ndefines the term “crime of violence” as:\\n\\n (a) an offense that has an element the use,\\n attempted use, or threatened use of physical\\n force against the person or property of\\n another, or\\n\\n (b) any other offense that is a felony and that,\\n by its nature, involves a substantial risk that\\n physical force against the person or property\\n of another may be used in the course of\\n committing the offense.\\n\\n(emphasis added). The only substantive difference is that the\\nfelony requirement applies to both subsections of\\n§ 924(c)(3) and only to subsection (b) of § 16, but this\\ndifference “does not affect the operative language used to\\ninterpret the statute’s requisite mental state.” Benally,\\n843 F.3d at 354 n.1; see also Smith v. City of Jackson, 544\\nU.S. 228, 233 (2005) (plurality) (holding that a term is given\\nthe same meaning “when Congress uses the same language\\nin two statutes having similar purposes.”).\\n\\n A “crime of violence” requires intentional conduct. In\\nLeocal v. Ashcroft, the Supreme Court discussed the mens\\nrea necessary to commit a “crime of violence” under\\n18 U.S.C. § 16. 543 U.S. 1 (2004). The Supreme Court\\nreasoned that § 16’s requirement that force be used “against”\\nsomeone or something suggests that “crimes of violence”\\nrequire “a higher degree of intent than negligent or merely\\naccidental conduct.” Leocal, 543 U.S. at 9–11.\\n\\n We have since interpreted Leocal’s reasoning to hold\\nthat “crimes of violence,” as defined in both § 16 and\\n\\n UNITED STATES V. BEGAY 11\\n\\n§ 924(c), require purposeful conduct. Benally, 843 F.3d at\\n353–54 (applying Leocal and Fernandez-Ruiz v. Gonzales,\\n466 F.3d 1121 (9th Cir. 2006) (en banc) to find that\\ninvoluntary manslaughter, which requires only gross\\nnegligence, is not a crime of violence under § 924(c));\\nCovarrubias Teposte v. Holder, 632 F.3d 1049, 1053 (9th\\nCir. 2011) (concluding from Leocal, Fernandez-Ruiz, and\\nUnited States v. Gomez-Leon, 545 F.3d 777 (9th Cir. 2008)\\nthat an intentional use of force was required for a crime of\\nviolence as defined in either subsection of 18 U.S.C. § 16);\\nGomez-Leon, 545 F.3d at 787 (holding that a crime of\\nviolence under 18 U.S.C. § 16 “must require proof of an\\nintentional use of force or a substantial risk that force will be\\nintentionally used during its commission”); Fernandez-Ruiz,\\n466 F.3d at 1130 (holding that crimes that can be committed\\nrecklessly are not “crimes of violence” for the purposes of\\n§ 16 because reckless conduct “is not purposeful”).\\n\\n The government argues that Voisine v. United States, —\\nU.S. — , 136 S.Ct. 2272 (2016) “implicitly overruled\\nFernandez-Ruiz.” As we recently explained, “Voisine\\nexpressly left open the question that Fernandez-Ruiz\\nanswered” and is not “so clearly irreconcilable with\\nFernandez-Ruiz’s reasoning that this three-judge panel is no\\nlonger bound by the precedent of our court.” United States\\nv. Orona, 923 F.3d 1197, 1203 (9th Cir. 2019). In Voisine,\\nthe Supreme Court held that the definition of “misdemeanor\\ncrime of domestic violence” in 18 U.S.C. § 921(a)(33)(A)\\nincludes reckless assaults. Id. at 2278. In doing so, however,\\nthe Court explicitly stated, “our decision today concerning\\n§ 921(a)(33)(A)’s scope does not resolve whether § 16\\nincludes reckless behavior. Courts have sometimes given\\nthese two statutory definitions divergent readings in light of\\ndifferences in their contexts and purposes, and we do not\\nforeclose that possibility with respect to the required mental\\n\\n12 UNITED STATES V. BEGAY\\n\\nstates.” Id. at 2280 n.4. Given that explicit qualification, the\\nunique “context[] and purpose[]” of domestic violence, and\\nthe differences in statutory text between the provision at\\nissue and the one interpreted in Voisine, we continue to apply\\nour case law interpreting 18 U.S.C. § 16 to 18 U.S.C.\\n§ 924(c)(3). 1 See Voisine, 136 S.Ct. at 2290 (limiting the\\nholding to the context of misdemeanor domestic violence);\\nUnited States v. Castleman, 572 U.S. 157, 162–63 (2014)\\n(same); Benally, 843 at 354 (continuing to apply our\\nprecedent interpreting § 16 in interpreting § 924(c)(3)\\n“because the wording of the two statutes is virtually\\nidentical”). It follows from our precedent that a crime of\\nviolence under 18 U.S.C. § 924(c)(3) requires the intentional\\nuse of force.\\n\\n The elements of second-degree murder are that the\\ndefendant (1) “unlawfully kill[ed] a human being” (2) “with\\nmalice aforethought.” 18 U.S.C. § 1111(a); Ninth Circuit\\nModel Criminal Jury Instruction 8.108. “[M]alice\\naforethought covers four different kinds of mental states:\\n(1) intent to kill; (2) intent to do serious bodily injury;\\n(3) depraved heart (i.e., reckless indifference); and (4) intent\\nto commit a felony.” United States v. Pineda-Doval,\\n614 F.3d 1019, 1038 (9th Cir. 2010). As such, second-degree\\nmurder may be committed recklessly—with a depraved\\nheart mental state—and need not be committed willfully or\\nintentionally. See United States v. Houser, 130 F.3d 867,\\n871–72 (9th Cir. 1997) (“Malice aforethought does not\\nrequire an element of willfulness if the existence of that\\n\\n\\n\\n 1\\n See also Gonzales-Ramirez v. Sessions, 727 F. App’x 404, 405 n.7\\n(9th Cir. 2018) (unpublished) (concluding that Voisine “does not affect\\nour § 16(a) case law.”).\\n\\n UNITED STATES V. BEGAY 13\\n\\nmalice is inferred from the fact that defendant acted\\nrecklessly with extreme disregard for human life.”).\\n\\n It is of no consequence that the recklessness required for\\nsecond-degree murder must be “extreme” and goes beyond\\nordinary recklessness. In Gomez-Leon, we made clear that,\\nin order to constitute a crime of violence, “the underlying\\noffense must require proof of an intentional use of force or a\\nsubstantial risk that force will be intentionally used during\\nits commission.” 545 F.3d at 787. “[O]ur precedent seems\\nsquarely to place crimes motivated by intent on a pedestal,\\nwhile pushing off other very dangerous and violent conduct\\nthat, because not intentional, does not qualify as a ‘crime of\\nviolence.’” Covarrubias, 632 F.3d at 1053. Reckless\\nconduct, no matter how extreme, is not intentional.\\n\\n Second-degree murder also does not involve a\\n“substantial risk that force will be intentionally used during\\nits commission.” Gomez-Leon, 545 F.3d at 787. In\\nCovarrubias, we held that a California offense prohibiting\\nthe malicious and willful discharge of a firearm at an\\ninhabited dwelling was not a “crime of violence” because it\\ncould be committed recklessly, not just intentionally.\\nCovarrubias, 632 F.3d at 1053–55. Although we conducted\\nour analysis under § 16(b) because the BIA rested its\\ndecision on subsection (b), id. at 1052, the analysis regarding\\nintent bears upon either subsection of § 16, and by analogy,\\n18 U.S.C. § 924(c)(3). See, e.g., Gomez-Leon, 545 F.3d\\nat 787 (requiring intentional use of force for a crime of\\nviolence under either subsection of § 16); Benally, 843 F.3d\\nat 354. In contrast to crimes like burglary that can be\\ncommitted only intentionally, “with a crime committed\\nrecklessly, it is unlikely that the reckless actor will, in\\nresponse to external events, form an intent to use force in\\nfurtherance of his crime.” Covarrubias, 632 F.3d at 1055.\\n\\n14 UNITED STATES V. BEGAY\\n\\n“Classic examples of second-degree murder include\\nshooting a gun into a room that the defendant knows to be\\noccupied, a game of Russian roulette, and driving a car at\\nvery high speeds along a crowded main street . . . .” Pineda-\\nDoval, 614 F.3d at 1039. For purposes of our analysis, these\\nexamples are substantively indistinguishable from the\\noffense—“Shooting at an Inhabited Dwelling or Vehicle”—\\nthat we held was not categorically a crime of violence in\\nCovarrubias. This risk that a crime could escalate to the use\\nof intentional force is no more substantial for a defendant\\nwho recklessly kills than it is for a defendant who recklessly\\nshoots at a house.\\n\\n The cases the government cites do not squarely address\\nwhether second-degree murder is a crime of violence.\\nInstead, in those cases, we found that other challenges to\\n§ 924(c) convictions lacked merit. See, e.g., United States v.\\nPercy, 250 F.3d 720 (9th Cir. 2001) (addressing the Sixth\\nAmendment right to counsel and prosecutorial misconduct);\\nHouser, 130 F.3d 867 (discussing the role of the jury and the\\nmens rea required for second-degree murder). More\\nrecently, in United States v. JJ, we reviewed a district court's\\ndecision to try a juvenile as an adult under the Federal\\nJuvenile Delinquency Act, 18 U.S.C. § 5031 et seq. United\\nStates v. JJ, 704 F.3d 1219 (9th Cir. 2013). In JJ, the\\nquestion of whether “second-degree murder, if committed by\\nan adult, would be a felony crime of violence” was not in\\ndispute. Id. at 1222. Far from “holding” that second-degree\\nmurder is a “crime of violence” for purposes of § 924(c), JJ\\nfocused on when it might be in the interest of justice to try a\\njuvenile as an adult. Id. Because second-degree murder can\\n\\n UNITED STATES V. BEGAY 15\\n\\nbe committed recklessly, rather than intentionally, it does not\\ncategorically constitute a crime of violence. 2\\n\\n B\\n\\n Second-degree murder is not categorically a crime of\\nviolence under the elements clause, 18 U.S.C.\\n§ 924(c)(3)(A). And, pursuant to Davis, second-degree\\nmurder cannot constitute a crime of violence under the\\nresidual clause, section 924(c)(3)(B), as the residual clause\\nis unconstitutionally vague. Begay’s § 924(c) conviction for\\ndischarging a firearm during and in relation to a crime of\\nviolence therefore cannot stand under either the elements\\nclause or residual clause of § 924(c)(3). We reverse Count\\nTwo of his conviction.\\n\\n III\\n\\n We review the legality of a restitution order de novo and\\nthe factual findings supporting the order for clear error.\\nUnited States v. Luis, 765 F.3d 1061, 1065 (9th Cir. 2014)\\n(quoting United States v. Brock-Davis, 504 F.3d 991, 996\\n(9th Cir. 2007)). Where, as here, a defendant raises an\\nobjection to a restitution order for the first time on appeal,\\nwe review for plain error. United States v. Van Alystne, 584\\nF.3d 803, 819 (9th Cir. 2009).\\n\\n 18 U.S.C. § 3663A directs a district court to impose\\nmandatory restitution where a defendant is convicted of a\\ncrime of violence, as defined in 18 U.S.C. § 16. 18 U.S.C.\\n\\n 2\\n In holding that second-degree murder does not qualify as a\\ncategorical crime of violence because it can be committed recklessly, we\\ndo not reach Begay's argument that second-degree murder does not\\nqualify as a categorical crime of violence for the additional reason that it\\ncan be committed without the use of violent force.\\n\\n16 UNITED STATES V. BEGAY\\n\\n§ 3663A(c)(1)(A)(i). We conclude that the district court\\nplainly erred in imposing mandatory restitution under\\n18 U.S.C. § 3663A because second-degree murder is not\\ncategorically a crime of violence, and vacate the mandatory\\nrestitution order. In light of this disposition, we need not\\naddress Begay’s claim that the district court failed to make\\nthe necessary findings for the restitution order.\\n\\n CONCLUSION\\n\\n We AFFIRM Count One of Begay’s conviction for\\nsecond-degree murder. We REVERSE Count Two of\\nBegay’s conviction for discharging a firearm during a “crime\\nof violence” under 18 U.S.C. § 924(c)(1)(A), VACATE the\\nmandatory restitution order imposed under 18 U.S.C.\\n§ 3663A, and REMAND for resentencing.\\n\\n AFFIRMED IN PART, REVERSED IN PART,\\nVACATED IN PART, AND REMANDED.\\n\\n UNITED STATES V. BEGAY 17\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426680/\", \"author_raw\": \"D.W. NELSON, Circuit Judge:\"}, {\"author\": \"SMITH, N.R., Circuit Judge, dissenting\", \"type\": \"dissent\", \"text\": \"SMITH, N.R., Circuit Judge, dissenting from Parts II & III\\nof the Majority’s opinion: 1\\n\\n MURDER 2 in the second-degree is NOT a crime of\\nviolence??? Yet attempted first-degree murder, 3 battery, 4\\nassault, 5 exhibiting a firearm, 6 criminal threats 7 (even\\nattempted criminal threats 8), and mailing threatening\\n\\n 1\\n I concur in Part I of the Majority’s opinion.\\n\\n 2\\n Murderers are “deserving of the most serious forms of\\npunishment.” Graham v. Florida, 560 U.S. 48, 69 (2010), as modified\\n(July 6, 2010). The Supreme Court recognized “a line between homicide\\nand other serious violent offenses against the individual,” because\\n“[s]erious nonhomicide crimes may be devastating in their harm but in\\nterms of moral depravity and of the injury to the person and to the public,\\nthey cannot be compared to murder in their severity and irrevocability.”\\nId. (citation, quotation marks, and alteration omitted).\\n 3\\n United States v. Studhorse, 883 F.3d 1198, 1206 (9th Cir.), cert.\\ndenied, 139 S. Ct. 127 (2018) (Wash. Rev. Code § 9A.32.030(1)).\\n\\n 4\\n United States v. Perez, No. 17-10216, 2019 WL 3022334, *5 (9th\\nCir. July 11, 2019) (Cal. Penal Code § 243(d)).\\n\\n 5\\n United States v. Vasquez-Gonzalez, 901 F.3d 1060, 1068 (9th Cir.\\n2018) (Cal. Penal Code § 245(a)(1)); United States v. Cabrera-Perez,\\n751 F.3d 1000, 1007 (9th Cir. 2014) (Ariz. Rev. Stat. §§ 13-1203, 13-\\n1204).\\n\\n 6\\n Bolanos v. Holder, 734 F.3d 875, 878 (9th Cir. 2013) (Cal. Penal\\nCode § 417.3).\\n\\n 7\\n Rosales-Rosales v. Ashcroft, 347 F.3d 714, 717 (9th Cir. 2003)\\n(Cal. Penal Code § 422).\\n 8\\n Arellano Hernandez v. Lynch, 831 F.3d 1127, 1132 (9th Cir. 2016)\\n(Cal. Penal Code §§ 422, 664).\\n\\n18 UNITED STATES V. BEGAY\\n\\ncommunications 9 are crimes of violence. How can this be?\\n“I feel like I am taking crazy pills.” 10\\n\\n This unbelievable result (arising because of the\\nMajority’s opinion) stems from the Majority’s\\nmisapplication of the categorical approach to conclude that\\nsecond-degree murder cannot serve as a predicate crime of\\nviolence under 18 U.S.C. § 924(c)(3)(A). Never mind that\\nthe Majority’s result is contrary to Supreme Court precedent\\nand our precedent 11 and further defies reality and logic. See\\nUnited States v. Hill, 890 F.3d 51, 56 (2d Cir. 2018), cert.\\ndenied, 139 S. Ct. 844, 202 L. Ed. 2d 612 (2019) (noting that\\n“the categorical approach must be grounded in reality, logic,\\nand precedent, not flights of fancy”).\\n\\n “Malice aforethought” 12 is a required element of first-\\nand second-degree murder. 18 U.S.C. § 1111(a). Rather\\nthan use “malice aforethought” as a clear (and easy)\\ndistinction for a crime of violence, the Majority plucks the\\n 9\\n United States v. De La Fuente, 353 F.3d 766, 770–71 (9th Cir.\\n2003) (18 U.S.C. § 876(c)).\\n\\n 10\\n Ben Stiller (Director), Zoolander [Film], United States:\\nParamount Pictures (2001).\\n\\n 11\\n Our sister circuit has also recognized that it would be illogical to\\nfind that “the most morally repugnant crime—murder—would not be a\\ncrime of violence while at the same time permitting many less-serious\\ncrimes to be so classified.” In re Irby, 858 F.3d 231, 237 (4th Cir. 2017)\\n(citation and quotation marks omitted) (analyzing crime of violence\\nunder the force provision).\\n\\n 12\\n “[M]alice aforethought covers four different kinds of mental\\nstates: (1) intent to kill; (2) intent to do serious bodily injury;\\n(3) depraved heart (i.e., reckless indifference); and (4) intent to commit\\na felony.” United States v. Pineda-Doval, 614 F.3d 1019, 1038 (9th Cir.\\n2010).\\n\\n UNITED STATES V. BEGAY 19\\n\\nterm “recklessness” out of the defined term of “depraved\\nheart murder” and rejects any notion that there are varying\\ndegrees of recklessness in order to hold that second-degree\\nmurder is not a crime of violence. As a result, the term\\n“crime of violence” now has no real or logical meaning. I\\ndo not understand how this can be.\\n\\n I.\\n\\n Prior to the Comprehensive Crime Control Act of 1984\\n(CCCA), when Congress enacted the term “crime of\\nviolence,” it outlined specific crimes that met the definition.\\nSee Narcotic Addict Rehabilitation Act of 1966, Pub. L. No.\\n89-793, tit. I, § 2901, 80 Stat. 1438 (1966) (defining “crime\\nof violence” as “voluntary manslaughter, murder, rape,\\nmayhem, kidnaping, robbery, burglary or housebreaking in\\nthe nighttime, extortion accompanied by threats of violence,\\nassault with a dangerous weapon or assault with intent to\\ncommit any offense punishable by imprisonment for more\\nthan one year, arson punishable as a felony, or an attempt or\\nconspiracy to commit any of the foregoing offenses”),\\nrepealed by Comprehensive Crime Control Act of 1984,\\nPub. L. No. 98-473, tit. II, § 218(a)(6), 98 Stat. 1837 (1984);\\nFederal Firearms Act of 1938, Pub. L. No. 75-785, § 6, 52\\nStat. 1250 (1938) (defining “crime of violence” as “murder,\\nmanslaughter, rape, mayhem, kidnaping, burglary,\\nhousebreaking; assault with intent to kill, commit rape, or\\nrob; assault with a dangerous weapon, or assault with intent\\nto commit any offense punishable by imprisonment for more\\nthan one year”), repealed by Gun Control Act of 1968, Pub.\\nL. No. 90-618, § 203, 82 Stat. 1214 (1968). Upon enacting\\nthe CCCA, Congress changed the definition of “crime of\\nviolence” to include “an offense that is a felony” and that\\n“has as an element the use, attempted use, or threatened use\\nof physical force against the person or property of another.”\\n\\n20 UNITED STATES V. BEGAY\\n\\n18 U.S.C. §§ 16(a), 924(c)(3)(A). However, Congress has\\nindicated that murder is crime of violence on several\\noccasions since the CCCA enactment. See 132 Cong. Rec.\\n7698 (April 16, 1986) (explaining the need to expand the\\nArmed Career Criminal Act beyond robbery and burglary to\\ninclude murder, rape, and heroin smuggling); H.R. Rep. 105-\\n86, *76–77 (1997) (noting that “[t]he age at which a juvenile\\nmay be proceeded against as an adult drops to 13 if the crime\\nof violence alleged is assault, murder [citing 18 U.S.C.\\n§ 1111(a)], attempted murder, or if the juvenile possessed a\\nfirearm during the offense, robbery, bank robbery,\\naggravated sexual abuse.” (footnotes omitted)).\\n\\n Because Congress no longer provided a specific list of\\noffenses that were “crimes of violence,” courts were left to\\napply the categorical approach to determine which offenses\\nmet the new definition. Thus, since the enactment of the\\nCCCA, the interpretation of the term “crime of violence” has\\nevolved in the Supreme Court and the Ninth Circuit.\\nNotably, as demonstrated below, the Ninth Circuit’s\\ninterpretation of included crimes has narrowed over time.\\n\\n Initially, we rejected an argument that aggravated assault\\nwith a recklessness mens rea was a not crime of violence.\\nSee United States v. Ceron-Sanchez, 222 F.3d 1169, 1172–\\n73 (9th Cir. 2000), overruled by Fernandez-Ruiz v.\\nGonzales, 466 F.3d 1121 (9th Cir. 2006) (en banc). One year\\nlater, we declined to extend the holding to cover “negligent\\nconduct,” instead concluding that a “crime of violence”\\ncould not be committed if a person acted “negligently—\\nrather than intentionally or recklessly.” United States v.\\nTrinidad-Aquino, 259 F.3d 1140, 1144–45 (9th Cir. 2001).\\nThereafter, in Leocal v. Ashcroft, the Supreme Court defined\\na crime of violence when it interpreted the phrase “use of\\nphysical force against another” to mean “a higher degree of\\n\\n UNITED STATES V. BEGAY 21\\n\\nintent than negligent or merely accidental conduct.”\\n543 U.S. 1, 9 (2004) (alterations omitted). 13 Thus, Leocal\\neffectively affirmed our decision in Trinidad-Aquino. In\\n2005, we determined that a crime of violence does not\\ninclude acts that are “gross[ly] negligen[t] (but without\\nmalice aforethought).” See Lara-Cazares v. Gonzales, 408\\nF.3d 1217, 1221 (9th Cir. 2005). The next year, we rejected\\nour prior case law and followed our sister circuits to\\nconclude that acts “involving the reckless use of force” were\\nalso not crimes of violence. See Fernandez-Ruiz, 466 F.3d\\nat 1129. 14\\n\\n Following Leocal and Fernandez-Ruiz, we next\\noverruled our prior precedent, which had made involuntary\\nmanslaughter, 15 under 18 U.S.C. § 1112, a crime of\\nviolence. United States v. Benally, 843 F.3d 350, 354 (9th\\nCir. 2016) (holding that United States v. Springfield, 829\\nF.2d 860 (9th Cir. 1987) was “clearly irreconcilable with the\\nreasoning and the results of Leocal and Fernandez-Ruiz and\\n\\n 13\\n Leocal did not address “whether a state or federal offense that\\nrequires proof of the reckless use of force against a person or property of\\nanother qualifies as a crime of violence under 18 U.S.C. § 16.” Leocal,\\n543 U.S. at 13.\\n\\n 14\\n In United States v. Orona, we concluded that Voisine v. United\\nStates, 136 S. Ct. 2272, 2279–82 (2016) did not overrule Fernandez-\\nRuiz. 923 F.3d 1197, 1202–03 (9th Cir. 2019).\\n\\n 15\\n Manslaughter can be a “lesser included offense” of second-degree\\nmurder. See United States v. Anderson, 201 F.3d 1145, 1148 (9th Cir.\\n2000). An involuntary manslaughter instruction is proper when the\\ndefense theory is that the killing was accidental or self-defense. Id.\\nat 1151–52. A voluntary manslaughter instruction is proper when\\ndefense “introduces evidence of sudden quarrel or heat of passion.”\\nUnited States v. Quintero, 21 F.3d 885, 890 (9th Cir. 1994).\\n\\n22 UNITED STATES V. BEGAY\\n\\nis no longer good law”). In Benally, we recognized that “[a]\\nconviction for involuntary manslaughter requires, at a\\nminimum, a mental state of gross negligence, defined as a\\nwanton or reckless disregard for human life.” 16 Id. at 353\\n(internal quotation marks and citation omitted).\\n\\n Against this historical backdrop of interpreting the term\\n“crime of violence,” the Majority now concludes “[s]econd-\\ndegree murder does not constitute a crime of violence under\\nthe elements clause—18 U.S.C. § 924(c)(3)(A)—because it\\ncan be committed recklessly.” Maj. Op. 9 However, the\\n“recklessness” (the Majority now cites) arises only when one\\ncan be found to possess malice aforethought because of a\\n“depraved-heart murder,” wherein “a defendant’s conduct\\nmust create a ‘very high degree of risk’ of injury to other\\npersons, he must be aware of that risk, and he cannot have a\\njustifiable reason for taking that risk.” Pineda-Doval,\\n614 F.3d at 1038 (footnote omitted). In making this\\ndecision, the Majority ignores the differences between\\n“gross negligence” and “malice aforethought,” concluding\\ninstead that “[i]t is of no consequence that the recklessness\\nrequired for second-degree murder must be ‘extreme’ and\\ngoes beyond ordinary recklessness.” 17 Maj. Op. 13. Thus,\\n 16\\n Although we have not yet addressed whether voluntary\\nmanslaughter under 18 U.S.C. § 1112 is a crime of violence, we have\\ndetermined that state voluntary manslaughter crimes are not crimes of\\nviolence, because they can be committed recklessly. See Quijada-\\nAguilar v. Lynch, 799 F.3d 1303, 1306–07 (9th Cir. 2015) (holding\\nCalifornia voluntary manslaughter is not a crime of violence, because a\\nconviction requires “merely reckless conduct”).\\n\\n 17\\n The Majority’s notion that the term “recklessness” effectively has\\none meaning conjures up visions of Alice in Wonderland: “When I use\\na word, . . . it means just what I choose it to mean—neither more nor\\nless.” Lewis Carroll, Through the Looking-Glass and What Alice Found\\nThere.\\n\\n UNITED STATES V. BEGAY 23\\n\\nits decision in effect equates the term “malice aforethought”\\nwith “gross negligence” or “accidental conduct.” That\\ncannot be correct. Instead, there are significant differences\\nbetween the mens rea necessary to support a conviction\\nusing ordinary recklessness and the recklessness necessary\\nto establish malice aforethought. That difference must be\\ntaken into account when defining a “crime of violence.”\\n\\n Post-Leocal, we have referenced terms such as “mere”\\nor “simple” recklessness in the context of applying the\\ncategorical analysis and have held that the ordinary form of\\nrecklessness does not fall within the definition of a “crime of\\nviolence.” See, e.g., Covarrubias Teposte v. Holder,\\n632 F.3d 1049, 1054 (9th Cir. 2011) (holding the California\\noffense of shooting at an inhabited dwelling 18 or vehicle is\\nnot a crime of violence because it “requires a merely reckless\\nmens rea”); Pineda-Doval, 614 F.3d at 1040 (noting that\\n“second-degree murder required a finding of extreme\\nrecklessness evincing disregard for human life, not simple\\nrecklessness” (emphasis added)); Fernandez-Ruiz, 466 F.3d\\nat 1124 (“We ordered rehearing en banc to resolve an inter-\\nand intra-circuit conflict as to whether . . . crimes involving\\nthe merely reckless use of force can be crimes of violence.”\\n(emphasis added)). These cases outline that the ordinary or\\n“simple” form of recklessness is consistent with “gross\\nnegligence.” 19\\n\\n\\n\\n 18\\n For purposes of the California Penal Code section 246, an\\n“inhabited dwelling” need not be occupied. Covarrubias Teposte,\\n632 F.3d at 1054.\\n 19\\n “‘Gross negligence’ is a relative term. It is doubtless to be\\nunderstood as meaning a greater want of care than is implied by the term\\n‘ordinary negligence;’ but, after all, it means the absence of the care that\\n\\n24 UNITED STATES V. BEGAY\\n\\n However, we have never directly addressed “deliberate”\\nor “extreme” recklessness in the same context, 20 which\\nclearly provides a higher standard than “mere recklessness.”\\nSee Pineda-Doval, 614 F.3d at 1039 (recognizing that\\n“merely reckless driving cannot provide the basis for a\\nsecond-degree murder conviction,” because “‘something\\nmore’ was required to establish the malice aforethought\\nnecessary to prove second-degree murder” (quoting United\\nStates v. Hernandez-Rodriguez, 975 F.2d 622, 627 (9th Cir.\\n1992)).\\n\\n This distinction between mere or simple recklessness\\nand “extreme” recklessness is critical. The depraved-heart\\n(i.e., reckless indifference) necessary for a conviction of\\nsecond-degree murder requires more than “mere\\nrecklessness”; it requires malice aforethought. See\\n18 U.S.C. § 1111(a) (“Murder is the unlawful killing of a\\nhuman being with malice aforethought.”); see also United\\nStates v. Lesina, 833 F.2d 156, 159 (9th Cir. 1987) (“The\\ndifference between that recklessness which displays\\ndepravity and such extreme and wanton disregard for human\\nlife as to constitute ‘malice’ and that recklessness that\\namounts only to [gross negligence] lies in the quality of\\nawareness of the risk.” (citation omitted)). “Malice\\naforethought does not mean simply hatred or ill will, but also\\nembraces the state of mind with which one intentionally\\ncommits a wrongful act without legal justification or excuse.\\nIt may be inferred from circumstances which show a wanton\\n\\nwas necessary under the circumstances.” Milwaukee & St. P.R. Co. v.\\nArms, 91 U.S. 489, 495 (1875).\\n\\n 20\\n In United States v. Cabrera-Perez, we seemed to recognize a\\ndifference in “the reckless mens rea” when we suggested that a crime of\\nviolence required a “mens rea of at least extreme recklessness.” 751 F.3d\\nat 1005.\\n\\n UNITED STATES V. BEGAY 25\\n\\nand depraved spirit, a mind bent on evil mischief without\\nregard to its consequences.” United States v. Celestine,\\n510 F.2d 457, 459 (9th Cir. 1975) (emphasis added) (internal\\nquotation marks and citation omitted); see also Lesina,\\n833 F.2d at 159 (noting that “disregard for human life\\nbecomes more callous, wanton or reckless, and more\\nprobative of malice aforethought, as it approaches a mental\\nstate comparable to deliberation and intent”). Notably, we\\nhave referred to extreme “reckless indifference” as\\n“depraved-heart murder.” Pineda-Doval, 614 F.3d at 1038.\\nAs even the Majority acknowledges, “[c]lassic examples of\\nsecond-degree murder include shooting a gun into a room\\nthat the defendant knows to be occupied, playing a game of\\nRussian roulette, and driving a car at very high speeds along\\na crowded main street.” Id. at 1039. Yet, none of these\\nexamples demonstrate a recklessness equivalent to gross\\nnegligence. See Fernandez-Ruiz, 466 F.3d at 1130.\\n\\n Although the specific facts of a case do not play into a\\ncategorical analysis, they do shed light on what conduct\\namounts to the recklessness of depraved heart murder. For\\nexample, we held that certain types of actions amounted to\\n“extreme disregard for life” in United States v. Houser,\\n130 F.3d 867 (9th Cir. 1997), United States v. Boise, 916\\nF.2d 497 (9th Cir. 1990), and Celestine, 510 F.2d 457. In\\nHouser, the defendant “took a handgun from his truck,\\npumped a cartridge into the chamber of the gun, [and] held\\nthe gun behind his back.” 130 F.3d at 869. He then\\napproached his girlfriend and, “[a]fter four or five seconds[,]\\n. . . brought his right arm to the left side of [his girlfriend’s]\\nneck and shot her.” 21 Id. We concluded that the district\\n\\n\\n 21\\n “Houser’s defense was that the shooting was wholly accidental;\\nthat he did not intend to use the gun but that [his girlfriend] grabbed it\\n\\n26 UNITED STATES V. BEGAY\\n\\ncourt’s “extreme disregard” instruction was proper. 22 Id.\\nat 871. In Boise, we held that “the jury could rationally\\nconclude beyond a reasonable doubt that Boise displayed ‘a\\nwanton and depraved spirit, a mind bent on evil mischief\\nwithout regard to its consequences,’” when he killed a small\\nchild through “two blunt force blows to the head.” 916 F.2d\\nat 500 (quoting Celestine, 510 F.2d at 459). Similarly, in\\nCelestine, we held that a jury was permitted to conclude\\n“that the homicide was accompanied by a callous and\\nwanton disregard of human life,” when the defendant killed\\na woman by beating her with his fists and by repeatedly\\nkicking a stick into her vagina. 510 F.2d at 458–59.\\n\\n These cases do not demonstrate the type of conduct\\nindicative of mere or simple recklessness that is equivalent\\nto gross negligence. See Leocal, 543 U.S. at 9; see also\\nFernandez-Ruiz, 466 F.3d at 1130 (“[T]he reckless use of\\nforce is ‘accidental’ and crimes of recklessness cannot be\\ncrimes of violence.”). Thus, the extreme recklessness\\nnecessary to equate to malice aforethought is not and cannot\\nbe “accidental.” See Fernandez-Ruiz, 466 F.3d at 1130; see\\nalso Leocal, 543 U.S. at 9. To the contrary, killing\\n“recklessly with extreme disregard for human life” most\\nnaturally suggests a higher degree of depravity or knowing\\ndisregard. If the underlying use of physical force were\\n“accidental,” the charge (or conviction) would be\\ninvoluntary manslaughter. See Ninth Cir. Crim. Jury Inst.\\n\\nand tried to wrest it from him, causing an accidental discharge.” Houser,\\n130 F.3d at 870.\\n\\n 22\\n In Houser, we concluded that the district court’s use of “the\\nexpression ‘extreme disregard for human life’ instead of ‘a wanton and\\ndepraved spirit, a mind bent on evil mischief without regard to its\\nconsequences,’” was not error because they were “functionally\\nequivalent.” 130 F.3d at 871 n.3.\\n\\n UNITED STATES V. BEGAY 27\\n\\n8.110 (requiring a jury to find that “the defendant acted with\\ngross negligence, defined as wanton or reckless disregard for\\nhuman life”). Thus, murder, which requires malice\\naforethought (with its higher degree of intent), must\\nnecessarily be a crime of violence.\\n\\n As Judge Wardlaw aptly recognized in her dissent in\\nFernandez-Ruiz, “[r]ecklessness is a distinct mens rea,\\nwhich lies closer to intentionality than to negligence.”\\n466 F.3d at 1141–42 (Wardlaw, J., dissenting). The\\nSupreme Court has also recognized that “the reckless\\ndisregard for human life . . . represents a highly culpable\\nmental state.” Tison v. Arizona, 481 U.S. 137, 157–58\\n(1987). The Majority’s suggestion that second-degree\\nmurder can be committed without any intentional conduct is\\nsimply not correct. Although a defendant may not have to\\nintended to kill (“express malice”), he or she nevertheless\\nkilled a human being with “an intentional act that had a high\\nprobability of resulting in death (implied malice).” See\\nPollard v. White, 119 F.3d 1430, 1434 (9th Cir. 1997)\\n(noting at a minimum “implied malice aforethought” was\\ndemonstrated by the defendant’s statement “that he was\\ngoing to ‘blow [the victim] away’” because it showed “an\\nintent to shoot when he pulled the trigger” (alteration in\\noriginal)). Again, these examples denote the difference\\nbetween gross negligence (mere recklessness) and malice\\naforethought (depraved heart murder). 23\\n\\n\\n 23\\n The difference between the required mens rea necessary for a\\nconviction of manslaughter or for murder also demonstrates this\\nimportant distinction. Under the federal definition, manslaughter is\\ndefined as an “unlawful killing of a human being without malice.”\\n18 U.S.C. § 1112)(a) (emphasis added). Killing without malice suggests\\nthat the mental state required to commit manslaughter is lesser than the\\nmental state required to commit some degree of murder. See United\\n\\n28 UNITED STATES V. BEGAY\\n\\n II.\\n\\n “[T]o construe statutes so as to avoid results glaringly\\nabsurd, has long been a judicial function.” Armstrong Paint\\n& Varnish Works v. Nu-Enamel Corp., 305 U.S. 315, 333\\n(1938). Yet, the Majority does not seem to even contemplate\\nthe far-reaching results of its decision. This Majority\\nopinion produces precisely those “glaringly absurd” results,\\nwhich we are instructed to avoid. See Griffin v. Oceanic\\nContractors, Inc., 458 U.S. 564, 575 (1982) (noting “that\\ninterpretations of a statute which would produce absurd\\nresults are to be avoided if alternative interpretations\\nconsistent with the legislative purpose are available”).\\n\\n We have already concluded that as exhibiting a firearm\\nin the presence of a passenger in a motor vehicle under\\nCalifornia Penal Code section 417.3, Bolanos, 734 F.3d at\\n878, assault under Arizona Revised Statute sections 13-\\n1203, 13-1204, Cabrera-Perez, 751 F.3d at 1007, and\\ncriminal threats under California Penal Code section 422,\\nRosales-Rosales, 347 F.3d at 717, are all crimes of violence.\\nAlthough none of these crimes can be committed\\n“recklessly,” how can our court logically conclude that\\nsecond-degree murder (the killing of a human being) is not\\na crime of violence when these lesser crimes are?\\n\\n\\n\\nStates v. Paul, 37 F.3d 496, 499 n.1 (9th Cir. 1994) (explaining voluntary\\nmanslaughter may involve “a defendant who killed unintentionally but\\nrecklessly with extreme disregard for human life . . . in the heat of\\npassion with adequate provocation”); see also Quintero, 21 F.3d at 890\\n(“Intent without malice, not the heat of passion, is the defining\\ncharacteristic of voluntary manslaughter.”). Yet, the Majority’s opinion\\nignores this difference in mens rea.\\n\\n UNITED STATES V. BEGAY 29\\n\\n Even more problematic with this decision: under the\\nMajority’s approach, attempted murder would be a crime of\\nviolence while the complete offense would not be. 24 See\\nBraxton v. United States, 500 U.S. 344, 351 n.**(1991)\\n(“Although a murder may be committed without an intent to\\nkill, an attempt to commit murder requires a specific intent\\nto kill.”). We reached this same conclusion in United States\\nv. Studhorse, 883 F.3d 1198. There, we first concluded that\\nthe crime could not be committed recklessly, because “the\\nmental state required for criminal attempt (specific intent) is\\nthe highest mental state,” “[r]egardless of the intent required\\nto commit the underlying crime.” Id. at 1203. Next, we\\nconcluded that Washington attempted first degree murder 25\\nfell under § 16(a)’s definition of “crime of violence.” See id.\\nat 1204 (Section 16(a) requires “an element the use,\\nattempted use, or threatened use of physical force against the\\nperson.” ). As with Studhorse, an attempt to commit murder\\nunder 18 U.S.C. § 1113 similarly would require a substantial\\nstep and an intent to commit the crime. United States v.\\nGoetzke, 494 F.3d 1231, 1237 (9th Cir. 2007) (per curiam).\\nIt is illogical to conclude that an attempted murder is a\\n\\n 24\\n This holding would also seem to conflict with convictions under\\n18 U.S.C. § 373, solicitation to commit a crime of violence, wherein a\\nperson solicits another to commit murder.\\n 25\\n Under Revised Code of Washington section 9A.32.030(1), first-\\ndegree murder includes “(b) Under circumstances manifesting an\\nextreme indifference to human life, he or she engages in conduct which\\ncreates a grave risk of death to any person, and thereby causes the death\\nof a person.” Under the Majority’s view, first-degree murder in\\nWashington would not be a crime of violence, which the Studhorse panel\\nrejected. 883 F.3d at 1204 (“[W]e easily conclude that Washington\\nattempted first degree murder falls within § 16(a)’s definition of a ‘crime\\nof violence.’”).\\n\\n30 UNITED STATES V. BEGAY\\n\\n“crime of violence” that could garner a higher sentence than\\nthat for the actual killing of a human being.\\n\\n The Majority’s opinion also conflicts with other statutes\\nand sentencing guidelines, which presume that murder is a\\ncrime of violence. 26 See, e.g., 18 U.S.C. § 3559(f)(1)\\n(increasing the mandatory minimum for violent crimes\\nagainst children “if the crime of violence is murder”);\\n 26\\n The Majority’s analysis could easily exclude first-degree murder\\nas a crime of violence. First- and second-degree murder both require that\\n(1) the defendant unlawfully killed another and (2) the defendant killed\\nwith malice aforethought (which can be committed “recklessly with\\nextreme disregard for human life”). See 8 U.S.C. § 1111(a); Ninth Cir.\\nCrim. Jury Inst. 8.107, 8.108. Although first degree murder can be:\\n\\n perpetrated by poison, lying in wait, or any other kind\\n of willful, deliberate, malicious, and premeditated\\n killing; or committed in the perpetration of, or attempt\\n to perpetrate, any arson, escape, murder, kidnapping,\\n treason, espionage, sabotage, aggravated sexual abuse\\n or sexual abuse, child abuse, burglary, or robbery; or\\n perpetrated as part of a pattern or practice of assault or\\n torture against a child or children; or perpetrated from\\n a premeditated design unlawfully and maliciously to\\n effect the death of any human being other than him\\n who is killed[,]\\n\\n18 U.S.C. § 1111(a), the“malice aforethought” element is still required.\\nId. Thus, if any degree of reckless conduct precludes a crime from being\\ndesignated as a crime of violence, by the Majority’s logic first-degree\\nmurder would not be a crime of violence either. It further defies reason\\nthat a defendant could be subject to the death penalty; yet not be found\\nto have committed a crime of violence. See 18 U.S.C. § 3591(a)(2)(D)\\n(The Federal Death Penalty Act recognizes that a murder could include\\nthe mens rea of “intentionally and specifically engag[ing] in an act of\\nviolence, knowing that the act created a grave risk of death to a person,\\nother than one of the participants in the offense, such that participation\\nin the act constituted a reckless disregard for human life and the victim\\ndied as a direct result of the act.”).\\n\\n UNITED STATES V. BEGAY 31\\n\\nU.S.S.G. § 4B1.2(a)(2) (“‘[C]rime of violence’ means . . .\\nmurder.”); U.S.S.G. § 2L1.2 cmt. n.2 (“‘Crime of violence’\\nmeans . . . murder . . . .”). Again, the question: why carve\\nout this exception for second-degree murder under § 16(a)\\nand § 924(c)(3)(A), when Congress has been clear that\\nmurder is a crime of violence?\\n\\n The Majority goes too far in extending the concept of\\nrecklessness to include malice aforethought. Malice\\naforethought, including “depraved heart” murder, requires\\nconduct well beyond mere recklessness. Murder should be\\na categorical match to a crime of violence because of the\\nnecessary element of malice aforethought. Holding\\notherwise makes the definition of “crime of violence”\\nvirtually meaningless and will lead to utterly absurd results\\nin future cases.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426680/\", \"author_raw\": \"SMITH, N.R., Circuit Judge, dissenting\"}]}"},"judge1":{"kind":"string","value":"DOROTHY W NELSON"},"judge2":{"kind":"string","value":"RICHARD R CLIFTON"},"judge3":{"kind":"string","value":"N RANDY SMITH"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649427/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Randly Begay\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-19\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 14-10080\n Plaintiff-Appellee,\n D.C. No.\n v. 3:13-cr-08073-\n NVW-1\nRANDLY IRVIN BEGAY, AKA Randly\nBegay,\n Defendant-Appellant. OPINION\n\n Appeal from the United States District Court\n for the District of Arizona\n Neil V. Wake, District Judge, Presiding\n\n Argued and Submission Deferred October 20, 2015\n Submitted May 26, 2016\n Submission Vacated and Deferred March 29, 2017\n Resubmitted August 19, 2019\n San Francisco, California\n\n Filed August 19, 2019\n\n Before: Dorothy W. Nelson, Richard R. Clifton,\n and N. Randy Smith, Circuit Judges.\n\n Opinion by Judge D.W. Nelson;\n Dissent by Judge N.R. Smith\n\n2 UNITED STATES V. BEGAY\n\n SUMMARY *\n\n\n Criminal Law\n\n The panel affirmed a conviction for second-degree\nmurder (18 U.S.C. §§ 1111 and 1153), reversed a conviction\nfor discharging a firearm during a “crime of violence”\n(18 U.S.C. § 924(c)(1)(A)), reversed a mandatory restitution\norder, and remanded for resentencing.\n\n Affirming the second-degree murder conviction, the\npanel held that the district court did not plainly err in failing\nto instruct the jury on absence of “heat of passion” as an\nelement of second-degree murder.\n\n The panel held that because second-degree murder can\nbe committed recklessly, it does not categorically constitute\na “crime of violence” under the elements clause, 18 U.S.C.\n§ 924(c)(3)(A). Because in light of United States v. Davis,\n139 S. Ct. 2319 (2019), second-degree murder likewise\ncannot constitute a crime of violence under the residual\nclause, 18 U.S.C. § 924(c)(3)(B), the panel concluded that\nthe defendant’s § 924(c) conviction cannot stand.\n\n The panel held that because second-degree murder is not\ncategorically a crime of violence, the district court erred in\nimposing mandatory restitution under 18 U.S.C. § 3663A.\n\n Dissenting from Parts II and III, Judge N.R. Smith would\nhold that second-degree murder is a crime of violence under\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n UNITED STATES V. BEGAY 3\n\n§ 924(c)(3)(A). He wrote that the majority goes too far in\nextending the concept of recklessness to include malice\naforethought, which, including “depraved heart” murder,\nrequires conduct well beyond mere recklessness.\n\n\n COUNSEL\n\nM. Edith Cunningham (argued), Assistant Federal Public\nDefender; Jon M. Sands, Federal Public Defender; Office of\nthe Federal Public Defender, Tucson, Arizona; for\nDefendant-Appellant.\n\nKarla Hotis Delord (argued), Assistant United States\nAttorney; Krissa M. Lanham, Deputy Appellate Chief;\nRobert L. Miskell, Appellate Chief; United States Attorney’s\nOffice, Phoenix, Arizona; for Plaintiff-Appellee.\n\n\n OPINION\n\nD.W. NELSON, Circuit Judge:\n\n Randly Irvin Begay appeals his jury conviction and\nsentence for second-degree murder under 18 U.S.C. §§ 1111\nand 1153 and discharging a firearm during a “crime of\nviolence” in violation of 18 U.S.C. § 924(c)(1)(A). We\naffirm Begay’s conviction for second-degree murder. We\nhold that second-degree murder is not a “crime of violence,”\nreverse the § 924(c) count of conviction, reverse the\nmandatory restitution order, and remand for resentencing.\n\n Begay’s conviction is AFFIRMED in part, REVERSED\nin part, and REMANDED for resentencing.\n\n4 UNITED STATES V. BEGAY\n\n FACTS AND PROCEEDINGS\n\n On March 4, 2013, Randly Begay, a Native American,\nshot Roderick Ben in the head with a handgun, killing Ben.\nBegay was charged with second-degree murder and\ndischarging a firearm during a “crime of violence.”\n\n After a few hours of drinking and smoking\nmethamphetamine, Begay, Ben, Begay’s nephew Lionel\nBegay, and Begay’s girlfriend Meghan Williams were\nsitting inside a van parked in front of Begay’s parents’\nresidence in Tuba City, Arizona, located within the Navajo\nNation Indian Reservation. While sitting inside the van,\nWilliams and Begay were arguing about Williams’ alleged\ninfidelity, including Begay’s accusations that she had been\ncheating on him with Ben. Begay stated he was tired of\npeople thinking that he was a “bitch” because of Williams\ncheating on him. During the argument, Begay pulled out a\ngun and laid it on his leg. Ben saw the gun and told Begay\nnot to shoot the windows of the van. Begay continued to\nargue with Williams, stating that he was not going to be a\n“bitch” anymore and that he was not scared to go to prison\nfor life. Begay then shot Ben in the head once.\n\n At trial, Begay’s theory of defense was that someone else\nin the car had shot Ben. In closing argument, Begay’s\nattorney briefly argued that Begay lacked the requisite\nmalice because he did not act deliberately. During closing\nargument, the government argued that Begay intentionally\nshot Ben because he was angry about Williams’ alleged\ninfidelity with Ben. Begay and the government submitted\njoint jury instructions, requesting an instruction on second-\ndegree murder only. Begay’s attorney did not request a\nvoluntary or involuntary manslaughter instruction, nor did\nhe object to the instructions as presented.\n\n UNITED STATES V. BEGAY 5\n\n The jury, instructed to find second-degree murder to be\na “crime of violence,” convicted Begay of second-degree\nmurder (Count One) and discharging a firearm during a\n“crime of violence” under 18 U.S.C. § 924(c)(1)(A)(iii)\n(Count Two). The district court sentenced Begay to 204\nmonths’ imprisonment on Count One and a consecutive term\nof 120 months’ imprisonment on Count Two. The district\ncourt imposed mandatory restitution under 18 U.S.C.\n§ 3663A in the amount of $23,622. Begay appeals his\nconviction on both counts as well as the restitution order.\n\n JURISDICTION\n\n An “Indian” who commits murder in “Indian country” is\nsubject to applicable federal criminal laws. 18 U.S.C.\n§ 1153(a). The location of the shooting here, the Navajo\nNation Indian Reservation, is “Indian country” for the\npurposes of § 1153. 18 U.S.C. § 1151. We have jurisdiction\nunder 28 U.S.C. § 1291.\n\n DISCUSSION\n\n I\n\n “[T]he Due Process Clause requires the prosecution to\nprove beyond a reasonable doubt the absence of the heat of\npassion on sudden provocation when the issue is properly\npresented in a homicide case.” Mullaney v. Wilbur, 421 U.S.\n684, 704 (1975). Where, as here, a defendant does not object\nto the jury instructions given by the district court, we review\nfor plain error. United States v. Matsumaru, 244 F.3d 1092,\n1102 (9th Cir. 2001).\n\n On appeal, Begay argues for the first time that the district\ncourt erred by failing to instruct the jury that, in order to\nestablish the element of malice, the government had to prove\n\n6 UNITED STATES V. BEGAY\n\nbeyond a reasonable doubt that Begay did not act in the heat\nof passion or upon a sudden quarrel. At trial, Begay’s\nattorney did not request a voluntary manslaughter\ninstruction, object to the absence of such an instruction, or\nintroduce evidence of sudden quarrel or heat of passion at\ntrial. Begay’s theory of defense was that he did not kill the\nvictim. Cf. United States v. Anderson, 201 F.3d 1145, 1153\n(9th Cir. 2000) (“A defendant arguing voluntary\nmanslaughter ‘attempts to negate the malice element by\nclaiming, in essence, that she was not acting maliciously\nbecause some extreme provocation . . . severely impaired her\ncapacity for self-control in committing the killing.’”\n(quoting United States v. Quintero, 21 F.3d 885, 890 (9th\nCir. 1994))).\n\n During closing argument, Begay’s attorney briefly\nargued that the government could not prove malice because\nBegay did not act deliberately, but did not argue that Begay\nacted in heat of passion or upon a sudden quarrel. The\nargument Begay’s attorney made regarding malice centered\non Begay’s lack of deliberation, arguing “to deliberate and\nto intend to do something is not to simply be high and pull\nthe trigger.” Begay’s attorney did not argue that Begay\nlacked malice because he shot Ben in heat of passion or\nbecause he was provoked. See Anderson, 201 F.3d at 1152\n(“A failure to give a jury instruction, even if error, does not\nseriously affect the fairness and integrity of judicial\nproceedings if the defense at trial made no argument relevant\nto the omitted instruction.”).\n\n Although the government introduced evidence at trial\nthat Begay may have been angry at the time of the shooting\nand discussed Begay’s anger during closing argument, the\nevidence, arguments, and jury instructions taken as a whole\ndid not “properly present” the issue of heat of passion or\n\n UNITED STATES V. BEGAY 7\n\nprovocation to the jury. See United States v. Roston,\n986 F.2d 1287, 1290 (9th Cir. 1993) (holding that the issue\nof heat of passion was not properly presented where there\nwas insufficient evidence of provocation and the defendant’s\ntheory of the case was that he did not kill the victim); cf.\nUnited States v. Lesina, 833 F.2d 156, 160 (9th Cir. 1987)\n(finding that the district court erred by failing to give\ndefendant’s requested instructions regarding accidental\nkilling and heat of passion where “that defense [was]\nraised”). As such, the district court did not plainly err in\nfailing to instruct the jury on absence of “heat of passion” as\nan element of second-degree murder. We therefore affirm\nBegay’s conviction under Count One for second-degree\nmurder.\n\n II\n\n “We review de novo whether a criminal conviction is a\n‘crime of violence’ [under section 924(c)(3)] and whether a\njury instruction misstated the elements of an offense.”\nUnited States v. Benally, 843 F.3d 350, 353 (9th Cir. 2016).\nWhere, as here, a party raises an argument for the first time\non appeal, we generally review for plain error; however, we\nare not limited to plain error review “when we are presented\nwith a question that ‘is purely one of law’ and where ‘the\nopposing party will suffer no prejudice as a result of the\nfailure to raise the issue in the trial court.’” United States v.\nSaavedra-Velazquez, 578 F.3d 1103, 1106 (9th Cir. 2009)\n(quoting United States v. Echavarria-Escobar, 270 F.3d\n1265, 1267–68 (9th Cir. 2001)). Here, whether second-\ndegree murder is a crime of violence is a pure question of\nlaw, and the government, which has fully briefed the issue,\nsuffers no prejudice. We therefore apply de novo review. See\nid. at 1106 (reviewing de novo whether California attempted\n\n8 UNITED STATES V. BEGAY\n\nrobbery is a crime of violence despite the petitioner’s failure\nto raise the issue before the district court).\n\n Begay was convicted of discharging a firearm during a\n“crime of violence” under 18 U.S.C. § 924(c). On appeal,\nBegay argues that second-degree murder does not qualify as\na “crime of violence.” To determine whether second-degree\nmurder is a “crime of violence” we apply the “categorical\napproach” laid out in Taylor v. United States, 495 U.S. 575\n(1990). Benally, 843 F.3d at 352. Based on the facts of this\ncase, it may be hard to understand how the shooting of Ben\nby Begay might not be a “crime of violence.” Under the\ncategorical approach, however, we do not look to the facts\nunderlying the conviction, but “compare the elements of the\nstatute forming the basis of the defendant’s conviction with\nthe elements of” a “crime of violence.” See Descamps v.\nUnited States, 570 U.S. 254, 257 (2013). The defendant’s\ncrime cannot be a categorical “crime of violence” if the\nconduct proscribed by the statute of conviction is broader\nthan the conduct encompassed by the statutory definition of\na “crime of violence.” See id.\n\n If the statute of conviction is overbroad, we may ask\nwhether the statute is divisible, and, if it is, apply the\n“modified categorical approach” to determine which\n“statutory phrase” formed the basis of the defendant’s\nconviction. Descamps, 570 U.S. at 263 (quoting Nijhawan v.\nHolder, 557 U.S. 29, 41 (2009)). Where, as here, the\ngovernment has not argued that the statute of conviction is\ndivisible, we need not conduct a modified categorical\nanalysis. See United States v. Walton, 881 F.3d 768, 774–75\n(9th Cir. 2018) (declining to conduct a modified categorical\nanalysis because the government failed to argue that the\nstatute of conviction was divisible).\n\n UNITED STATES V. BEGAY 9\n\n Accordingly, we compare the elements of 18 U.S.C.\n§ 1111, the second-degree murder statute, to the definition\nof a “crime of violence” found in § 924(c)(3). A “crime of\nviolence” is defined as a felony that\n\n (A) has as an element the use, attempted use,\n or threatened use of physical force against the\n person or property of another, or\n\n (B) that by its nature, involves a substantial\n risk that physical force against the person or\n property of another may be used in the course\n of committing the offense.\n\n18 U.S.C. § 924(c)(3). Subsection (A) is commonly known\nas the “elements clause” and subsection (B) is commonly\nknown as the “residual clause.” Because the Supreme Court\ndeclared 18 U.S.C. § 924 (c)(3)(B) unconstitutionally vague\nin United States v. Davis, 139 S.Ct. 2319, 2336 (2019), we\nneed only determine whether second-degree murder\nconstitutes a crime of violence under the “elements clause”\nin subsection (A).\n\n A\n\n Second-degree murder does not constitute a crime of\nviolence under the elements clause—18 U.S.C.\n§ 924(c)(3)(A)—because it can be committed recklessly.\n\n “[B]ecause the wording of [18 U.S.C. § 924(c)(3) and\n18 U.S.C. § 16] is virtually identical, we interpret their plain\nlanguage in the same manner.” Benally, 843 F.3d at 354\n(analyzing the required mental state for § 924(c)(3) by\nlooking to case law interpreting § 16); see also Davis,\n139 S.Ct. at 2326 (“Like § 924(c)(3), § 16 contains an\nelements clause and a residual clause. The only difference is\n\n10 UNITED STATES V. BEGAY\n\nthat § 16’s elements clause, unlike § 924(c)(3)’s elements\nclause, isn’t limited to felonies . . . .”). 18 U.S.C. § 16\ndefines the term “crime of violence” as:\n\n (a) an offense that has an element the use,\n attempted use, or threatened use of physical\n force against the person or property of\n another, or\n\n (b) any other offense that is a felony and that,\n by its nature, involves a substantial risk that\n physical force against the person or property\n of another may be used in the course of\n committing the offense.\n\n(emphasis added). The only substantive difference is that the\nfelony requirement applies to both subsections of\n§ 924(c)(3) and only to subsection (b) of § 16, but this\ndifference “does not affect the operative language used to\ninterpret the statute’s requisite mental state.” Benally,\n843 F.3d at 354 n.1; see also Smith v. City of Jackson, 544\nU.S. 228, 233 (2005) (plurality) (holding that a term is given\nthe same meaning “when Congress uses the same language\nin two statutes having similar purposes.”).\n\n A “crime of violence” requires intentional conduct. In\nLeocal v. Ashcroft, the Supreme Court discussed the mens\nrea necessary to commit a “crime of violence” under\n18 U.S.C. § 16. 543 U.S. 1 (2004). The Supreme Court\nreasoned that § 16’s requirement that force be used “against”\nsomeone or something suggests that “crimes of violence”\nrequire “a higher degree of intent than negligent or merely\naccidental conduct.” Leocal, 543 U.S. at 9–11.\n\n We have since interpreted Leocal’s reasoning to hold\nthat “crimes of violence,” as defined in both § 16 and\n\n UNITED STATES V. BEGAY 11\n\n§ 924(c), require purposeful conduct. Benally, 843 F.3d at\n353–54 (applying Leocal and Fernandez-Ruiz v. Gonzales,\n466 F.3d 1121 (9th Cir. 2006) (en banc) to find that\ninvoluntary manslaughter, which requires only gross\nnegligence, is not a crime of violence under § 924(c));\nCovarrubias Teposte v. Holder, 632 F.3d 1049, 1053 (9th\nCir. 2011) (concluding from Leocal, Fernandez-Ruiz, and\nUnited States v. Gomez-Leon, 545 F.3d 777 (9th Cir. 2008)\nthat an intentional use of force was required for a crime of\nviolence as defined in either subsection of 18 U.S.C. § 16);\nGomez-Leon, 545 F.3d at 787 (holding that a crime of\nviolence under 18 U.S.C. § 16 “must require proof of an\nintentional use of force or a substantial risk that force will be\nintentionally used during its commission”); Fernandez-Ruiz,\n466 F.3d at 1130 (holding that crimes that can be committed\nrecklessly are not “crimes of violence” for the purposes of\n§ 16 because reckless conduct “is not purposeful”).\n\n The government argues that Voisine v. United States, —\nU.S. — , 136 S.Ct. 2272 (2016) “implicitly overruled\nFernandez-Ruiz.” As we recently explained, “Voisine\nexpressly left open the question that Fernandez-Ruiz\nanswered” and is not “so clearly irreconcilable with\nFernandez-Ruiz’s reasoning that this three-judge panel is no\nlonger bound by the precedent of our court.” United States\nv. Orona, 923 F.3d 1197, 1203 (9th Cir. 2019). In Voisine,\nthe Supreme Court held that the definition of “misdemeanor\ncrime of domestic violence” in 18 U.S.C. § 921(a)(33)(A)\nincludes reckless assaults. Id. at 2278. In doing so, however,\nthe Court explicitly stated, “our decision today concerning\n§ 921(a)(33)(A)’s scope does not resolve whether § 16\nincludes reckless behavior. Courts have sometimes given\nthese two statutory definitions divergent readings in light of\ndifferences in their contexts and purposes, and we do not\nforeclose that possibility with respect to the required mental\n\n12 UNITED STATES V. BEGAY\n\nstates.” Id. at 2280 n.4. Given that explicit qualification, the\nunique “context[] and purpose[]” of domestic violence, and\nthe differences in statutory text between the provision at\nissue and the one interpreted in Voisine, we continue to apply\nour case law interpreting 18 U.S.C. § 16 to 18 U.S.C.\n§ 924(c)(3). 1 See Voisine, 136 S.Ct. at 2290 (limiting the\nholding to the context of misdemeanor domestic violence);\nUnited States v. Castleman, 572 U.S. 157, 162–63 (2014)\n(same); Benally, 843 at 354 (continuing to apply our\nprecedent interpreting § 16 in interpreting § 924(c)(3)\n“because the wording of the two statutes is virtually\nidentical”). It follows from our precedent that a crime of\nviolence under 18 U.S.C. § 924(c)(3) requires the intentional\nuse of force.\n\n The elements of second-degree murder are that the\ndefendant (1) “unlawfully kill[ed] a human being” (2) “with\nmalice aforethought.” 18 U.S.C. § 1111(a); Ninth Circuit\nModel Criminal Jury Instruction 8.108. “[M]alice\naforethought covers four different kinds of mental states:\n(1) intent to kill; (2) intent to do serious bodily injury;\n(3) depraved heart (i.e., reckless indifference); and (4) intent\nto commit a felony.” United States v. Pineda-Doval,\n614 F.3d 1019, 1038 (9th Cir. 2010). As such, second-degree\nmurder may be committed recklessly—with a depraved\nheart mental state—and need not be committed willfully or\nintentionally. See United States v. Houser, 130 F.3d 867,\n871–72 (9th Cir. 1997) (“Malice aforethought does not\nrequire an element of willfulness if the existence of that\n\n\n\n 1\n See also Gonzales-Ramirez v. Sessions, 727 F. App’x 404, 405 n.7\n(9th Cir. 2018) (unpublished) (concluding that Voisine “does not affect\nour § 16(a) case law.”).\n\n UNITED STATES V. BEGAY 13\n\nmalice is inferred from the fact that defendant acted\nrecklessly with extreme disregard for human life.”).\n\n It is of no consequence that the recklessness required for\nsecond-degree murder must be “extreme” and goes beyond\nordinary recklessness. In Gomez-Leon, we made clear that,\nin order to constitute a crime of violence, “the underlying\noffense must require proof of an intentional use of force or a\nsubstantial risk that force will be intentionally used during\nits commission.” 545 F.3d at 787. “[O]ur precedent seems\nsquarely to place crimes motivated by intent on a pedestal,\nwhile pushing off other very dangerous and violent conduct\nthat, because not intentional, does not qualify as a ‘crime of\nviolence.’” Covarrubias, 632 F.3d at 1053. Reckless\nconduct, no matter how extreme, is not intentional.\n\n Second-degree murder also does not involve a\n“substantial risk that force will be intentionally used during\nits commission.” Gomez-Leon, 545 F.3d at 787. In\nCovarrubias, we held that a California offense prohibiting\nthe malicious and willful discharge of a firearm at an\ninhabited dwelling was not a “crime of violence” because it\ncould be committed recklessly, not just intentionally.\nCovarrubias, 632 F.3d at 1053–55. Although we conducted\nour analysis under § 16(b) because the BIA rested its\ndecision on subsection (b), id. at 1052, the analysis regarding\nintent bears upon either subsection of § 16, and by analogy,\n18 U.S.C. § 924(c)(3). See, e.g., Gomez-Leon, 545 F.3d\nat 787 (requiring intentional use of force for a crime of\nviolence under either subsection of § 16); Benally, 843 F.3d\nat 354. In contrast to crimes like burglary that can be\ncommitted only intentionally, “with a crime committed\nrecklessly, it is unlikely that the reckless actor will, in\nresponse to external events, form an intent to use force in\nfurtherance of his crime.” Covarrubias, 632 F.3d at 1055.\n\n14 UNITED STATES V. BEGAY\n\n“Classic examples of second-degree murder include\nshooting a gun into a room that the defendant knows to be\noccupied, a game of Russian roulette, and driving a car at\nvery high speeds along a crowded main street . . . .” Pineda-\nDoval, 614 F.3d at 1039. For purposes of our analysis, these\nexamples are substantively indistinguishable from the\noffense—“Shooting at an Inhabited Dwelling or Vehicle”—\nthat we held was not categorically a crime of violence in\nCovarrubias. This risk that a crime could escalate to the use\nof intentional force is no more substantial for a defendant\nwho recklessly kills than it is for a defendant who recklessly\nshoots at a house.\n\n The cases the government cites do not squarely address\nwhether second-degree murder is a crime of violence.\nInstead, in those cases, we found that other challenges to\n§ 924(c) convictions lacked merit. See, e.g., United States v.\nPercy, 250 F.3d 720 (9th Cir. 2001) (addressing the Sixth\nAmendment right to counsel and prosecutorial misconduct);\nHouser, 130 F.3d 867 (discussing the role of the jury and the\nmens rea required for second-degree murder). More\nrecently, in United States v. JJ, we reviewed a district court's\ndecision to try a juvenile as an adult under the Federal\nJuvenile Delinquency Act, 18 U.S.C. § 5031 et seq. United\nStates v. JJ, 704 F.3d 1219 (9th Cir. 2013). In JJ, the\nquestion of whether “second-degree murder, if committed by\nan adult, would be a felony crime of violence” was not in\ndispute. Id. at 1222. Far from “holding” that second-degree\nmurder is a “crime of violence” for purposes of § 924(c), JJ\nfocused on when it might be in the interest of justice to try a\njuvenile as an adult. Id. Because second-degree murder can\n\n UNITED STATES V. BEGAY 15\n\nbe committed recklessly, rather than intentionally, it does not\ncategorically constitute a crime of violence. 2\n\n B\n\n Second-degree murder is not categorically a crime of\nviolence under the elements clause, 18 U.S.C.\n§ 924(c)(3)(A). And, pursuant to Davis, second-degree\nmurder cannot constitute a crime of violence under the\nresidual clause, section 924(c)(3)(B), as the residual clause\nis unconstitutionally vague. Begay’s § 924(c) conviction for\ndischarging a firearm during and in relation to a crime of\nviolence therefore cannot stand under either the elements\nclause or residual clause of § 924(c)(3). We reverse Count\nTwo of his conviction.\n\n III\n\n We review the legality of a restitution order de novo and\nthe factual findings supporting the order for clear error.\nUnited States v. Luis, 765 F.3d 1061, 1065 (9th Cir. 2014)\n(quoting United States v. Brock-Davis, 504 F.3d 991, 996\n(9th Cir. 2007)). Where, as here, a defendant raises an\nobjection to a restitution order for the first time on appeal,\nwe review for plain error. United States v. Van Alystne, 584\nF.3d 803, 819 (9th Cir. 2009).\n\n 18 U.S.C. § 3663A directs a district court to impose\nmandatory restitution where a defendant is convicted of a\ncrime of violence, as defined in 18 U.S.C. § 16. 18 U.S.C.\n\n 2\n In holding that second-degree murder does not qualify as a\ncategorical crime of violence because it can be committed recklessly, we\ndo not reach Begay's argument that second-degree murder does not\nqualify as a categorical crime of violence for the additional reason that it\ncan be committed without the use of violent force.\n\n16 UNITED STATES V. BEGAY\n\n§ 3663A(c)(1)(A)(i). We conclude that the district court\nplainly erred in imposing mandatory restitution under\n18 U.S.C. § 3663A because second-degree murder is not\ncategorically a crime of violence, and vacate the mandatory\nrestitution order. In light of this disposition, we need not\naddress Begay’s claim that the district court failed to make\nthe necessary findings for the restitution order.\n\n CONCLUSION\n\n We AFFIRM Count One of Begay’s conviction for\nsecond-degree murder. We REVERSE Count Two of\nBegay’s conviction for discharging a firearm during a “crime\nof violence” under 18 U.S.C. § 924(c)(1)(A), VACATE the\nmandatory restitution order imposed under 18 U.S.C.\n§ 3663A, and REMAND for resentencing.\n\n AFFIRMED IN PART, REVERSED IN PART,\nVACATED IN PART, AND REMANDED.\n\n UNITED STATES V. BEGAY 17\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Randly Begay\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-19\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 14-10080\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:13-cr-08073-\\n NVW-1\\nRANDLY IRVIN BEGAY, AKA Randly\\nBegay,\\n Defendant-Appellant. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Neil V. Wake, District Judge, Presiding\\n\\n Argued and Submission Deferred October 20, 2015\\n Submitted May 26, 2016\\n Submission Vacated and Deferred March 29, 2017\\n Resubmitted August 19, 2019\\n San Francisco, California\\n\\n Filed August 19, 2019\\n\\n Before: Dorothy W. Nelson, Richard R. Clifton,\\n and N. Randy Smith, Circuit Judges.\\n\\n Opinion by Judge D.W. Nelson;\\n Dissent by Judge N.R. Smith\\n\\n2 UNITED STATES V. BEGAY\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n The panel affirmed a conviction for second-degree\\nmurder (18 U.S.C. §§ 1111 and 1153), reversed a conviction\\nfor discharging a firearm during a “crime of violence”\\n(18 U.S.C. § 924(c)(1)(A)), reversed a mandatory restitution\\norder, and remanded for resentencing.\\n\\n Affirming the second-degree murder conviction, the\\npanel held that the district court did not plainly err in failing\\nto instruct the jury on absence of “heat of passion” as an\\nelement of second-degree murder.\\n\\n The panel held that because second-degree murder can\\nbe committed recklessly, it does not categorically constitute\\na “crime of violence” under the elements clause, 18 U.S.C.\\n§ 924(c)(3)(A). Because in light of United States v. Davis,\\n139 S. Ct. 2319 (2019), second-degree murder likewise\\ncannot constitute a crime of violence under the residual\\nclause, 18 U.S.C. § 924(c)(3)(B), the panel concluded that\\nthe defendant’s § 924(c) conviction cannot stand.\\n\\n The panel held that because second-degree murder is not\\ncategorically a crime of violence, the district court erred in\\nimposing mandatory restitution under 18 U.S.C. § 3663A.\\n\\n Dissenting from Parts II and III, Judge N.R. Smith would\\nhold that second-degree murder is a crime of violence under\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. BEGAY 3\\n\\n§ 924(c)(3)(A). He wrote that the majority goes too far in\\nextending the concept of recklessness to include malice\\naforethought, which, including “depraved heart” murder,\\nrequires conduct well beyond mere recklessness.\\n\\n\\n COUNSEL\\n\\nM. Edith Cunningham (argued), Assistant Federal Public\\nDefender; Jon M. Sands, Federal Public Defender; Office of\\nthe Federal Public Defender, Tucson, Arizona; for\\nDefendant-Appellant.\\n\\nKarla Hotis Delord (argued), Assistant United States\\nAttorney; Krissa M. Lanham, Deputy Appellate Chief;\\nRobert L. Miskell, Appellate Chief; United States Attorney’s\\nOffice, Phoenix, Arizona; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nD.W. NELSON, Circuit Judge:\\n\\n Randly Irvin Begay appeals his jury conviction and\\nsentence for second-degree murder under 18 U.S.C. §§ 1111\\nand 1153 and discharging a firearm during a “crime of\\nviolence” in violation of 18 U.S.C. § 924(c)(1)(A). We\\naffirm Begay’s conviction for second-degree murder. We\\nhold that second-degree murder is not a “crime of violence,”\\nreverse the § 924(c) count of conviction, reverse the\\nmandatory restitution order, and remand for resentencing.\\n\\n Begay’s conviction is AFFIRMED in part, REVERSED\\nin part, and REMANDED for resentencing.\\n\\n4 UNITED STATES V. BEGAY\\n\\n FACTS AND PROCEEDINGS\\n\\n On March 4, 2013, Randly Begay, a Native American,\\nshot Roderick Ben in the head with a handgun, killing Ben.\\nBegay was charged with second-degree murder and\\ndischarging a firearm during a “crime of violence.”\\n\\n After a few hours of drinking and smoking\\nmethamphetamine, Begay, Ben, Begay’s nephew Lionel\\nBegay, and Begay’s girlfriend Meghan Williams were\\nsitting inside a van parked in front of Begay’s parents’\\nresidence in Tuba City, Arizona, located within the Navajo\\nNation Indian Reservation. While sitting inside the van,\\nWilliams and Begay were arguing about Williams’ alleged\\ninfidelity, including Begay’s accusations that she had been\\ncheating on him with Ben. Begay stated he was tired of\\npeople thinking that he was a “bitch” because of Williams\\ncheating on him. During the argument, Begay pulled out a\\ngun and laid it on his leg. Ben saw the gun and told Begay\\nnot to shoot the windows of the van. Begay continued to\\nargue with Williams, stating that he was not going to be a\\n“bitch” anymore and that he was not scared to go to prison\\nfor life. Begay then shot Ben in the head once.\\n\\n At trial, Begay’s theory of defense was that someone else\\nin the car had shot Ben. In closing argument, Begay’s\\nattorney briefly argued that Begay lacked the requisite\\nmalice because he did not act deliberately. During closing\\nargument, the government argued that Begay intentionally\\nshot Ben because he was angry about Williams’ alleged\\ninfidelity with Ben. Begay and the government submitted\\njoint jury instructions, requesting an instruction on second-\\ndegree murder only. Begay’s attorney did not request a\\nvoluntary or involuntary manslaughter instruction, nor did\\nhe object to the instructions as presented.\\n\\n UNITED STATES V. BEGAY 5\\n\\n The jury, instructed to find second-degree murder to be\\na “crime of violence,” convicted Begay of second-degree\\nmurder (Count One) and discharging a firearm during a\\n“crime of violence” under 18 U.S.C. § 924(c)(1)(A)(iii)\\n(Count Two). The district court sentenced Begay to 204\\nmonths’ imprisonment on Count One and a consecutive term\\nof 120 months’ imprisonment on Count Two. The district\\ncourt imposed mandatory restitution under 18 U.S.C.\\n§ 3663A in the amount of $23,622. Begay appeals his\\nconviction on both counts as well as the restitution order.\\n\\n JURISDICTION\\n\\n An “Indian” who commits murder in “Indian country” is\\nsubject to applicable federal criminal laws. 18 U.S.C.\\n§ 1153(a). The location of the shooting here, the Navajo\\nNation Indian Reservation, is “Indian country” for the\\npurposes of § 1153. 18 U.S.C. § 1151. We have jurisdiction\\nunder 28 U.S.C. § 1291.\\n\\n DISCUSSION\\n\\n I\\n\\n “[T]he Due Process Clause requires the prosecution to\\nprove beyond a reasonable doubt the absence of the heat of\\npassion on sudden provocation when the issue is properly\\npresented in a homicide case.” Mullaney v. Wilbur, 421 U.S.\\n684, 704 (1975). Where, as here, a defendant does not object\\nto the jury instructions given by the district court, we review\\nfor plain error. United States v. Matsumaru, 244 F.3d 1092,\\n1102 (9th Cir. 2001).\\n\\n On appeal, Begay argues for the first time that the district\\ncourt erred by failing to instruct the jury that, in order to\\nestablish the element of malice, the government had to prove\\n\\n6 UNITED STATES V. BEGAY\\n\\nbeyond a reasonable doubt that Begay did not act in the heat\\nof passion or upon a sudden quarrel. At trial, Begay’s\\nattorney did not request a voluntary manslaughter\\ninstruction, object to the absence of such an instruction, or\\nintroduce evidence of sudden quarrel or heat of passion at\\ntrial. Begay’s theory of defense was that he did not kill the\\nvictim. Cf. United States v. Anderson, 201 F.3d 1145, 1153\\n(9th Cir. 2000) (“A defendant arguing voluntary\\nmanslaughter ‘attempts to negate the malice element by\\nclaiming, in essence, that she was not acting maliciously\\nbecause some extreme provocation . . . severely impaired her\\ncapacity for self-control in committing the killing.’”\\n(quoting United States v. Quintero, 21 F.3d 885, 890 (9th\\nCir. 1994))).\\n\\n During closing argument, Begay’s attorney briefly\\nargued that the government could not prove malice because\\nBegay did not act deliberately, but did not argue that Begay\\nacted in heat of passion or upon a sudden quarrel. The\\nargument Begay’s attorney made regarding malice centered\\non Begay’s lack of deliberation, arguing “to deliberate and\\nto intend to do something is not to simply be high and pull\\nthe trigger.” Begay’s attorney did not argue that Begay\\nlacked malice because he shot Ben in heat of passion or\\nbecause he was provoked. See Anderson, 201 F.3d at 1152\\n(“A failure to give a jury instruction, even if error, does not\\nseriously affect the fairness and integrity of judicial\\nproceedings if the defense at trial made no argument relevant\\nto the omitted instruction.”).\\n\\n Although the government introduced evidence at trial\\nthat Begay may have been angry at the time of the shooting\\nand discussed Begay’s anger during closing argument, the\\nevidence, arguments, and jury instructions taken as a whole\\ndid not “properly present” the issue of heat of passion or\\n\\n UNITED STATES V. BEGAY 7\\n\\nprovocation to the jury. See United States v. Roston,\\n986 F.2d 1287, 1290 (9th Cir. 1993) (holding that the issue\\nof heat of passion was not properly presented where there\\nwas insufficient evidence of provocation and the defendant’s\\ntheory of the case was that he did not kill the victim); cf.\\nUnited States v. Lesina, 833 F.2d 156, 160 (9th Cir. 1987)\\n(finding that the district court erred by failing to give\\ndefendant’s requested instructions regarding accidental\\nkilling and heat of passion where “that defense [was]\\nraised”). As such, the district court did not plainly err in\\nfailing to instruct the jury on absence of “heat of passion” as\\nan element of second-degree murder. We therefore affirm\\nBegay’s conviction under Count One for second-degree\\nmurder.\\n\\n II\\n\\n “We review de novo whether a criminal conviction is a\\n‘crime of violence’ [under section 924(c)(3)] and whether a\\njury instruction misstated the elements of an offense.”\\nUnited States v. Benally, 843 F.3d 350, 353 (9th Cir. 2016).\\nWhere, as here, a party raises an argument for the first time\\non appeal, we generally review for plain error; however, we\\nare not limited to plain error review “when we are presented\\nwith a question that ‘is purely one of law’ and where ‘the\\nopposing party will suffer no prejudice as a result of the\\nfailure to raise the issue in the trial court.’” United States v.\\nSaavedra-Velazquez, 578 F.3d 1103, 1106 (9th Cir. 2009)\\n(quoting United States v. Echavarria-Escobar, 270 F.3d\\n1265, 1267–68 (9th Cir. 2001)). Here, whether second-\\ndegree murder is a crime of violence is a pure question of\\nlaw, and the government, which has fully briefed the issue,\\nsuffers no prejudice. We therefore apply de novo review. See\\nid. at 1106 (reviewing de novo whether California attempted\\n\\n8 UNITED STATES V. BEGAY\\n\\nrobbery is a crime of violence despite the petitioner’s failure\\nto raise the issue before the district court).\\n\\n Begay was convicted of discharging a firearm during a\\n“crime of violence” under 18 U.S.C. § 924(c). On appeal,\\nBegay argues that second-degree murder does not qualify as\\na “crime of violence.” To determine whether second-degree\\nmurder is a “crime of violence” we apply the “categorical\\napproach” laid out in Taylor v. United States, 495 U.S. 575\\n(1990). Benally, 843 F.3d at 352. Based on the facts of this\\ncase, it may be hard to understand how the shooting of Ben\\nby Begay might not be a “crime of violence.” Under the\\ncategorical approach, however, we do not look to the facts\\nunderlying the conviction, but “compare the elements of the\\nstatute forming the basis of the defendant’s conviction with\\nthe elements of” a “crime of violence.” See Descamps v.\\nUnited States, 570 U.S. 254, 257 (2013). The defendant’s\\ncrime cannot be a categorical “crime of violence” if the\\nconduct proscribed by the statute of conviction is broader\\nthan the conduct encompassed by the statutory definition of\\na “crime of violence.” See id.\\n\\n If the statute of conviction is overbroad, we may ask\\nwhether the statute is divisible, and, if it is, apply the\\n“modified categorical approach” to determine which\\n“statutory phrase” formed the basis of the defendant’s\\nconviction. Descamps, 570 U.S. at 263 (quoting Nijhawan v.\\nHolder, 557 U.S. 29, 41 (2009)). Where, as here, the\\ngovernment has not argued that the statute of conviction is\\ndivisible, we need not conduct a modified categorical\\nanalysis. See United States v. Walton, 881 F.3d 768, 774–75\\n(9th Cir. 2018) (declining to conduct a modified categorical\\nanalysis because the government failed to argue that the\\nstatute of conviction was divisible).\\n\\n UNITED STATES V. BEGAY 9\\n\\n Accordingly, we compare the elements of 18 U.S.C.\\n§ 1111, the second-degree murder statute, to the definition\\nof a “crime of violence” found in § 924(c)(3). A “crime of\\nviolence” is defined as a felony that\\n\\n (A) has as an element the use, attempted use,\\n or threatened use of physical force against the\\n person or property of another, or\\n\\n (B) that by its nature, involves a substantial\\n risk that physical force against the person or\\n property of another may be used in the course\\n of committing the offense.\\n\\n18 U.S.C. § 924(c)(3). Subsection (A) is commonly known\\nas the “elements clause” and subsection (B) is commonly\\nknown as the “residual clause.” Because the Supreme Court\\ndeclared 18 U.S.C. § 924 (c)(3)(B) unconstitutionally vague\\nin United States v. Davis, 139 S.Ct. 2319, 2336 (2019), we\\nneed only determine whether second-degree murder\\nconstitutes a crime of violence under the “elements clause”\\nin subsection (A).\\n\\n A\\n\\n Second-degree murder does not constitute a crime of\\nviolence under the elements clause—18 U.S.C.\\n§ 924(c)(3)(A)—because it can be committed recklessly.\\n\\n “[B]ecause the wording of [18 U.S.C. § 924(c)(3) and\\n18 U.S.C. § 16] is virtually identical, we interpret their plain\\nlanguage in the same manner.” Benally, 843 F.3d at 354\\n(analyzing the required mental state for § 924(c)(3) by\\nlooking to case law interpreting § 16); see also Davis,\\n139 S.Ct. at 2326 (“Like § 924(c)(3), § 16 contains an\\nelements clause and a residual clause. The only difference is\\n\\n10 UNITED STATES V. BEGAY\\n\\nthat § 16’s elements clause, unlike § 924(c)(3)’s elements\\nclause, isn’t limited to felonies . . . .”). 18 U.S.C. § 16\\ndefines the term “crime of violence” as:\\n\\n (a) an offense that has an element the use,\\n attempted use, or threatened use of physical\\n force against the person or property of\\n another, or\\n\\n (b) any other offense that is a felony and that,\\n by its nature, involves a substantial risk that\\n physical force against the person or property\\n of another may be used in the course of\\n committing the offense.\\n\\n(emphasis added). The only substantive difference is that the\\nfelony requirement applies to both subsections of\\n§ 924(c)(3) and only to subsection (b) of § 16, but this\\ndifference “does not affect the operative language used to\\ninterpret the statute’s requisite mental state.” Benally,\\n843 F.3d at 354 n.1; see also Smith v. City of Jackson, 544\\nU.S. 228, 233 (2005) (plurality) (holding that a term is given\\nthe same meaning “when Congress uses the same language\\nin two statutes having similar purposes.”).\\n\\n A “crime of violence” requires intentional conduct. In\\nLeocal v. Ashcroft, the Supreme Court discussed the mens\\nrea necessary to commit a “crime of violence” under\\n18 U.S.C. § 16. 543 U.S. 1 (2004). The Supreme Court\\nreasoned that § 16’s requirement that force be used “against”\\nsomeone or something suggests that “crimes of violence”\\nrequire “a higher degree of intent than negligent or merely\\naccidental conduct.” Leocal, 543 U.S. at 9–11.\\n\\n We have since interpreted Leocal’s reasoning to hold\\nthat “crimes of violence,” as defined in both § 16 and\\n\\n UNITED STATES V. BEGAY 11\\n\\n§ 924(c), require purposeful conduct. Benally, 843 F.3d at\\n353–54 (applying Leocal and Fernandez-Ruiz v. Gonzales,\\n466 F.3d 1121 (9th Cir. 2006) (en banc) to find that\\ninvoluntary manslaughter, which requires only gross\\nnegligence, is not a crime of violence under § 924(c));\\nCovarrubias Teposte v. Holder, 632 F.3d 1049, 1053 (9th\\nCir. 2011) (concluding from Leocal, Fernandez-Ruiz, and\\nUnited States v. Gomez-Leon, 545 F.3d 777 (9th Cir. 2008)\\nthat an intentional use of force was required for a crime of\\nviolence as defined in either subsection of 18 U.S.C. § 16);\\nGomez-Leon, 545 F.3d at 787 (holding that a crime of\\nviolence under 18 U.S.C. § 16 “must require proof of an\\nintentional use of force or a substantial risk that force will be\\nintentionally used during its commission”); Fernandez-Ruiz,\\n466 F.3d at 1130 (holding that crimes that can be committed\\nrecklessly are not “crimes of violence” for the purposes of\\n§ 16 because reckless conduct “is not purposeful”).\\n\\n The government argues that Voisine v. United States, —\\nU.S. — , 136 S.Ct. 2272 (2016) “implicitly overruled\\nFernandez-Ruiz.” As we recently explained, “Voisine\\nexpressly left open the question that Fernandez-Ruiz\\nanswered” and is not “so clearly irreconcilable with\\nFernandez-Ruiz’s reasoning that this three-judge panel is no\\nlonger bound by the precedent of our court.” United States\\nv. Orona, 923 F.3d 1197, 1203 (9th Cir. 2019). In Voisine,\\nthe Supreme Court held that the definition of “misdemeanor\\ncrime of domestic violence” in 18 U.S.C. § 921(a)(33)(A)\\nincludes reckless assaults. Id. at 2278. In doing so, however,\\nthe Court explicitly stated, “our decision today concerning\\n§ 921(a)(33)(A)’s scope does not resolve whether § 16\\nincludes reckless behavior. Courts have sometimes given\\nthese two statutory definitions divergent readings in light of\\ndifferences in their contexts and purposes, and we do not\\nforeclose that possibility with respect to the required mental\\n\\n12 UNITED STATES V. BEGAY\\n\\nstates.” Id. at 2280 n.4. Given that explicit qualification, the\\nunique “context[] and purpose[]” of domestic violence, and\\nthe differences in statutory text between the provision at\\nissue and the one interpreted in Voisine, we continue to apply\\nour case law interpreting 18 U.S.C. § 16 to 18 U.S.C.\\n§ 924(c)(3). 1 See Voisine, 136 S.Ct. at 2290 (limiting the\\nholding to the context of misdemeanor domestic violence);\\nUnited States v. Castleman, 572 U.S. 157, 162–63 (2014)\\n(same); Benally, 843 at 354 (continuing to apply our\\nprecedent interpreting § 16 in interpreting § 924(c)(3)\\n“because the wording of the two statutes is virtually\\nidentical”). It follows from our precedent that a crime of\\nviolence under 18 U.S.C. § 924(c)(3) requires the intentional\\nuse of force.\\n\\n The elements of second-degree murder are that the\\ndefendant (1) “unlawfully kill[ed] a human being” (2) “with\\nmalice aforethought.” 18 U.S.C. § 1111(a); Ninth Circuit\\nModel Criminal Jury Instruction 8.108. “[M]alice\\naforethought covers four different kinds of mental states:\\n(1) intent to kill; (2) intent to do serious bodily injury;\\n(3) depraved heart (i.e., reckless indifference); and (4) intent\\nto commit a felony.” United States v. Pineda-Doval,\\n614 F.3d 1019, 1038 (9th Cir. 2010). As such, second-degree\\nmurder may be committed recklessly—with a depraved\\nheart mental state—and need not be committed willfully or\\nintentionally. See United States v. Houser, 130 F.3d 867,\\n871–72 (9th Cir. 1997) (“Malice aforethought does not\\nrequire an element of willfulness if the existence of that\\n\\n\\n\\n 1\\n See also Gonzales-Ramirez v. Sessions, 727 F. App’x 404, 405 n.7\\n(9th Cir. 2018) (unpublished) (concluding that Voisine “does not affect\\nour § 16(a) case law.”).\\n\\n UNITED STATES V. BEGAY 13\\n\\nmalice is inferred from the fact that defendant acted\\nrecklessly with extreme disregard for human life.”).\\n\\n It is of no consequence that the recklessness required for\\nsecond-degree murder must be “extreme” and goes beyond\\nordinary recklessness. In Gomez-Leon, we made clear that,\\nin order to constitute a crime of violence, “the underlying\\noffense must require proof of an intentional use of force or a\\nsubstantial risk that force will be intentionally used during\\nits commission.” 545 F.3d at 787. “[O]ur precedent seems\\nsquarely to place crimes motivated by intent on a pedestal,\\nwhile pushing off other very dangerous and violent conduct\\nthat, because not intentional, does not qualify as a ‘crime of\\nviolence.’” Covarrubias, 632 F.3d at 1053. Reckless\\nconduct, no matter how extreme, is not intentional.\\n\\n Second-degree murder also does not involve a\\n“substantial risk that force will be intentionally used during\\nits commission.” Gomez-Leon, 545 F.3d at 787. In\\nCovarrubias, we held that a California offense prohibiting\\nthe malicious and willful discharge of a firearm at an\\ninhabited dwelling was not a “crime of violence” because it\\ncould be committed recklessly, not just intentionally.\\nCovarrubias, 632 F.3d at 1053–55. Although we conducted\\nour analysis under § 16(b) because the BIA rested its\\ndecision on subsection (b), id. at 1052, the analysis regarding\\nintent bears upon either subsection of § 16, and by analogy,\\n18 U.S.C. § 924(c)(3). See, e.g., Gomez-Leon, 545 F.3d\\nat 787 (requiring intentional use of force for a crime of\\nviolence under either subsection of § 16); Benally, 843 F.3d\\nat 354. In contrast to crimes like burglary that can be\\ncommitted only intentionally, “with a crime committed\\nrecklessly, it is unlikely that the reckless actor will, in\\nresponse to external events, form an intent to use force in\\nfurtherance of his crime.” Covarrubias, 632 F.3d at 1055.\\n\\n14 UNITED STATES V. BEGAY\\n\\n“Classic examples of second-degree murder include\\nshooting a gun into a room that the defendant knows to be\\noccupied, a game of Russian roulette, and driving a car at\\nvery high speeds along a crowded main street . . . .” Pineda-\\nDoval, 614 F.3d at 1039. For purposes of our analysis, these\\nexamples are substantively indistinguishable from the\\noffense—“Shooting at an Inhabited Dwelling or Vehicle”—\\nthat we held was not categorically a crime of violence in\\nCovarrubias. This risk that a crime could escalate to the use\\nof intentional force is no more substantial for a defendant\\nwho recklessly kills than it is for a defendant who recklessly\\nshoots at a house.\\n\\n The cases the government cites do not squarely address\\nwhether second-degree murder is a crime of violence.\\nInstead, in those cases, we found that other challenges to\\n§ 924(c) convictions lacked merit. See, e.g., United States v.\\nPercy, 250 F.3d 720 (9th Cir. 2001) (addressing the Sixth\\nAmendment right to counsel and prosecutorial misconduct);\\nHouser, 130 F.3d 867 (discussing the role of the jury and the\\nmens rea required for second-degree murder). More\\nrecently, in United States v. JJ, we reviewed a district court's\\ndecision to try a juvenile as an adult under the Federal\\nJuvenile Delinquency Act, 18 U.S.C. § 5031 et seq. United\\nStates v. JJ, 704 F.3d 1219 (9th Cir. 2013). In JJ, the\\nquestion of whether “second-degree murder, if committed by\\nan adult, would be a felony crime of violence” was not in\\ndispute. Id. at 1222. Far from “holding” that second-degree\\nmurder is a “crime of violence” for purposes of § 924(c), JJ\\nfocused on when it might be in the interest of justice to try a\\njuvenile as an adult. Id. Because second-degree murder can\\n\\n UNITED STATES V. BEGAY 15\\n\\nbe committed recklessly, rather than intentionally, it does not\\ncategorically constitute a crime of violence. 2\\n\\n B\\n\\n Second-degree murder is not categorically a crime of\\nviolence under the elements clause, 18 U.S.C.\\n§ 924(c)(3)(A). And, pursuant to Davis, second-degree\\nmurder cannot constitute a crime of violence under the\\nresidual clause, section 924(c)(3)(B), as the residual clause\\nis unconstitutionally vague. Begay’s § 924(c) conviction for\\ndischarging a firearm during and in relation to a crime of\\nviolence therefore cannot stand under either the elements\\nclause or residual clause of § 924(c)(3). We reverse Count\\nTwo of his conviction.\\n\\n III\\n\\n We review the legality of a restitution order de novo and\\nthe factual findings supporting the order for clear error.\\nUnited States v. Luis, 765 F.3d 1061, 1065 (9th Cir. 2014)\\n(quoting United States v. Brock-Davis, 504 F.3d 991, 996\\n(9th Cir. 2007)). Where, as here, a defendant raises an\\nobjection to a restitution order for the first time on appeal,\\nwe review for plain error. United States v. Van Alystne, 584\\nF.3d 803, 819 (9th Cir. 2009).\\n\\n 18 U.S.C. § 3663A directs a district court to impose\\nmandatory restitution where a defendant is convicted of a\\ncrime of violence, as defined in 18 U.S.C. § 16. 18 U.S.C.\\n\\n 2\\n In holding that second-degree murder does not qualify as a\\ncategorical crime of violence because it can be committed recklessly, we\\ndo not reach Begay's argument that second-degree murder does not\\nqualify as a categorical crime of violence for the additional reason that it\\ncan be committed without the use of violent force.\\n\\n16 UNITED STATES V. BEGAY\\n\\n§ 3663A(c)(1)(A)(i). We conclude that the district court\\nplainly erred in imposing mandatory restitution under\\n18 U.S.C. § 3663A because second-degree murder is not\\ncategorically a crime of violence, and vacate the mandatory\\nrestitution order. In light of this disposition, we need not\\naddress Begay’s claim that the district court failed to make\\nthe necessary findings for the restitution order.\\n\\n CONCLUSION\\n\\n We AFFIRM Count One of Begay’s conviction for\\nsecond-degree murder. We REVERSE Count Two of\\nBegay’s conviction for discharging a firearm during a “crime\\nof violence” under 18 U.S.C. § 924(c)(1)(A), VACATE the\\nmandatory restitution order imposed under 18 U.S.C.\\n§ 3663A, and REMAND for resentencing.\\n\\n AFFIRMED IN PART, REVERSED IN PART,\\nVACATED IN PART, AND REMANDED.\\n\\n UNITED STATES V. BEGAY 17\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4080,"cells":{"id":{"kind":"number","value":4649428,"string":"4,649,428"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Joseph Shayota"},"decision_date":{"kind":"string","value":"2019-08-19"},"docket_number":{"kind":"string","value":"17-10270"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Mary M. Schroeder, Diarmuid F. O’Scannlain, and Johnnie B. Rawlinson, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"Opinion by Judge O'Scannlain\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10270\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 5:15-cr-00264-LHK-1\\n\\nJOSEPH SHAYOTA,\\n Defendant-Appellant.\\n\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10271\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 5:15-cr-00264-LHK-2\\n\\nADRIANA SHAYOTA,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n Lucy H. Koh, District Judge, Presiding\\n\\n Argued and Submitted February 14, 2019\\n San Francisco, California\\n\\n Filed August 19, 2019\\n\\n2 UNITED STATES V. SHAYOTA\\n\\n Before: Mary M. Schroeder, Diarmuid F. O’Scannlain,\\n and Johnnie B. Rawlinson, Circuit Judges.\\n\\n Opinion by Judge O’Scannlain;\\n Concurrence by Judge O’Scannlain\\n\\n\\n SUMMARY*\\n\\n\\n Criminal Law\\n\\n The panel affirmed the district court’s judgment in a case\\nin which the panel was asked to decide whether prior civil\\ndeposition testimony of a witness, who has subsequently\\ninvoked his Fifth Amendment right against self-\\nincrimination, may be introduced against defendants in a\\ncriminal trial without violating their Confrontation Clause\\nright to confront the witnesses against them.\\n\\n At the defendants’ trial, the district court admitted the\\ncivil deposition testimony of two individuals after they\\ninvoked their Fifth Amendment privilege not to testify. The\\ndistrict court concluded that their invocation of their Fifth\\nAmendment privilege rendered them unavailable for purposes\\nof the Confrontation Clause. The defendants argued that the\\ngovernment’s inherent discretion to grant a witness immunity\\nand thereby prevent him from invoking the Fifth Amendment\\nprivilege renders the witness effectively available to the\\ngovernment for testimony at trial. The panel did not need to\\nresolve that issue because even if the district court erred by\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. SHAYOTA 3\\n\\nconcluding that the witnesses were unavailable, the error was\\nharmless because the outcome of the trial would not have\\nchanged had the depositions been excluded.\\n\\n The panel addressed other arguments in a memorandum\\ndisposition.\\n\\n Specially concurring, Judge O’Scannlain wrote separately\\nto call attention to this court’s precedent regarding the\\n“unavailability” requirement of the Confrontation Clause.\\nObserving that the right of confrontation is “most naturally\\nread as a reference to the right of confrontation at common\\nlaw, admitting only those exceptions established at the time\\nof the founding,” Crawford v. Washington, 541 U.S. 36, 54\\n(2004), Judge O’Scannlain suggested that this court revisit\\nits prior decisions to perform the historical analysis that\\nCrawford demands.\\n\\n\\n COUNSEL\\n\\nTheodore Sampsell-Jones (argued), Sampsell-Jones Law,\\nMinneapolis, Minnesota; Dennis P. Riordan and Donald M.\\nHorgan, Riordan & Horgan, San Francisco, California; for\\nDefendant-Appellant Joseph Shayota.\\n\\nJohn D. Cline (argued), Law Office of John D. Cline, San\\nFrancisco, California, for Defendant-Appellant Adriana\\nShayota.\\n\\nJonas Lerman (argued), Assistant United States Attorney; J.\\nDouglas Wilson, Chief, Appellate Division; United States\\nAttorney’s Office, San Francisco, California; for Plaintiff-\\nAppellee.\\n\\n4 UNITED STATES V. SHAYOTA\\n\\n OPINION\\n\\nO’SCANNLAIN, Circuit Judge:\\n\\n We are asked to decide whether prior civil deposition\\ntestimony of a witness, who has subsequently invoked his\\nFifth Amendment right against self-incrimination, may be\\nintroduced against defendants in a criminal trial without\\nviolating their Confrontation Clause right to confront the\\nwitnesses against them.\\n\\n I\\n\\n At the time of the events giving rise to this case,\\ndefendants Joseph Shayota and his wife Adriana Shayota ran\\nBaja Exporting, LLC (“Baja Exporting”)—a California\\ncompany that imported, exported, and distributed snacks and\\ndrinks to gas stations, convenience stores, and bodegas, both\\nin the United States and abroad. In 2009, Baja Exporting\\ncontracted with Living Essentials, LLC and its related entities\\n(collectively “Living Essentials”) to sell the liquid dietary\\nsupplement 5-Hour Energy in Mexico.\\n\\n Unfortunately, the product did not sell well. The parties\\nterminated their sales agreement in 2010, and Baja Exporting\\nwas left with excess bottles of Spanish-labeled 5-Hour\\nEnergy. Rather than dispose of the bottles, Baja Exporting\\ntried to sell them in the United States, despite not having\\napproval from Living Essentials to do so. When the product\\nagain failed to sell, the Shayotas and their associates\\nrelabeled the bottles in English and sold them without\\nauthorization from Living Essentials.\\n\\n UNITED STATES V. SHAYOTA 5\\n\\n By December 2011, Baja Exporting had sold the last of its\\n5-Hour Energy supply. Given the potential market in the\\nUnited States, the Shayotas and their associates then\\nconspired to create a counterfeit version of the drink, pass it\\noff as genuine 5-Hour Energy, and distribute it across the\\ncountry.\\n\\n A\\n\\n The scheme lasted from early 2012 to November 2012\\nand involved several key players. Joseph Shayota oversaw\\nand financed the operation. Adriana Shayota was in charge\\nof the accounting; she collected invoices for the counterfeit\\nproduct and made wire transfers to cover costs. Walid Jamil\\n(Joseph’s brother-in-law) and Justin Shayota (Joseph and\\nAdriana’s nephew) coordinated the repackaging and\\nrelabeling of the counterfeit drink at Baja Exporting’s San\\nDiego warehouse. They worked closely with Leslie Roman,\\nwho supplied blank bottles and manufactured counterfeit 5-\\nHour Energy labels. Jamil hired others to create the\\ncounterfeit drink and to manufacture boxes for the finished\\nproduct. Finally, Justin sent the finished product either to\\nBaja Exporting or to Dan Dee Company—a cash-and-carry\\nwarehouse owned by Kevin Attiq—for distribution across the\\ncountry.\\n\\n Living Essentials became aware of the scheme in 2012,\\nafter it noticed a mysterious drop-off in its California sales.\\nLiving Essentials hired private investigators, who ultimately\\nraided warehouses controlled by Baja Exporting and Jamil\\nand discovered boxes of counterfeit 5-Hour Energy.\\nConsequently, Living Essentials sued the Shayotas, Jamil,\\nand others, alleging numerous violations of laws pertaining\\nto trademark infringement and false advertising.\\n\\n6 UNITED STATES V. SHAYOTA\\n\\n During discovery, Living Essentials deposed numerous\\nparticipants in the scheme, including Jamil and Roman.\\nBecause the Shayotas were parties to the suit, their counsel\\nattended the depositions and questioned the witnesses. The\\ncivil suit ultimately ended in a settlement in which Baja\\nExporting agreed to pay $6 million to Living Essentials.\\n\\n B\\n\\n Subsequent to the civil proceedings, the government\\nlaunched a criminal investigation into the scheme. A grand\\njury returned an indictment against the Shayotas, Jamil, and\\nRoman, among others. The government ultimately filed a\\ntwo-count Superseding Information charging each defendant\\nwith (1) conspiracy to traffic in counterfeit goods, in violation\\nof 18 U.S.C. § 2320(a), and (2) conspiracy to commit\\ncopyright infringement and to introduce misbranded food\\ninto interstate commerce, in violation of 17 U.S.C. § 506,\\n18 U.S.C. § 2319, and 21 U.S.C. §§ 331 and 333. Jamil and\\nRoman pleaded guilty; the Shayotas proceeded to trial.\\n\\n As trial approached, the government notified defense\\ncounsel of its intent to offer into evidence statements made by\\nJamil and Roman during their civil depositions. The\\nShayotas moved to exclude the testimony on the ground that\\nits admission would violate the Sixth Amendment’s\\nConfrontation Clause if Jamil and Roman refused to testify,\\nbut the district court denied the motion. The depositions were\\nintroduced at trial after Jamil and Roman invoked their Fifth\\nAmendment privilege not to testify, and the jury ultimately\\nfound the Shayotas guilty of the charged offenses.\\n\\n The Shayotas timely appealed their convictions.\\n\\n UNITED STATES V. SHAYOTA 7\\n\\n II\\n\\n The Shayotas argue that admission of Jamil’s and\\nRoman’s deposition testimony violated the Sixth\\nAmendment’s Confrontation Clause.1\\n\\n The Sixth Amendment to the United States Constitution\\nprovides in pertinent part: “[I]n all criminal prosecutions the\\naccused shall enjoy the right . . . to be confronted with the\\nwitnesses against him.” U.S. Const. amend. VI. Generally,\\nthis means that the government must produce at trial any\\nwitnesses who have offered testimonial evidence against the\\naccused. But where, as here, the prosecution chooses not to\\nproduce such witnesses and offers their hearsay statements\\ninstead, a defendant’s right of confrontation is satisfied if\\n(1) the declarant is unavailable, and (2) the defendant had a\\nprior opportunity to confront him through cross-examination.\\nCrawford v. Washington, 541 U.S. 36, 68 (2004). The district\\ncourt concluded that Jamil and Roman were unavailable\\nbecause they had validly invoked their Fifth Amendment\\nprivilege against self-incrimination. The court also ruled that\\nthe Shayotas had a prior opportunity to cross-examine Jamil\\nand Roman in the civil suit, given that the Shayotas sent their\\nlawyers to attend Jamil’s and Roman’s depositions in that\\ncase.\\n\\n The Shayotas dispute the district court’s legal conclusion\\nthat Jamil and Roman were unavailable. They argue that\\nJamil and Roman were in fact available because the\\n\\n\\n 1\\n We address the Shayotas’ other arguments on appeal in a\\nmemorandum disposition issued concurrently with this opinion. See\\nUnited States v. Shayota, Nos. 17-10270 & 17-10271, — F. App’x — (9th\\nCir. 2019).\\n\\n8 UNITED STATES V. SHAYOTA\\n\\ngovernment could have granted them immunity and\\ncompelled them to testify.\\n\\n We have held on numerous occasions that a witness’s\\nassertion of Fifth Amendment privilege renders him\\n“unavailable” for purposes of the Confrontation Clause. See,\\ne.g., United States v. Wilmore, 381 F.3d 868, 872 (9th Cir.\\n2004); see also California v. Green, 399 U.S. 149, 167–68\\n(1970) (explaining in dicta that once a witness has been\\nproduced at trial and “claimed his privilege against\\ncompulsory self-incrimination, . . . nothing in the\\nConfrontation Clause prohibit[s] the State from . . . relying on\\nhis prior testimony to prove its case”). But we have yet to\\nconfront the Shayotas’ precise argument here—that the\\ngovernment’s inherent discretion to grant a witness immunity\\nand thereby prevent him from invoking the privilege against\\nself-incrimination renders the witness effectively available to\\nthe government for testimony at trial. Nevertheless, we need\\nnot resolve the issue here: even if the district court erred by\\nconcluding that Jamil and Roman were unavailable, the error\\nwas harmless.\\n\\n “[C]onfrontation [C]lause violations are subject to\\nharmless error analysis.” United States v. Bernard S.,\\n795 F.2d 749, 756 (9th Cir. 1986). “Whether a violation of\\nthe [C]onfrontation [C]lause is harmless depends on a variety\\nof factors including: (1) the importance of the evidence to the\\nprosecution’s case; (2) whether the evidence was cumulative;\\n(3) the presence of corroborating evidence; (4) the overall\\nstrength of the prosecution’s case.” Id. To be sure, the\\ntestimony of Jamil and Roman was an important component\\nof the trial. But the testimony was also corroborated by live\\nwitness testimony, including that of Kevin Attiq and Justin\\nShayota who described the Shayotas’ involvement in the\\n\\n UNITED STATES V. SHAYOTA 9\\n\\nscheme, and a bounty of other circumstantial evidence\\nrelating to the Shayotas’ management of the counterfeit\\nproduct and exchange of invoices with other members of the\\nconspiracy. Considering the other evidence presented at trial,\\nthe prosecution’s case against the Shayotas was strong. We\\nconclude that the outcome of the trial would not have\\nchanged had the depositions been excluded.2\\n\\n III\\n\\n The judgment of the district court is AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426681/\", \"author_raw\": \"Opinion by Judge O'Scannlain\"}, {\"author\": \"O’SCANNLAIN, Circuit Judge, specially concurring\", \"type\": \"concurrence\", \"text\": \"O’SCANNLAIN, Circuit Judge, specially concurring:\\n\\n I write separately to call attention to our court’s precedent\\nregarding the “unavailability” requirement of the Sixth\\nAmendment’s Confrontation Clause. We have held that a\\nwitness is unavailable for purposes of the Confrontation\\nClause if he invokes his Fifth Amendment privilege and\\nrefuses to testify. See, e.g., United States v. Wilmore,\\n381 F.3d 868, 872 (9th Cir. 2004). Presumably we have done\\nso because such a witness is considered unavailable under the\\nformer testimony exception to the federal hearsay rule. See\\nid. (citing Fed. R. Evid. 804(b)(1)). But the right of\\nconfrontation is “most naturally read as a reference to the\\nright of confrontation at common law, admitting only those\\n\\n\\n 2\\n We also reject on harmless error grounds the Shayotas’ argument\\nthat former testimony cannot be admitted unless the parties and issues in\\nthe prior civil proceeding were identical to those at the subsequent\\ncriminal trial and the defendants were physically present during the prior\\ntestimony.\\n\\n10 UNITED STATES V. SHAYOTA\\n\\nexceptions established at the time of the founding.”\\nCrawford v. Washington, 541 U.S. 36, 54 (2004). Thus, a\\ndetermination of whether the former testimony of a witness\\nwho has since invoked his Fifth Amendment privilege is\\nadmissible necessarily requires an analysis of whether such\\ntestimony would have been admitted at the time of the\\nfounding. Our court has yet to do that work, but history\\nsuggests that the scope of unavailability may be narrower\\nthan our court has recognized. I respectfully suggest that we\\nshould revisit our prior decisions to perform the historical\\nanalysis that Crawford demands.\\n\\n I\\n\\n To understand how the Confrontation Clause applies, we\\nmust first understand its history. I will not duplicate the\\nefforts of courts and commentators who have thoroughly\\nrecounted the development of the right of confrontation from\\nsixteenth-century England to the founding. See, e.g.,\\nCrawford, 541 U.S. at 42–49; Daniel H. Pollitt, The Right of\\nConfrontation: Its History and Modern Dress, 8 J. Pub. L.\\n386 (1959). It is enough to say that the common law right\\nemerged in England in the mid-1600s as a reaction to the\\nwidespread use of ex parte affidavits and depositions instead\\nof face-to-face examination in criminal trials against the\\naccused. See Crawford, 541 U.S. at 44–45. To prevent such\\nabuses from taking root in the United States, the First\\nCongress in 1789 included “the Confrontation Clause in the\\nproposal that became the Sixth Amendment.” Id. at 49; see\\nalso Mattox v. United States, 156 U.S. 237, 242 (1895)\\n(recognizing that the primary object of the Clause was to\\n“prevent depositions or ex parte affidavits, such as were\\nsometimes admitted in civil cases, [from] being used against\\nthe prisoner in lieu of a personal examination and cross-\\n\\n UNITED STATES V. SHAYOTA 11\\n\\nexamination of the witness”). Thus, “[t]he right of\\nconfrontation did not originate with the provision in the Sixth\\nAmendment, but was a common-law right having recognized\\nexceptions.” Salinger v. United States, 272 U.S. 542, 548\\n(1926); accord 3 Joseph Story, Commentaries on the\\nConstitution of the United States 662 (1st ed. 1833).\\n\\n In England, such exceptions relied on a “relatively strict\\nrule[] of unavailability.” Crawford, 541 U.S. at 44; see also\\nTim Donaldson, Gradually Exploded: Confrontation vs. The\\nFormer Testimony Rule, 46 St. Mary’s L.J. 137, 156 (2015)\\n(collecting cases). Lord Morley’s Case, 6 How. St. Tr. 769\\n(H.L. 1666), for example, explained that a coroner’s out-of-\\ncourt examinations of witnesses were admissible if “the\\nwitnesses . . . were dead or unable to travel,” or if they were\\n“absent . . . by the means or procurement of the prisoner.” Id.\\nat 770–71. Such a rule persisted in England well into the\\n1700s. See 2 William Hawkins, A Treatise of the Pleas of the\\nCrown 429 (1721) (explaining that depositions “may be given\\nin evidence . . . if it be made out by Oath to the Satisfaction\\nof the Court, that such Informer is dead, or unable to travel,\\nor kept away by the Means or Procurement of the Prisoner”).\\nOther sources of English law suggest that depositions could\\nalso be admitted if the witness was beyond the jurisdictional\\nreach of the court. See Henry Bathurst, The Theory of\\nEvidence 34 (1761) (“[I]f the Witnesses examined on a\\nCoroner’s Inquest are . . . beyond Sea, their Depositions may\\nbe read.”).\\n\\n The American tradition similarly permitted the use of\\ndeposition testimony, at least when the witness was dead and\\nperhaps also when he was missing or outside of the court’s\\njurisdiction. See Cline v. State, 36 Tex. Crim. 320, 360–61\\n(Tex. Crim. App. 1896) (Henderson, J., dissenting)\\n\\n12 UNITED STATES V. SHAYOTA\\n\\n(collecting cases); see also Johnston v. State, 10 Tenn. 58,\\n59–60 (Err. & App. 1821) (holding that “depositions . . . may\\nbe read on trial against a prisoner” when the witness is proven\\ndead); State v. Atkins, 1 Tenn. 229, 229 (Super. L. & Eq.\\n1807) (per curiam) (same). But see United States v. Angell,\\n11 F. 34, 43 (C.C.D. N.H. 1881) (“I have found no case\\nwhere the testimony of a witness, absent but living, given at\\na former trial, has been allowed to be proved at a subsequent\\ntrial.”); People v. Newman, 5 Hill 295, 296 (N.Y. Sup. Ct.\\n1843) (“[N]othing short of the witness’[s] death can be\\nreceived to let in his testimony given on a former trial.”).\\n\\n Notably absent from founding-era case law,1 however, are\\ndecisions permitting the adverse use of former testimony of\\nwitnesses who claimed a testimonial privilege—even though\\nsuch privileges existed at common law, see, e.g., Quinn v.\\nUnited States, 349 U.S. 155, 161 (1955) (explaining that the\\nprivilege against self-incrimination was “firmly established\\n. . . in the common law of England” by the mid-1600s and\\n“[t]ransplanted to this country as part of our legal heritage”).\\nThe absence of such cases suggests that there may not have\\nbeen a common-law exception to the right of confrontation\\nfor former testimony of witnesses who claimed a privilege\\nagainst self-incrimination.\\n\\n II\\n\\n\\n\\n 1\\n There are cases from the early 1900s which admit former testimony\\nby a witness who claims a privilege. But in addition to being farther in\\ntime from the founding, these cases do not discuss the right of\\nconfrontation at all. See, e.g., State v. Stewart, 85 Kan. 404 (1911). Such\\ncases therefore have little—if any—bearing on the question of\\nunavailability under the Confrontation Clause at the time of the founding.\\n\\n UNITED STATES V. SHAYOTA 13\\n\\n The Supreme Court has yet to consider the historical\\nlimitations of the unavailability requirement. Rather, in\\ndecisions that predate Crawford, the Court has provided\\nconflicting guidance on the scope of the unavailability\\nrequirement. On the one hand, in Douglas v. Alabama,\\n380 U.S. 415 (1965), the Court held that the defendant’s right\\nof confrontation was violated by the admission of a\\nconfession that implicated the defendant and that was given\\nby a witness who invoked the Fifth Amendment. Id. at 420.\\n\\n By contrast, in California v. Green, the Court observed\\nthat the Confrontation Clause would not be violated if a\\nwitness “claimed his privilege against compulsory self-\\nincrimination.” 399 U.S. 149, 167 (1970); accord Lee v.\\nIllinois, 476 U.S. 530, 549–50 (1986) (Blackmun, J.,\\ndissenting) (concluding that a witness was unavailable\\nbecause he would have invoked his Fifth Amendment\\nprivilege). And, most recently, in Lilly v. Virginia, 527 U.S.\\n116 (1999), a plurality of the Court “assumed” without\\ndeciding that a witness is “unavailable . . . for Confrontation\\nClause purposes” when he invokes his Fifth Amendment right\\nagainst self-incrimination. Id. at 124 n.1. The need for such\\nan assumption suggests that the question of unavailability is\\nfar from settled and requires further examination by lower\\ncourts.\\n\\n III\\n\\n Rather than consider the common law, our court has\\nrelied on the former testimony exception to the hearsay\\nrule—Federal Rule of Evidence 804—to define the contours\\nof unavailability under the Confrontation Clause. See\\nWilmore, 381 F.3d at 872 (citing Fed. R. Evid. 804(a) and\\n\\n14 UNITED STATES V. SHAYOTA\\n\\nrelying only on circuit precedent that predates Crawford).2 In\\nmy view, such reasoning is problematic.\\n\\n Crawford all but rejected the view that the Confrontation\\nClause is satisfied so long as the evidence fits neatly within\\na federal hearsay exception. As the Court explained, the\\nConfrontation Clause’s application does not “depend[] upon\\nthe law of Evidence.” 541 U.S. at 50; see also id. at 51\\n(acknowledging that “ex parte examinations might sometimes\\nbe admissible under modern hearsay rules, but the Framers\\ncertainly would not have condoned them” (emphasis added)).\\nA contrary conclusion would improperly conflate the\\nrequirements of the Confrontation Clause and the hearsay rule\\nwhen the two often serve distinct functions in the presentation\\nof evidence and the preservation of rights. See id. at 51\\n(“Leaving the regulation of out-of-court statements to the law\\nof evidence would render the Confrontation Clause powerless\\nto prevent even the most flagrant inquisitorial practices.”).\\nWe must, therefore, do more than look only to the\\nrequirements of Rule 804 when deciding whether the\\nadmission of certain testimony violates a defendant’s right of\\nconfrontation.\\n\\n Relying on dicta in United States v. Yida, 498 F.3d 945,\\n950 (9th Cir. 2007), however, the government argues that\\n“Rule 804(b)(1) implements the command of the . . .\\n\\nConfrontation Clause,” and that satisfying Rule 804 therefore\\nsatisfies the right of confrontation. But Yida acknowledges\\nnothing more than the obvious: Rule 804 incorporates the\\n\\n\\n 2\\n Federal Rule of Evidence 804 permits the introduction of former\\ntestimony if the witness is “unavailable” and defines unavailability to\\ninclude a valid assertion of privilege. See Fed. R. Evid. 804(a)(1), (b)(1).\\n\\n UNITED STATES V. SHAYOTA 15\\n\\nConstitution’s general requirements of unavailability and a\\nprior opportunity for cross-examination. Yida expressed no\\nview on whether Rule 804(a)’s enumerated classes of\\nunavailable witnesses echo those existing at common law.\\n\\n Indeed, there are sound reasons to conclude that\\nunavailability for purposes of the Confrontation Clause ought\\nnot to include the assertion of a testimonial privilege.\\nConsidering the common law sources described above in part\\nI, the scope of unavailability at common law appeared to be\\ndefined by circumstances outside the prosecution’s\\ncontrol—the death of the witness, for example, or the\\nprosecution’s inability to procure his attendance due to\\njurisdictional limitations. Yet the government is not so\\npowerless to prevent a claim of privilege, as it can grant the\\nwitness immunity or enter into a plea agreement with the\\nwitness to provide for his testimony. In either case, the\\nwitness would be unable to invoke the Fifth Amendment and\\ncould then be made to testify. See Kastigar v. United States,\\n406 U.S. 441, 458 (1972) (“[U]se and derivative-use\\nimmunity is constitutionally sufficient to compel testimony\\nover a claim of the privilege.”); 5 Wayne R. LaFave,\\nCriminal Procedure § 21.2(e), at 60 (2d ed. Supp. 2007) (“[I]t\\nis generally accepted that ‘when a defendant breaches his plea\\nagreement, the Government has the option to either seek\\nspecific performance of the agreement or treat it as\\nunenforceable’ (at least absent language in the plea agreement\\nspecifying fewer or other remedies).”)\\n\\n To be sure, such choices “entail significant costs.” Lee,\\n476 U.S. at 550 (Blackmun, J., dissenting). “A plea\\nagreement necessarily compromises the community’s\\nlegitimate correctional interests, and a grant of immunity\\nplaces a heavy evidentiary burden on any future prosecution\\n\\n16 UNITED STATES V. SHAYOTA\\n\\nof the witness.” Id. But nothing entitles a prosecutor to\\nintroduce any evidence he wishes; the Constitution, by\\ncontrast, entitles the defendant to confront the witnesses\\nagainst him unless a well-recognized exception to the right of\\nconfrontation applies. That there may be significant costs\\nshould not lead us to ignore Sixth Amendment protections.\\nCf. United States v. Stevens, 559 U.S. 460, 470 (2010) (“The\\nFirst Amendment’s guarantee of free speech does not extend\\nonly to categories of speech that survive an ad hoc balancing\\nof relative social costs and benefits. The First Amendment\\nitself reflects a judgment by the American people that the\\nbenefits of its restrictions on the Government outweigh the\\ncosts.”).\\n\\n IV\\n\\n The district court’s decision to admit the testimony of\\nJamil and Roman finds support in our case law. But “the\\nSixth Amendment demands what the common law required.”\\nCrawford, 541 U.S. at 68. In future cases, I hope that our\\ncourt, inspired by Crawford, might reconsider the scope of\\nthe unavailability requirement in light of the common law.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426681/\", \"author_raw\": \"O’SCANNLAIN, Circuit Judge, specially concurring\"}]}"},"judge1":{"kind":"string","value":"MARY M SCHROEDER"},"judge2":{"kind":"string","value":"DIARMUID F O'SCANNLAIN"},"judge3":{"kind":"string","value":"JOHNNIE B RAWLINSON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":1,"string":"1"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649428/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Joseph Shayota\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-19\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: Yes\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-10270\n Plaintiff-Appellee,\n D.C. No.\n v. 5:15-cr-00264-LHK-1\n\nJOSEPH SHAYOTA,\n Defendant-Appellant.\n\n\n\nUNITED STATES OF AMERICA, No. 17-10271\n Plaintiff-Appellee,\n D.C. No.\n v. 5:15-cr-00264-LHK-2\n\nADRIANA SHAYOTA,\n Defendant-Appellant. OPINION\n\n\n Appeal from the United States District Court\n for the Northern District of California\n Lucy H. Koh, District Judge, Presiding\n\n Argued and Submitted February 14, 2019\n San Francisco, California\n\n Filed August 19, 2019\n\n2 UNITED STATES V. SHAYOTA\n\n Before: Mary M. Schroeder, Diarmuid F. O’Scannlain,\n and Johnnie B. Rawlinson, Circuit Judges.\n\n Opinion by Judge O’Scannlain;\n Concurrence by Judge O’Scannlain\n\n\n SUMMARY*\n\n\n Criminal Law\n\n The panel affirmed the district court’s judgment in a case\nin which the panel was asked to decide whether prior civil\ndeposition testimony of a witness, who has subsequently\ninvoked his Fifth Amendment right against self-\nincrimination, may be introduced against defendants in a\ncriminal trial without violating their Confrontation Clause\nright to confront the witnesses against them.\n\n At the defendants’ trial, the district court admitted the\ncivil deposition testimony of two individuals after they\ninvoked their Fifth Amendment privilege not to testify. The\ndistrict court concluded that their invocation of their Fifth\nAmendment privilege rendered them unavailable for purposes\nof the Confrontation Clause. The defendants argued that the\ngovernment’s inherent discretion to grant a witness immunity\nand thereby prevent him from invoking the Fifth Amendment\nprivilege renders the witness effectively available to the\ngovernment for testimony at trial. The panel did not need to\nresolve that issue because even if the district court erred by\n\n *\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\n UNITED STATES V. SHAYOTA 3\n\nconcluding that the witnesses were unavailable, the error was\nharmless because the outcome of the trial would not have\nchanged had the depositions been excluded.\n\n The panel addressed other arguments in a memorandum\ndisposition.\n\n Specially concurring, Judge O’Scannlain wrote separately\nto call attention to this court’s precedent regarding the\n“unavailability” requirement of the Confrontation Clause.\nObserving that the right of confrontation is “most naturally\nread as a reference to the right of confrontation at common\nlaw, admitting only those exceptions established at the time\nof the founding,” Crawford v. Washington, 541 U.S. 36, 54\n(2004), Judge O’Scannlain suggested that this court revisit\nits prior decisions to perform the historical analysis that\nCrawford demands.\n\n\n COUNSEL\n\nTheodore Sampsell-Jones (argued), Sampsell-Jones Law,\nMinneapolis, Minnesota; Dennis P. Riordan and Donald M.\nHorgan, Riordan & Horgan, San Francisco, California; for\nDefendant-Appellant Joseph Shayota.\n\nJohn D. Cline (argued), Law Office of John D. Cline, San\nFrancisco, California, for Defendant-Appellant Adriana\nShayota.\n\nJonas Lerman (argued), Assistant United States Attorney; J.\nDouglas Wilson, Chief, Appellate Division; United States\nAttorney’s Office, San Francisco, California; for Plaintiff-\nAppellee.\n\n4 UNITED STATES V. SHAYOTA\n\n OPINION\n\nO’SCANNLAIN, Circuit Judge:\n\n We are asked to decide whether prior civil deposition\ntestimony of a witness, who has subsequently invoked his\nFifth Amendment right against self-incrimination, may be\nintroduced against defendants in a criminal trial without\nviolating their Confrontation Clause right to confront the\nwitnesses against them.\n\n I\n\n At the time of the events giving rise to this case,\ndefendants Joseph Shayota and his wife Adriana Shayota ran\nBaja Exporting, LLC (“Baja Exporting”)—a California\ncompany that imported, exported, and distributed snacks and\ndrinks to gas stations, convenience stores, and bodegas, both\nin the United States and abroad. In 2009, Baja Exporting\ncontracted with Living Essentials, LLC and its related entities\n(collectively “Living Essentials”) to sell the liquid dietary\nsupplement 5-Hour Energy in Mexico.\n\n Unfortunately, the product did not sell well. The parties\nterminated their sales agreement in 2010, and Baja Exporting\nwas left with excess bottles of Spanish-labeled 5-Hour\nEnergy. Rather than dispose of the bottles, Baja Exporting\ntried to sell them in the United States, despite not having\napproval from Living Essentials to do so. When the product\nagain failed to sell, the Shayotas and their associates\nrelabeled the bottles in English and sold them without\nauthorization from Living Essentials.\n\n UNITED STATES V. SHAYOTA 5\n\n By December 2011, Baja Exporting had sold the last of its\n5-Hour Energy supply. Given the potential market in the\nUnited States, the Shayotas and their associates then\nconspired to create a counterfeit version of the drink, pass it\noff as genuine 5-Hour Energy, and distribute it across the\ncountry.\n\n A\n\n The scheme lasted from early 2012 to November 2012\nand involved several key players. Joseph Shayota oversaw\nand financed the operation. Adriana Shayota was in charge\nof the accounting; she collected invoices for the counterfeit\nproduct and made wire transfers to cover costs. Walid Jamil\n(Joseph’s brother-in-law) and Justin Shayota (Joseph and\nAdriana’s nephew) coordinated the repackaging and\nrelabeling of the counterfeit drink at Baja Exporting’s San\nDiego warehouse. They worked closely with Leslie Roman,\nwho supplied blank bottles and manufactured counterfeit 5-\nHour Energy labels. Jamil hired others to create the\ncounterfeit drink and to manufacture boxes for the finished\nproduct. Finally, Justin sent the finished product either to\nBaja Exporting or to Dan Dee Company—a cash-and-carry\nwarehouse owned by Kevin Attiq—for distribution across the\ncountry.\n\n Living Essentials became aware of the scheme in 2012,\nafter it noticed a mysterious drop-off in its California sales.\nLiving Essentials hired private investigators, who ultimately\nraided warehouses controlled by Baja Exporting and Jamil\nand discovered boxes of counterfeit 5-Hour Energy.\nConsequently, Living Essentials sued the Shayotas, Jamil,\nand others, alleging numerous violations of laws pertaining\nto trademark infringement and false advertising.\n\n6 UNITED STATES V. SHAYOTA\n\n During discovery, Living Essentials deposed numerous\nparticipants in the scheme, including Jamil and Roman.\nBecause the Shayotas were parties to the suit, their counsel\nattended the depositions and questioned the witnesses. The\ncivil suit ultimately ended in a settlement in which Baja\nExporting agreed to pay $6 million to Living Essentials.\n\n B\n\n Subsequent to the civil proceedings, the government\nlaunched a criminal investigation into the scheme. A grand\njury returned an indictment against the Shayotas, Jamil, and\nRoman, among others. The government ultimately filed a\ntwo-count Superseding Information charging each defendant\nwith (1) conspiracy to traffic in counterfeit goods, in violation\nof 18 U.S.C. § 2320(a), and (2) conspiracy to commit\ncopyright infringement and to introduce misbranded food\ninto interstate commerce, in violation of 17 U.S.C. § 506,\n18 U.S.C. § 2319, and 21 U.S.C. §§ 331 and 333. Jamil and\nRoman pleaded guilty; the Shayotas proceeded to trial.\n\n As trial approached, the government notified defense\ncounsel of its intent to offer into evidence statements made by\nJamil and Roman during their civil depositions. The\nShayotas moved to exclude the testimony on the ground that\nits admission would violate the Sixth Amendment’s\nConfrontation Clause if Jamil and Roman refused to testify,\nbut the district court denied the motion. The depositions were\nintroduced at trial after Jamil and Roman invoked their Fifth\nAmendment privilege not to testify, and the jury ultimately\nfound the Shayotas guilty of the charged offenses.\n\n The Shayotas timely appealed their convictions.\n\n UNITED STATES V. SHAYOTA 7\n\n II\n\n The Shayotas argue that admission of Jamil’s and\nRoman’s deposition testimony violated the Sixth\nAmendment’s Confrontation Clause.1\n\n The Sixth Amendment to the United States Constitution\nprovides in pertinent part: “[I]n all criminal prosecutions the\naccused shall enjoy the right . . . to be confronted with the\nwitnesses against him.” U.S. Const. amend. VI. Generally,\nthis means that the government must produce at trial any\nwitnesses who have offered testimonial evidence against the\naccused. But where, as here, the prosecution chooses not to\nproduce such witnesses and offers their hearsay statements\ninstead, a defendant’s right of confrontation is satisfied if\n(1) the declarant is unavailable, and (2) the defendant had a\nprior opportunity to confront him through cross-examination.\nCrawford v. Washington, 541 U.S. 36, 68 (2004). The district\ncourt concluded that Jamil and Roman were unavailable\nbecause they had validly invoked their Fifth Amendment\nprivilege against self-incrimination. The court also ruled that\nthe Shayotas had a prior opportunity to cross-examine Jamil\nand Roman in the civil suit, given that the Shayotas sent their\nlawyers to attend Jamil’s and Roman’s depositions in that\ncase.\n\n The Shayotas dispute the district court’s legal conclusion\nthat Jamil and Roman were unavailable. They argue that\nJamil and Roman were in fact available because the\n\n\n 1\n We address the Shayotas’ other arguments on appeal in a\nmemorandum disposition issued concurrently with this opinion. See\nUnited States v. Shayota, Nos. 17-10270 & 17-10271, — F. App’x — (9th\nCir. 2019).\n\n8 UNITED STATES V. SHAYOTA\n\ngovernment could have granted them immunity and\ncompelled them to testify.\n\n We have held on numerous occasions that a witness’s\nassertion of Fifth Amendment privilege renders him\n“unavailable” for purposes of the Confrontation Clause. See,\ne.g., United States v. Wilmore, 381 F.3d 868, 872 (9th Cir.\n2004); see also California v. Green, 399 U.S. 149, 167–68\n(1970) (explaining in dicta that once a witness has been\nproduced at trial and “claimed his privilege against\ncompulsory self-incrimination, . . . nothing in the\nConfrontation Clause prohibit[s] the State from . . . relying on\nhis prior testimony to prove its case”). But we have yet to\nconfront the Shayotas’ precise argument here—that the\ngovernment’s inherent discretion to grant a witness immunity\nand thereby prevent him from invoking the privilege against\nself-incrimination renders the witness effectively available to\nthe government for testimony at trial. Nevertheless, we need\nnot resolve the issue here: even if the district court erred by\nconcluding that Jamil and Roman were unavailable, the error\nwas harmless.\n\n “[C]onfrontation [C]lause violations are subject to\nharmless error analysis.” United States v. Bernard S.,\n795 F.2d 749, 756 (9th Cir. 1986). “Whether a violation of\nthe [C]onfrontation [C]lause is harmless depends on a variety\nof factors including: (1) the importance of the evidence to the\nprosecution’s case; (2) whether the evidence was cumulative;\n(3) the presence of corroborating evidence; (4) the overall\nstrength of the prosecution’s case.” Id. To be sure, the\ntestimony of Jamil and Roman was an important component\nof the trial. But the testimony was also corroborated by live\nwitness testimony, including that of Kevin Attiq and Justin\nShayota who described the Shayotas’ involvement in the\n\n UNITED STATES V. SHAYOTA 9\n\nscheme, and a bounty of other circumstantial evidence\nrelating to the Shayotas’ management of the counterfeit\nproduct and exchange of invoices with other members of the\nconspiracy. Considering the other evidence presented at trial,\nthe prosecution’s case against the Shayotas was strong. We\nconclude that the outcome of the trial would not have\nchanged had the depositions been excluded.2\n\n III\n\n The judgment of the district court is AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Joseph Shayota\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-19\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: Yes\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10270\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 5:15-cr-00264-LHK-1\\n\\nJOSEPH SHAYOTA,\\n Defendant-Appellant.\\n\\n\\n\\nUNITED STATES OF AMERICA, No. 17-10271\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 5:15-cr-00264-LHK-2\\n\\nADRIANA SHAYOTA,\\n Defendant-Appellant. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n Lucy H. Koh, District Judge, Presiding\\n\\n Argued and Submitted February 14, 2019\\n San Francisco, California\\n\\n Filed August 19, 2019\\n\\n2 UNITED STATES V. SHAYOTA\\n\\n Before: Mary M. Schroeder, Diarmuid F. O’Scannlain,\\n and Johnnie B. Rawlinson, Circuit Judges.\\n\\n Opinion by Judge O’Scannlain;\\n Concurrence by Judge O’Scannlain\\n\\n\\n SUMMARY*\\n\\n\\n Criminal Law\\n\\n The panel affirmed the district court’s judgment in a case\\nin which the panel was asked to decide whether prior civil\\ndeposition testimony of a witness, who has subsequently\\ninvoked his Fifth Amendment right against self-\\nincrimination, may be introduced against defendants in a\\ncriminal trial without violating their Confrontation Clause\\nright to confront the witnesses against them.\\n\\n At the defendants’ trial, the district court admitted the\\ncivil deposition testimony of two individuals after they\\ninvoked their Fifth Amendment privilege not to testify. The\\ndistrict court concluded that their invocation of their Fifth\\nAmendment privilege rendered them unavailable for purposes\\nof the Confrontation Clause. The defendants argued that the\\ngovernment’s inherent discretion to grant a witness immunity\\nand thereby prevent him from invoking the Fifth Amendment\\nprivilege renders the witness effectively available to the\\ngovernment for testimony at trial. The panel did not need to\\nresolve that issue because even if the district court erred by\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. SHAYOTA 3\\n\\nconcluding that the witnesses were unavailable, the error was\\nharmless because the outcome of the trial would not have\\nchanged had the depositions been excluded.\\n\\n The panel addressed other arguments in a memorandum\\ndisposition.\\n\\n Specially concurring, Judge O’Scannlain wrote separately\\nto call attention to this court’s precedent regarding the\\n“unavailability” requirement of the Confrontation Clause.\\nObserving that the right of confrontation is “most naturally\\nread as a reference to the right of confrontation at common\\nlaw, admitting only those exceptions established at the time\\nof the founding,” Crawford v. Washington, 541 U.S. 36, 54\\n(2004), Judge O’Scannlain suggested that this court revisit\\nits prior decisions to perform the historical analysis that\\nCrawford demands.\\n\\n\\n COUNSEL\\n\\nTheodore Sampsell-Jones (argued), Sampsell-Jones Law,\\nMinneapolis, Minnesota; Dennis P. Riordan and Donald M.\\nHorgan, Riordan & Horgan, San Francisco, California; for\\nDefendant-Appellant Joseph Shayota.\\n\\nJohn D. Cline (argued), Law Office of John D. Cline, San\\nFrancisco, California, for Defendant-Appellant Adriana\\nShayota.\\n\\nJonas Lerman (argued), Assistant United States Attorney; J.\\nDouglas Wilson, Chief, Appellate Division; United States\\nAttorney’s Office, San Francisco, California; for Plaintiff-\\nAppellee.\\n\\n4 UNITED STATES V. SHAYOTA\\n\\n OPINION\\n\\nO’SCANNLAIN, Circuit Judge:\\n\\n We are asked to decide whether prior civil deposition\\ntestimony of a witness, who has subsequently invoked his\\nFifth Amendment right against self-incrimination, may be\\nintroduced against defendants in a criminal trial without\\nviolating their Confrontation Clause right to confront the\\nwitnesses against them.\\n\\n I\\n\\n At the time of the events giving rise to this case,\\ndefendants Joseph Shayota and his wife Adriana Shayota ran\\nBaja Exporting, LLC (“Baja Exporting”)—a California\\ncompany that imported, exported, and distributed snacks and\\ndrinks to gas stations, convenience stores, and bodegas, both\\nin the United States and abroad. In 2009, Baja Exporting\\ncontracted with Living Essentials, LLC and its related entities\\n(collectively “Living Essentials”) to sell the liquid dietary\\nsupplement 5-Hour Energy in Mexico.\\n\\n Unfortunately, the product did not sell well. The parties\\nterminated their sales agreement in 2010, and Baja Exporting\\nwas left with excess bottles of Spanish-labeled 5-Hour\\nEnergy. Rather than dispose of the bottles, Baja Exporting\\ntried to sell them in the United States, despite not having\\napproval from Living Essentials to do so. When the product\\nagain failed to sell, the Shayotas and their associates\\nrelabeled the bottles in English and sold them without\\nauthorization from Living Essentials.\\n\\n UNITED STATES V. SHAYOTA 5\\n\\n By December 2011, Baja Exporting had sold the last of its\\n5-Hour Energy supply. Given the potential market in the\\nUnited States, the Shayotas and their associates then\\nconspired to create a counterfeit version of the drink, pass it\\noff as genuine 5-Hour Energy, and distribute it across the\\ncountry.\\n\\n A\\n\\n The scheme lasted from early 2012 to November 2012\\nand involved several key players. Joseph Shayota oversaw\\nand financed the operation. Adriana Shayota was in charge\\nof the accounting; she collected invoices for the counterfeit\\nproduct and made wire transfers to cover costs. Walid Jamil\\n(Joseph’s brother-in-law) and Justin Shayota (Joseph and\\nAdriana’s nephew) coordinated the repackaging and\\nrelabeling of the counterfeit drink at Baja Exporting’s San\\nDiego warehouse. They worked closely with Leslie Roman,\\nwho supplied blank bottles and manufactured counterfeit 5-\\nHour Energy labels. Jamil hired others to create the\\ncounterfeit drink and to manufacture boxes for the finished\\nproduct. Finally, Justin sent the finished product either to\\nBaja Exporting or to Dan Dee Company—a cash-and-carry\\nwarehouse owned by Kevin Attiq—for distribution across the\\ncountry.\\n\\n Living Essentials became aware of the scheme in 2012,\\nafter it noticed a mysterious drop-off in its California sales.\\nLiving Essentials hired private investigators, who ultimately\\nraided warehouses controlled by Baja Exporting and Jamil\\nand discovered boxes of counterfeit 5-Hour Energy.\\nConsequently, Living Essentials sued the Shayotas, Jamil,\\nand others, alleging numerous violations of laws pertaining\\nto trademark infringement and false advertising.\\n\\n6 UNITED STATES V. SHAYOTA\\n\\n During discovery, Living Essentials deposed numerous\\nparticipants in the scheme, including Jamil and Roman.\\nBecause the Shayotas were parties to the suit, their counsel\\nattended the depositions and questioned the witnesses. The\\ncivil suit ultimately ended in a settlement in which Baja\\nExporting agreed to pay $6 million to Living Essentials.\\n\\n B\\n\\n Subsequent to the civil proceedings, the government\\nlaunched a criminal investigation into the scheme. A grand\\njury returned an indictment against the Shayotas, Jamil, and\\nRoman, among others. The government ultimately filed a\\ntwo-count Superseding Information charging each defendant\\nwith (1) conspiracy to traffic in counterfeit goods, in violation\\nof 18 U.S.C. § 2320(a), and (2) conspiracy to commit\\ncopyright infringement and to introduce misbranded food\\ninto interstate commerce, in violation of 17 U.S.C. § 506,\\n18 U.S.C. § 2319, and 21 U.S.C. §§ 331 and 333. Jamil and\\nRoman pleaded guilty; the Shayotas proceeded to trial.\\n\\n As trial approached, the government notified defense\\ncounsel of its intent to offer into evidence statements made by\\nJamil and Roman during their civil depositions. The\\nShayotas moved to exclude the testimony on the ground that\\nits admission would violate the Sixth Amendment’s\\nConfrontation Clause if Jamil and Roman refused to testify,\\nbut the district court denied the motion. The depositions were\\nintroduced at trial after Jamil and Roman invoked their Fifth\\nAmendment privilege not to testify, and the jury ultimately\\nfound the Shayotas guilty of the charged offenses.\\n\\n The Shayotas timely appealed their convictions.\\n\\n UNITED STATES V. SHAYOTA 7\\n\\n II\\n\\n The Shayotas argue that admission of Jamil’s and\\nRoman’s deposition testimony violated the Sixth\\nAmendment’s Confrontation Clause.1\\n\\n The Sixth Amendment to the United States Constitution\\nprovides in pertinent part: “[I]n all criminal prosecutions the\\naccused shall enjoy the right . . . to be confronted with the\\nwitnesses against him.” U.S. Const. amend. VI. Generally,\\nthis means that the government must produce at trial any\\nwitnesses who have offered testimonial evidence against the\\naccused. But where, as here, the prosecution chooses not to\\nproduce such witnesses and offers their hearsay statements\\ninstead, a defendant’s right of confrontation is satisfied if\\n(1) the declarant is unavailable, and (2) the defendant had a\\nprior opportunity to confront him through cross-examination.\\nCrawford v. Washington, 541 U.S. 36, 68 (2004). The district\\ncourt concluded that Jamil and Roman were unavailable\\nbecause they had validly invoked their Fifth Amendment\\nprivilege against self-incrimination. The court also ruled that\\nthe Shayotas had a prior opportunity to cross-examine Jamil\\nand Roman in the civil suit, given that the Shayotas sent their\\nlawyers to attend Jamil’s and Roman’s depositions in that\\ncase.\\n\\n The Shayotas dispute the district court’s legal conclusion\\nthat Jamil and Roman were unavailable. They argue that\\nJamil and Roman were in fact available because the\\n\\n\\n 1\\n We address the Shayotas’ other arguments on appeal in a\\nmemorandum disposition issued concurrently with this opinion. See\\nUnited States v. Shayota, Nos. 17-10270 & 17-10271, — F. App’x — (9th\\nCir. 2019).\\n\\n8 UNITED STATES V. SHAYOTA\\n\\ngovernment could have granted them immunity and\\ncompelled them to testify.\\n\\n We have held on numerous occasions that a witness’s\\nassertion of Fifth Amendment privilege renders him\\n“unavailable” for purposes of the Confrontation Clause. See,\\ne.g., United States v. Wilmore, 381 F.3d 868, 872 (9th Cir.\\n2004); see also California v. Green, 399 U.S. 149, 167–68\\n(1970) (explaining in dicta that once a witness has been\\nproduced at trial and “claimed his privilege against\\ncompulsory self-incrimination, . . . nothing in the\\nConfrontation Clause prohibit[s] the State from . . . relying on\\nhis prior testimony to prove its case”). But we have yet to\\nconfront the Shayotas’ precise argument here—that the\\ngovernment’s inherent discretion to grant a witness immunity\\nand thereby prevent him from invoking the privilege against\\nself-incrimination renders the witness effectively available to\\nthe government for testimony at trial. Nevertheless, we need\\nnot resolve the issue here: even if the district court erred by\\nconcluding that Jamil and Roman were unavailable, the error\\nwas harmless.\\n\\n “[C]onfrontation [C]lause violations are subject to\\nharmless error analysis.” United States v. Bernard S.,\\n795 F.2d 749, 756 (9th Cir. 1986). “Whether a violation of\\nthe [C]onfrontation [C]lause is harmless depends on a variety\\nof factors including: (1) the importance of the evidence to the\\nprosecution’s case; (2) whether the evidence was cumulative;\\n(3) the presence of corroborating evidence; (4) the overall\\nstrength of the prosecution’s case.” Id. To be sure, the\\ntestimony of Jamil and Roman was an important component\\nof the trial. But the testimony was also corroborated by live\\nwitness testimony, including that of Kevin Attiq and Justin\\nShayota who described the Shayotas’ involvement in the\\n\\n UNITED STATES V. SHAYOTA 9\\n\\nscheme, and a bounty of other circumstantial evidence\\nrelating to the Shayotas’ management of the counterfeit\\nproduct and exchange of invoices with other members of the\\nconspiracy. Considering the other evidence presented at trial,\\nthe prosecution’s case against the Shayotas was strong. We\\nconclude that the outcome of the trial would not have\\nchanged had the depositions been excluded.2\\n\\n III\\n\\n The judgment of the district court is AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4081,"cells":{"id":{"kind":"number","value":4649429,"string":"4,649,429"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Chemehuevi Indian Tribe v. John McMahon"},"decision_date":{"kind":"string","value":"2019-08-19"},"docket_number":{"kind":"string","value":"17-56791"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Kim McLane Wardlaw and Andrew D. Hurwitz, Circuit Judges, and Edward R. Korman, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"HURWITZ, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCHEMEHUEVI INDIAN TRIBE, on its No. 17-56791\\nown behalf and on behalf of its\\nmembers parens patriae; CHELSEA D.C. No.\\nLYNN BUNIM; TOMMIE ROBERT 5:15-cv-01538-\\nOCHOA; JASMINE SANSOUCIE; NAOMI DMG-FFM\\nLOPEZ,\\n Plaintiffs-Appellants,\\n OPINION\\n v.\\n\\nJOHN MCMAHON, in his official\\ncapacity as Sheriff of San\\nBernardino County; RONALD\\nSINDELAR, in his official capacity as\\ndeputy sheriff for San Bernardino\\nCounty,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Dolly M. Gee, District Judge, Presiding\\n\\n Argued and Submitted May 13, 2019\\n Pasadena, California\\n\\n Filed August 19, 2019\\n\\f2 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\n Before: Kim McLane Wardlaw and Andrew D. Hurwitz,\\n Circuit Judges, and Edward R. Korman, * District Judge.\\n\\n Opinion by Judge Hurwitz\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights/Indian Country\\n\\n The panel affirmed in part and vacated in part the district\\ncourt’s dismissal of a complaint and remanded in an action\\nbrought pursuant to 42 U.S.C. § 1983 by the Chemehuevi\\nIndian Tribe and four of its enrolled members alleging\\nviolations of various federal statutory and constitutional\\nrights in connection with citations by San Bernardino\\nCounty Sheriff’s Deputies of four Tribe members for\\nviolating California regulatory traffic laws.\\n\\n The panel first analyzed the history and establishment of\\nthe Chemehuevi Reservation and concluded that the area\\nwhere the Tribe members were cited was within the\\nboundaries of the Reservation and hence was “Indian\\ncountry” under 18 U.S.C. § 1151(a). Accordingly, the panel\\nheld that San Bernardino County did not have jurisdiction to\\nenforce California regulatory traffic laws within that area.\\n\\n\\n\\n *\\n The Honorable Edward R. Korman, United States District Judge\\nfor the Eastern District of New York, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 3\\n\\n The panel held that the individual plaintiffs, but not the\\nTribe, could challenge the citations under § 1983. The panel\\nheld that because § 1983 was designed to secure private\\nrights against government encroachment, tribal members\\ncould use it to vindicate their individual rights, but not the\\ntribe’s communal rights. The panel therefore vacated the\\ndistrict court’s judgment dismissing the complaint as to the\\nindividuals, but affirmed the judgment as to the Tribe.\\n\\n\\n COUNSEL\\n\\nLester John Marston (argued), Rapport and Marston, Ukiah,\\nCalifornia, for Plaintiffs-Appellants.\\n\\nShaun M. Murphy (argued) and Katelyn K. Empey, Slovak\\nBaron Empey Murphy & Pinkey LLP, Palm Springs,\\nCalifornia; Michelle Blakemore and Miles Kowalski, San\\nBernardino County Counsel, San Bernardino, California; for\\nDefendants-Appellees.\\n\\n\\n OPINION\\n\\nHURWITZ, Circuit Judge:\\n\\n In 2015, San Bernardino County Sheriff’s Deputies cited\\nfour enrolled members of the Chemehuevi Indian Tribe for\\nviolating California regulatory traffic laws. Two of the\\nTribe’s members were cited on Section 36 of Township\\n5 North, Range 24 East (“Section 36”), a one square mile\\nplot the Tribe claims is part of its Reservation; two were\\ncited elsewhere on the Reservation.\\n\\f4 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\n It is undisputed that the Sheriff cannot enforce regulatory\\ntraffic laws in “Indian country.” See 18 U.S.C. § 1162;\\n28 U.S.C. § 1360. “Indian country” includes, but is not\\nlimited to, land within the boundaries of a reservation.\\n18 U.S.C. § 1151. The issues for decision today are\\n(1) whether the individual Tribe members and the Tribe can\\nchallenge the citations through a 42 U.S.C. § 1983 action;\\nand, if so, (2) whether Section 36 is Indian country. We hold\\nthat the individual plaintiffs, but not the Tribe, can challenge\\nthe citations under § 1983. And, we conclude that all the\\ncitations occurred within Indian country. We therefore\\nvacate the district court’s judgment dismissing the complaint\\nas to the individuals but affirm the judgment as to the Tribe.\\n\\n I. Background.\\n\\nA. Facts.\\n\\n Chelsea Lynn Bunim, Jasmine Sansoucie, Tommie\\nRobert Ochoa, and Naomi Lopez are enrolled members of\\nthe Chemehuevi Tribe. Each was stopped and cited by a San\\nBernardino County Sheriff’s Deputy for violating a\\nCalifornia regulatory traffic law. Deputy Sheriff Ronald\\nSindelar stopped and cited Bunim on Section 36 for driving\\nwithout a valid registration. Sindelar impounded Bunim’s\\ncar, leaving her alone on the roadside. Deputy Sindelar also\\nstopped Sansoucie on Section 36, citing her for driving with\\na suspended license.\\n\\n Deputy Sindelar cited Ochoa for driving without a valid\\nregistration and failing to provide evidence of financial\\nresponsibility. Sindelar had Ochoa’s car towed, leaving him\\nalone on the roadside. Deputy Sheriff J. Wagner cited Lopez\\nfor driving without a valid registration. Both of these\\ncitations were issued at locations that the parties agree are\\ninside the boundaries of the Chemehuevi Reservation.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 5\\n\\n Bunim, Sansoucie, Ochoa, Lopez, and the Tribe sued the\\nSheriff and the Deputies under 42 U.S.C. § 1983, alleging\\nviolations of various federal statutory and constitutional\\nrights. The complaint sought monetary damages, a\\ndeclaratory judgment, and injunctive relief. The defendants\\nargued that none of the plaintiffs’ claims was cognizable\\nunder § 1983. In addition, in response to the claims raised\\nby Bunim and Sansoucie, the defendants argued that Section\\n36 was outside the Reservation boundaries, and therefore\\nwithin the County’s regulatory jurisdiction.\\n\\nB. Procedural History.\\n\\n The district court initially entered a preliminary\\ninjunction prohibiting the defendants from “citing, arresting,\\nimpounding the vehicles of, and prosecuting Chemehuevi\\ntribal members for on-reservation violations” of California\\nregulatory vehicle laws, including violations occurring on\\nSection 36. The court determined there were “at least\\nserious questions going to the merits” of whether Section 36\\nwas “Indian country.”\\n\\n But, the court later granted summary judgment to the\\ndefendants, concluding that Section 36 was not part of the\\nChemehuevi Reservation and therefore not Indian country\\nunder 18 U.S.C. § 1151(a). In a motion to amend the\\njudgment, Ochoa and Lopez noted that they were not\\nticketed in Section 36. The district court denied the motion,\\nholding that even if the plaintiffs were cited on the\\nReservation, they failed to allege “a well-established\\nconstitutional violation for purposes of their section 1983\\nclaim.” The court reasoned that § 1983 “is concerned with\\nthe relationship between individuals and the state, not the\\ndistribution of power between state, federal, or tribal\\ngovernments,” and therefore neither the “right to be free of\\n\\f6 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nstate regulation” nor “the right to tribal government” is\\n“within the scope of section 1983.” 1\\n\\n The Tribe and the individual plaintiffs timely appealed.\\nWe have jurisdiction of that appeal under 28 U.S.C. § 1291\\nand review the summary judgment de novo. Parravano v.\\nBabbitt, 70 F.3d 539, 543 (9th Cir. 1995).\\n\\n II. Discussion.\\n\\nA. Is Section 36 Indian Country?\\n\\n We turn first to the question whether Section 36 is in the\\nChemehuevi Reservation, and thus Indian country under\\n18 U.S.C. § 1151(a). But it is important also to note at the\\noutset what issues are not before us. We need not—and do\\nnot—decide today who holds title to Section 36. Indian\\ncountry includes “all land within the limits of any Indian\\nreservation . . . notwithstanding the issuance of any patent.”\\n18 U.S.C. § 1151(a). “[A]djudicating reservation\\nboundaries is conceptually quite distinct from adjudicating\\ntitle to the same lands. One inquiry does not necessarily\\nhave anything in common with the other, as title and\\nreservation status are not congruent concepts in Indian law.”\\nNavajo Tribe of Indians v. New Mexico, 809 F.2d 1455, 1475\\n(10th Cir. 1987) (internal quotation marks and footnote\\nomitted); see Solem v. Bartlett, 465 U.S. 463, 466–68\\n(1984).\\n\\n\\n 1\\n The complaint also alleged that the citations “constitute[d] racial\\ndiscrimination in direct violation” of the Equal Protection Clause. The\\ndistrict court held that “the specter of racial animus” was not “sufficient\\nto create a triable issue of fact that Defendants violated” the Fourteenth\\nAmendment. The plaintiffs do not challenge that ruling on appeal.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 7\\n\\n Our inquiry as to the reservation status of Section 36\\nbegins in 1853. After California gained statehood, Congress\\nordered a survey of its public lands and granted the State title\\nto sections 16 and 36 of each township. 2 Act of March 3,\\n1853, ch. 145, 10 Stat. 244, 245–46. But, the 1853 Act\\nspecifically excluded from that grant any land “in the\\noccupation or possession of any Indian tribe.” Id. at 246–47.\\nThe Surveyor General approved a survey of the land at issue\\nin this case in 1895.\\n\\n While that survey was being conducted, Congress\\nordered the Secretary of the Interior “to select a reservation”\\nfor each California Mission Indian tribe. Mission Indian\\nRelief Act, ch. 65, 26 Stat. 712, 712 (1891). The\\nreservations were to “include, as far as practicable, the lands\\nand villages which have been in the actual occupation and\\npossession of said Indians.” Id. Although the Secretary was\\nalso instructed to “cause a patent to issue for each”\\nreservation, and thus transfer title to the land to the United\\nStates as trustee for the tribes, the Act provided that the\\nreservations would be “valid when approved by the\\n[Executive Branch].” Id. In 1905, Congress authorized the\\nSecretary “to investigate through an inspector . . . existing\\nconditions of the California Indians and to report to\\nCongress at the next session some plan to improve the\\nsame.” Act of March 3, 1905, ch. 1479, 33 Stat. 1048, 1058.\\n\\n Special Agent C.E. Kelsey was then dispatched to visit\\nthe Chemehuevi Tribe and identify territory for a\\nreservation. In 1907, Kelsey issued a report to the\\n\\n 2\\n See U.S. Geological Survey, The Public Land Survey System, The\\nNational Map Small Scale (Jan. 18, 2018),\\nhttps://nationalmap.gov/small_scale/a_plss.html (explaining township,\\nrange, and section designations).\\n\\f8 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nCommissioner of Indian Affairs, identifying land to be\\nincluded in the reservation. He specifically recommended\\nthat the reservation include the eastern half of Township\\n5 North, Range 24 East (“E. 1/2 of T. 5 N., R. 24 E.”)—\\nwhich contains Section 36. Kelsey noted that this land was\\nthe “present location” of the tribal members and that “there\\nis no question but they have occupied this land since\\nprimeval times.” The Commissioner forwarded Kelsey’s\\nrecommendation to the Secretary of the Interior.\\n\\n In an executive order (the “1907 Order”), the Secretary\\nthen “direct[ed] that the lands referred to” by Kelsey and the\\nCommissioner “be withdrawn from all form of settlement,”\\nand created the Chemehuevi Reservation. The Secretary\\nalso asked Congress “to authorize the addition of certain\\nlands to the Mission Indian Reservations.” Although\\nCongress did not act upon this proposed legislation, it\\nsubsequently recognized the existence of the Chemehuevi\\nReservation in the Parker Dam Act, ch. 522, 54 Stat. 744\\n(1940).\\n\\n It is clear that a Chemehuevi Reservation was validly\\nestablished by the Secretary’s 1907 Order, notwithstanding\\nthe absence of subsequent Congressional approval. Indeed,\\nthe Supreme Court has expressly so recognized:\\n\\n Congress and the Executive have ever since\\n recognized these as Indian Reservations. . . .\\n They have been uniformly and universally\\n treated as reservations by map makers,\\n surveyors, and the public. We can give but\\n short shrift at this late date to the argument\\n that the reservations . . . are invalid because\\n they were originally set apart by the\\n Executive.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 9\\n\\nArizona v. California, 373 U.S. 546, 598 (1963); see also id.\\nat 596 & n.100. 3\\n\\n The defendants argue that the 1907 Order was invalid ab\\ninitio because Section 36 had already been deeded to\\nCalifornia. The factual premise of that argument, however,\\nis subject to question. The 1853 Act excluded any land “in\\nthe occupation or possession of any Indian tribe,” 10 Stat.\\nat 246–47, and the Kelsey survey, adopted by the Secretary,\\ndocuments that Section 36 falls in that exception. The\\ndistrict court erred in excluding the Kelsey report as hearsay.\\nIt is plainly admissible as an ancient document, Fed. R. Evid.\\n803(16), which may contain multiple levels of hearsay.\\n30B Charles Alan Wright & Arthur R. Miller, Federal\\nPractice & Procedure § 6935 (2018 ed.) (“[E]xclusion of\\nstatements in qualifying ancient documents on the grounds\\nthat the author lacked firsthand knowledge, or (relatedly)\\nthat the document contains hearsay-within-hearsay should\\nbe rare.”). Review of historical documents is typical—\\nindeed often necessary—in cases involving the boundaries\\nof Indian reservations. See, e.g., Idaho v. United States,\\n533 U.S. 262, 265–71 (2001).\\n\\n But, as noted, we need not today decide the extent of the\\n1853 land grant. “[E]xecutive orders must be liberally\\nconstrued in favor of establishing Indian rights,”\\nConfederated Tribes of Chehalis Indian Reservation v.\\n\\n 3\\n The defendants cite Arizona v. California, 460 U.S. 605, 636 n.26\\n(1983), for the proposition that “the secretarial orders do not constitute\\n‘final determinations.’” But, “the secretarial orders” at issue in that\\nlitigation were issued in the 1960s and 1970s. See id. at 631–32. In 1919\\nand 1927, Congress “prohibited future changes in Indian reservations by\\nexecutive order.” United States v. S. Pac. Transp. Co., 543 F.2d 676,\\n686 & n.15 (9th Cir. 1976). That prohibition plainly does not affect the\\n1907 Order.\\n\\f10 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nWashington, 96 F.3d 334, 340 (9th Cir. 1996), and are\\n“interpreted as the Indians would have understood them,”\\nParravano, 70 F.3d at 544. Given the language of the 1853\\nAct, the Kelsey report identifying Section 36 as land\\noccupied historically by Indians, and the express inclusion\\nof Section 36 in the 1907 Order, the Chemehuevi Tribe (and\\nindeed, the Secretary of the Interior) surely understood\\nSection 36 to be within the Reservation.\\n\\n Nor can we conclude that the boundaries of the\\nReservation as established in the 1907 Order were later\\ndiminished. “We do not lightly infer diminishment of\\nreservations.” Confederated Tribes of Chehalis, 96 F.3d\\nat 343–44. After 1927, Congress prohibited any change to\\nthe boundaries of existing executive-order reservations\\nexcept by Congressional act. Act of March 3, 1927, ch. 299,\\n§ 4, 44 Stat. 1347 (codified at 25 U.S.C. § 398d); see S. Pac.\\nTransp., 543 F.2d at 686 & n.15. There is no such act\\nremoving Section 36 from the Chemehuevi Reservation.\\n\\n The defendants also rely on a patent issued to the Tribe\\nby the Bureau of Land Management in 2010, which excluded\\n“[t]hose lands granted to the State of California . . . on July\\n10, 1895”—the date on which the government survey was\\nfinalized—“located in . . . sec. 36, T. 5 N., R. 24 E.” But, as\\nnoted above, we do not today adjudicate title. More\\nimportantly, because the 2010 patent was issued over a\\ncentury after the Reservation was established, it provides no\\nevidence of the intent of the Executive or the understanding\\nof the Tribe in 1907. Nor can it, nor does it purport to,\\ndiminish the Chemehuevi Reservation. The patent cites the\\n1907 Order, then grants some of the land covered by that\\norder to the Tribe in trust—an issue of ownership. It is silent\\nas to the reservation status of any land excluded from the\\npatent.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 11\\n\\n We therefore conclude that Section 36 is within the\\nChemehuevi Reservation and hence “Indian country” under\\n18 U.S.C. § 1151(a).\\n\\nB. Can the Plaintiffs Sue Under § 1983?\\n\\n California cannot enforce state law that regulates—but\\ndoes not prohibit—tribal members’ conduct inside a\\nreservation. Confederated Tribes of Colville Reservation v.\\nWashington, 938 F.2d 146, 147 (9th Cir. 1991) (citing\\n18 U.S.C. § 1162; 28 U.S.C. § 1360). The defendants\\nconcede that the citations at issue involved regulatory laws\\nand therefore could not be issued against enrolled members\\nof the Tribe within the boundaries of the Reservation. See\\nid. at 148. But, they argue that even such citations cannot be\\nthe subject of a § 1983 action.\\n\\n We disagree. Section 1983 allows any “person” to sue\\nfor the “deprivation of any rights, privileges, or immunities\\nsecured by the Constitution and laws.” 42 U.S.C. § 1983.\\nBecause § 1983 “was designed to secure private rights\\nagainst government encroachment,” Inyo Cty. v. Paiute-\\nShoshone Indians of the Bishop Cmty. of the Bishop Colony,\\n538 U.S. 701, 712 (2003), tribal members can use it to\\nvindicate their “individual rights,” but not the tribe’s\\n“communal rights,” Skokomish Indian Tribe v. United\\nStates, 410 F.3d 506, 515–16 (9th Cir. 2005) (en banc). And,\\n“traditional section 1983 suit[s]”—for example, those\\nchallenging an arrest on tribal land—seek to vindicate an\\n“individual right.” See id. at 516 n.8 (citing Romero v.\\nKitsap Cty., 931 F.2d 624, 627 n.5 (9th Cir. 1991)).\\n\\n Bunim, Sansoucie, Ochoa, and Lopez’s claims are\\n“traditional” § 1983 suits. Each was stopped and detained\\nby a San Bernardino County Deputy; some had their vehicles\\nseized. They contend that their detentions and citations\\n\\f12 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nviolated the Constitution and federal statutes. They have a\\ncause of action under § 1983 against the defendants.\\n\\n The Tribe, however, does not have a § 1983 claim. An\\nIndian tribe “may not sue under § 1983 to vindicate” a\\n“sovereign right,” such as its right to be free of state\\nregulation and control. Inyo Cty., 538 U.S. at 712. Nor can\\nthe Tribe assert its members’ individual rights as parens\\npatriae in a § 1983 action. To assert parens patriae\\nstanding, the Tribe would have to “articulate an interest apart\\nfrom the interests of particular private parties,” i.e., “be more\\nthan a nominal party,” and “express a quasi-sovereign\\ninterest.” Missouri ex rel. Koster v. Harris, 847 F.3d 646,\\n651 (9th Cir. 2017). That requirement is inconsistent with a\\n§ 1983 action: quasi-sovereign interests are not individual\\nrights.\\n\\n III. Conclusion.\\n\\n The Chemehuevi Reservation, as established by the 1907\\nOrder, includes Section 36. Section 36 is therefore Indian\\ncountry, and San Bernardino County does not have\\njurisdiction to enforce California regulatory laws within it.\\nThe individual plaintiffs may bring § 1983 claims against the\\ndefendants. The Tribe, however, cannot assert its sovereign\\nrights under that statute. 4\\n\\n AFFIRMED in part, VACATED and REMANDED\\nin part. Each party shall bear its own costs.\\n\\n\\n\\n\\n 4\\n We take no position on any defenses, including immunity, the\\ndefendants might have to the claims raised by the individuals.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4426682/\", \"author_raw\": \"HURWITZ, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"KIM MCLANE WARDLAW"},"judge2":{"kind":"string","value":"ANDREW D HURWITZ"},"judge3":{"kind":"string","value":"EDWARD R KORMAN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649429/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Chemehuevi Indian Tribe v. John McMahon\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-19\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCHEMEHUEVI INDIAN TRIBE, on its No. 17-56791\nown behalf and on behalf of its\nmembers parens patriae; CHELSEA D.C. No.\nLYNN BUNIM; TOMMIE ROBERT 5:15-cv-01538-\nOCHOA; JASMINE SANSOUCIE; NAOMI DMG-FFM\nLOPEZ,\n Plaintiffs-Appellants,\n OPINION\n v.\n\nJOHN MCMAHON, in his official\ncapacity as Sheriff of San\nBernardino County; RONALD\nSINDELAR, in his official capacity as\ndeputy sheriff for San Bernardino\nCounty,\n Defendants-Appellees.\n\n Appeal from the United States District Court\n for the Central District of California\n Dolly M. Gee, District Judge, Presiding\n\n Argued and Submitted May 13, 2019\n Pasadena, California\n\n Filed August 19, 2019\n\f2 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\n\n Before: Kim McLane Wardlaw and Andrew D. Hurwitz,\n Circuit Judges, and Edward R. Korman, * District Judge.\n\n Opinion by Judge Hurwitz\n\n\n SUMMARY **\n\n\n Civil Rights/Indian Country\n\n The panel affirmed in part and vacated in part the district\ncourt’s dismissal of a complaint and remanded in an action\nbrought pursuant to 42 U.S.C. § 1983 by the Chemehuevi\nIndian Tribe and four of its enrolled members alleging\nviolations of various federal statutory and constitutional\nrights in connection with citations by San Bernardino\nCounty Sheriff’s Deputies of four Tribe members for\nviolating California regulatory traffic laws.\n\n The panel first analyzed the history and establishment of\nthe Chemehuevi Reservation and concluded that the area\nwhere the Tribe members were cited was within the\nboundaries of the Reservation and hence was “Indian\ncountry” under 18 U.S.C. § 1151(a). Accordingly, the panel\nheld that San Bernardino County did not have jurisdiction to\nenforce California regulatory traffic laws within that area.\n\n\n\n *\n The Honorable Edward R. Korman, United States District Judge\nfor the Eastern District of New York, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 3\n\n The panel held that the individual plaintiffs, but not the\nTribe, could challenge the citations under § 1983. The panel\nheld that because § 1983 was designed to secure private\nrights against government encroachment, tribal members\ncould use it to vindicate their individual rights, but not the\ntribe’s communal rights. The panel therefore vacated the\ndistrict court’s judgment dismissing the complaint as to the\nindividuals, but affirmed the judgment as to the Tribe.\n\n\n COUNSEL\n\nLester John Marston (argued), Rapport and Marston, Ukiah,\nCalifornia, for Plaintiffs-Appellants.\n\nShaun M. Murphy (argued) and Katelyn K. Empey, Slovak\nBaron Empey Murphy & Pinkey LLP, Palm Springs,\nCalifornia; Michelle Blakemore and Miles Kowalski, San\nBernardino County Counsel, San Bernardino, California; for\nDefendants-Appellees.\n\n\n OPINION\n\nHURWITZ, Circuit Judge:\n\n In 2015, San Bernardino County Sheriff’s Deputies cited\nfour enrolled members of the Chemehuevi Indian Tribe for\nviolating California regulatory traffic laws. Two of the\nTribe’s members were cited on Section 36 of Township\n5 North, Range 24 East (“Section 36”), a one square mile\nplot the Tribe claims is part of its Reservation; two were\ncited elsewhere on the Reservation.\n\f4 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\n\n It is undisputed that the Sheriff cannot enforce regulatory\ntraffic laws in “Indian country.” See 18 U.S.C. § 1162;\n28 U.S.C. § 1360. “Indian country” includes, but is not\nlimited to, land within the boundaries of a reservation.\n18 U.S.C. § 1151. The issues for decision today are\n(1) whether the individual Tribe members and the Tribe can\nchallenge the citations through a 42 U.S.C. § 1983 action;\nand, if so, (2) whether Section 36 is Indian country. We hold\nthat the individual plaintiffs, but not the Tribe, can challenge\nthe citations under § 1983. And, we conclude that all the\ncitations occurred within Indian country. We therefore\nvacate the district court’s judgment dismissing the complaint\nas to the individuals but affirm the judgment as to the Tribe.\n\n I. Background.\n\nA. Facts.\n\n Chelsea Lynn Bunim, Jasmine Sansoucie, Tommie\nRobert Ochoa, and Naomi Lopez are enrolled members of\nthe Chemehuevi Tribe. Each was stopped and cited by a San\nBernardino County Sheriff’s Deputy for violating a\nCalifornia regulatory traffic law. Deputy Sheriff Ronald\nSindelar stopped and cited Bunim on Section 36 for driving\nwithout a valid registration. Sindelar impounded Bunim’s\ncar, leaving her alone on the roadside. Deputy Sindelar also\nstopped Sansoucie on Section 36, citing her for driving with\na suspended license.\n\n Deputy Sindelar cited Ochoa for driving without a valid\nregistration and failing to provide evidence of financial\nresponsibility. Sindelar had Ochoa’s car towed, leaving him\nalone on the roadside. Deputy Sheriff J. Wagner cited Lopez\nfor driving without a valid registration. Both of these\ncitations were issued at locations that the parties agree are\ninside the boundaries of the Chemehuevi Reservation.\n\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 5\n\n Bunim, Sansoucie, Ochoa, Lopez, and the Tribe sued the\nSheriff and the Deputies under 42 U.S.C. § 1983, alleging\nviolations of various federal statutory and constitutional\nrights. The complaint sought monetary damages, a\ndeclaratory judgment, and injunctive relief. The defendants\nargued that none of the plaintiffs’ claims was cognizable\nunder § 1983. In addition, in response to the claims raised\nby Bunim and Sansoucie, the defendants argued that Section\n36 was outside the Reservation boundaries, and therefore\nwithin the County’s regulatory jurisdiction.\n\nB. Procedural History.\n\n The district court initially entered a preliminary\ninjunction prohibiting the defendants from “citing, arresting,\nimpounding the vehicles of, and prosecuting Chemehuevi\ntribal members for on-reservation violations” of California\nregulatory vehicle laws, including violations occurring on\nSection 36. The court determined there were “at least\nserious questions going to the merits” of whether Section 36\nwas “Indian country.”\n\n But, the court later granted summary judgment to the\ndefendants, concluding that Section 36 was not part of the\nChemehuevi Reservation and therefore not Indian country\nunder 18 U.S.C. § 1151(a). In a motion to amend the\njudgment, Ochoa and Lopez noted that they were not\nticketed in Section 36. The district court denied the motion,\nholding that even if the plaintiffs were cited on the\nReservation, they failed to allege “a well-established\nconstitutional violation for purposes of their section 1983\nclaim.” The court reasoned that § 1983 “is concerned with\nthe relationship between individuals and the state, not the\ndistribution of power between state, federal, or tribal\ngovernments,” and therefore neither the “right to be free of\n\f6 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\n\nstate regulation” nor “the right to tribal government” is\n“within the scope of section 1983.” 1\n\n The Tribe and the individual plaintiffs timely appealed.\nWe have jurisdiction of that appeal under 28 U.S.C. § 1291\nand review the summary judgment de novo. Parravano v.\nBabbitt, 70 F.3d 539, 543 (9th Cir. 1995).\n\n II. Discussion.\n\nA. Is Section 36 Indian Country?\n\n We turn first to the question whether Section 36 is in the\nChemehuevi Reservation, and thus Indian country under\n18 U.S.C. § 1151(a). But it is important also to note at the\noutset what issues are not before us. We need not—and do\nnot—decide today who holds title to Section 36. Indian\ncountry includes “all land within the limits of any Indian\nreservation . . . notwithstanding the issuance of any patent.”\n18 U.S.C. § 1151(a). “[A]djudicating reservation\nboundaries is conceptually quite distinct from adjudicating\ntitle to the same lands. One inquiry does not necessarily\nhave anything in common with the other, as title and\nreservation status are not congruent concepts in Indian law.”\nNavajo Tribe of Indians v. New Mexico, 809 F.2d 1455, 1475\n(10th Cir. 1987) (internal quotation marks and footnote\nomitted); see Solem v. Bartlett, 465 U.S. 463, 466–68\n(1984).\n\n\n 1\n The complaint also alleged that the citations “constitute[d] racial\ndiscrimination in direct violation” of the Equal Protection Clause. The\ndistrict court held that “the specter of racial animus” was not “sufficient\nto create a triable issue of fact that Defendants violated” the Fourteenth\nAmendment. The plaintiffs do not challenge that ruling on appeal.\n\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 7\n\n Our inquiry as to the reservation status of Section 36\nbegins in 1853. After California gained statehood, Congress\nordered a survey of its public lands and granted the State title\nto sections 16 and 36 of each township. 2 Act of March 3,\n1853, ch. 145, 10 Stat. 244, 245–46. But, the 1853 Act\nspecifically excluded from that grant any land “in the\noccupation or possession of any Indian tribe.” Id. at 246–47.\nThe Surveyor General approved a survey of the land at issue\nin this case in 1895.\n\n While that survey was being conducted, Congress\nordered the Secretary of the Interior “to select a reservation”\nfor each California Mission Indian tribe. Mission Indian\nRelief Act, ch. 65, 26 Stat. 712, 712 (1891). The\nreservations were to “include, as far as practicable, the lands\nand villages which have been in the actual occupation and\npossession of said Indians.” Id. Although the Secretary was\nalso instructed to “cause a patent to issue for each”\nreservation, and thus transfer title to the land to the United\nStates as trustee for the tribes, the Act provided that the\nreservations would be “valid when approved by the\n[Executive Branch].” Id. In 1905, Congress authorized the\nSecretary “to investigate through an inspector . . . existing\nconditions of the California Indians and to report to\nCongress at the next session some plan to improve the\nsame.” Act of March 3, 1905, ch. 1479, 33 Stat. 1048, 1058.\n\n Special Agent C.E. Kelsey was then dispatched to visit\nthe Chemehuevi Tribe and identify territory for a\nreservation. In 1907, Kelsey issued a report to the\n\n 2\n See U.S. Geological Survey, The Public Land Survey System, The\nNational Map Small Scale (Jan. 18, 2018),\nhttps://nationalmap.gov/small_scale/a_plss.html (explaining township,\nrange, and section designations).\n\f8 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\n\nCommissioner of Indian Affairs, identifying land to be\nincluded in the reservation. He specifically recommended\nthat the reservation include the eastern half of Township\n5 North, Range 24 East (“E. 1/2 of T. 5 N., R. 24 E.”)—\nwhich contains Section 36. Kelsey noted that this land was\nthe “present location” of the tribal members and that “there\nis no question but they have occupied this land since\nprimeval times.” The Commissioner forwarded Kelsey’s\nrecommendation to the Secretary of the Interior.\n\n In an executive order (the “1907 Order”), the Secretary\nthen “direct[ed] that the lands referred to” by Kelsey and the\nCommissioner “be withdrawn from all form of settlement,”\nand created the Chemehuevi Reservation. The Secretary\nalso asked Congress “to authorize the addition of certain\nlands to the Mission Indian Reservations.” Although\nCongress did not act upon this proposed legislation, it\nsubsequently recognized the existence of the Chemehuevi\nReservation in the Parker Dam Act, ch. 522, 54 Stat. 744\n(1940).\n\n It is clear that a Chemehuevi Reservation was validly\nestablished by the Secretary’s 1907 Order, notwithstanding\nthe absence of subsequent Congressional approval. Indeed,\nthe Supreme Court has expressly so recognized:\n\n Congress and the Executive have ever since\n recognized these as Indian Reservations. . . .\n They have been uniformly and universally\n treated as reservations by map makers,\n surveyors, and the public. We can give but\n short shrift at this late date to the argument\n that the reservations . . . are invalid because\n they were originally set apart by the\n Executive.\n\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 9\n\nArizona v. California, 373 U.S. 546, 598 (1963); see also id.\nat 596 & n.100. 3\n\n The defendants argue that the 1907 Order was invalid ab\ninitio because Section 36 had already been deeded to\nCalifornia. The factual premise of that argument, however,\nis subject to question. The 1853 Act excluded any land “in\nthe occupation or possession of any Indian tribe,” 10 Stat.\nat 246–47, and the Kelsey survey, adopted by the Secretary,\ndocuments that Section 36 falls in that exception. The\ndistrict court erred in excluding the Kelsey report as hearsay.\nIt is plainly admissible as an ancient document, Fed. R. Evid.\n803(16), which may contain multiple levels of hearsay.\n30B Charles Alan Wright & Arthur R. Miller, Federal\nPractice & Procedure § 6935 (2018 ed.) (“[E]xclusion of\nstatements in qualifying ancient documents on the grounds\nthat the author lacked firsthand knowledge, or (relatedly)\nthat the document contains hearsay-within-hearsay should\nbe rare.”). Review of historical documents is typical—\nindeed often necessary—in cases involving the boundaries\nof Indian reservations. See, e.g., Idaho v. United States,\n533 U.S. 262, 265–71 (2001).\n\n But, as noted, we need not today decide the extent of the\n1853 land grant. “[E]xecutive orders must be liberally\nconstrued in favor of establishing Indian rights,”\nConfederated Tribes of Chehalis Indian Reservation v.\n\n 3\n The defendants cite Arizona v. California, 460 U.S. 605, 636 n.26\n(1983), for the proposition that “the secretarial orders do not constitute\n‘final determinations.’” But, “the secretarial orders” at issue in that\nlitigation were issued in the 1960s and 1970s. See id. at 631–32. In 1919\nand 1927, Congress “prohibited future changes in Indian reservations by\nexecutive order.” United States v. S. Pac. Transp. Co., 543 F.2d 676,\n686 & n.15 (9th Cir. 1976). That prohibition plainly does not affect the\n1907 Order.\n\f10 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\n\nWashington, 96 F.3d 334, 340 (9th Cir. 1996), and are\n“interpreted as the Indians would have understood them,”\nParravano, 70 F.3d at 544. Given the language of the 1853\nAct, the Kelsey report identifying Section 36 as land\noccupied historically by Indians, and the express inclusion\nof Section 36 in the 1907 Order, the Chemehuevi Tribe (and\nindeed, the Secretary of the Interior) surely understood\nSection 36 to be within the Reservation.\n\n Nor can we conclude that the boundaries of the\nReservation as established in the 1907 Order were later\ndiminished. “We do not lightly infer diminishment of\nreservations.” Confederated Tribes of Chehalis, 96 F.3d\nat 343–44. After 1927, Congress prohibited any change to\nthe boundaries of existing executive-order reservations\nexcept by Congressional act. Act of March 3, 1927, ch. 299,\n§ 4, 44 Stat. 1347 (codified at 25 U.S.C. § 398d); see S. Pac.\nTransp., 543 F.2d at 686 & n.15. There is no such act\nremoving Section 36 from the Chemehuevi Reservation.\n\n The defendants also rely on a patent issued to the Tribe\nby the Bureau of Land Management in 2010, which excluded\n“[t]hose lands granted to the State of California . . . on July\n10, 1895”—the date on which the government survey was\nfinalized—“located in . . . sec. 36, T. 5 N., R. 24 E.” But, as\nnoted above, we do not today adjudicate title. More\nimportantly, because the 2010 patent was issued over a\ncentury after the Reservation was established, it provides no\nevidence of the intent of the Executive or the understanding\nof the Tribe in 1907. Nor can it, nor does it purport to,\ndiminish the Chemehuevi Reservation. The patent cites the\n1907 Order, then grants some of the land covered by that\norder to the Tribe in trust—an issue of ownership. It is silent\nas to the reservation status of any land excluded from the\npatent.\n\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 11\n\n We therefore conclude that Section 36 is within the\nChemehuevi Reservation and hence “Indian country” under\n18 U.S.C. § 1151(a).\n\nB. Can the Plaintiffs Sue Under § 1983?\n\n California cannot enforce state law that regulates—but\ndoes not prohibit—tribal members’ conduct inside a\nreservation. Confederated Tribes of Colville Reservation v.\nWashington, 938 F.2d 146, 147 (9th Cir. 1991) (citing\n18 U.S.C. § 1162; 28 U.S.C. § 1360). The defendants\nconcede that the citations at issue involved regulatory laws\nand therefore could not be issued against enrolled members\nof the Tribe within the boundaries of the Reservation. See\nid. at 148. But, they argue that even such citations cannot be\nthe subject of a § 1983 action.\n\n We disagree. Section 1983 allows any “person” to sue\nfor the “deprivation of any rights, privileges, or immunities\nsecured by the Constitution and laws.” 42 U.S.C. § 1983.\nBecause § 1983 “was designed to secure private rights\nagainst government encroachment,” Inyo Cty. v. Paiute-\nShoshone Indians of the Bishop Cmty. of the Bishop Colony,\n538 U.S. 701, 712 (2003), tribal members can use it to\nvindicate their “individual rights,” but not the tribe’s\n“communal rights,” Skokomish Indian Tribe v. United\nStates, 410 F.3d 506, 515–16 (9th Cir. 2005) (en banc). And,\n“traditional section 1983 suit[s]”—for example, those\nchallenging an arrest on tribal land—seek to vindicate an\n“individual right.” See id. at 516 n.8 (citing Romero v.\nKitsap Cty., 931 F.2d 624, 627 n.5 (9th Cir. 1991)).\n\n Bunim, Sansoucie, Ochoa, and Lopez’s claims are\n“traditional” § 1983 suits. Each was stopped and detained\nby a San Bernardino County Deputy; some had their vehicles\nseized. They contend that their detentions and citations\n\f12 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\n\nviolated the Constitution and federal statutes. They have a\ncause of action under § 1983 against the defendants.\n\n The Tribe, however, does not have a § 1983 claim. An\nIndian tribe “may not sue under § 1983 to vindicate” a\n“sovereign right,” such as its right to be free of state\nregulation and control. Inyo Cty., 538 U.S. at 712. Nor can\nthe Tribe assert its members’ individual rights as parens\npatriae in a § 1983 action. To assert parens patriae\nstanding, the Tribe would have to “articulate an interest apart\nfrom the interests of particular private parties,” i.e., “be more\nthan a nominal party,” and “express a quasi-sovereign\ninterest.” Missouri ex rel. Koster v. Harris, 847 F.3d 646,\n651 (9th Cir. 2017). That requirement is inconsistent with a\n§ 1983 action: quasi-sovereign interests are not individual\nrights.\n\n III. Conclusion.\n\n The Chemehuevi Reservation, as established by the 1907\nOrder, includes Section 36. Section 36 is therefore Indian\ncountry, and San Bernardino County does not have\njurisdiction to enforce California regulatory laws within it.\nThe individual plaintiffs may bring § 1983 claims against the\ndefendants. The Tribe, however, cannot assert its sovereign\nrights under that statute. 4\n\n AFFIRMED in part, VACATED and REMANDED\nin part. Each party shall bear its own costs.\n\n\n\n\n 4\n We take no position on any defenses, including immunity, the\ndefendants might have to the claims raised by the individuals.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Chemehuevi Indian Tribe v. John McMahon\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-19\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCHEMEHUEVI INDIAN TRIBE, on its No. 17-56791\\nown behalf and on behalf of its\\nmembers parens patriae; CHELSEA D.C. No.\\nLYNN BUNIM; TOMMIE ROBERT 5:15-cv-01538-\\nOCHOA; JASMINE SANSOUCIE; NAOMI DMG-FFM\\nLOPEZ,\\n Plaintiffs-Appellants,\\n OPINION\\n v.\\n\\nJOHN MCMAHON, in his official\\ncapacity as Sheriff of San\\nBernardino County; RONALD\\nSINDELAR, in his official capacity as\\ndeputy sheriff for San Bernardino\\nCounty,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Dolly M. Gee, District Judge, Presiding\\n\\n Argued and Submitted May 13, 2019\\n Pasadena, California\\n\\n Filed August 19, 2019\\n\\f2 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\n Before: Kim McLane Wardlaw and Andrew D. Hurwitz,\\n Circuit Judges, and Edward R. Korman, * District Judge.\\n\\n Opinion by Judge Hurwitz\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights/Indian Country\\n\\n The panel affirmed in part and vacated in part the district\\ncourt’s dismissal of a complaint and remanded in an action\\nbrought pursuant to 42 U.S.C. § 1983 by the Chemehuevi\\nIndian Tribe and four of its enrolled members alleging\\nviolations of various federal statutory and constitutional\\nrights in connection with citations by San Bernardino\\nCounty Sheriff’s Deputies of four Tribe members for\\nviolating California regulatory traffic laws.\\n\\n The panel first analyzed the history and establishment of\\nthe Chemehuevi Reservation and concluded that the area\\nwhere the Tribe members were cited was within the\\nboundaries of the Reservation and hence was “Indian\\ncountry” under 18 U.S.C. § 1151(a). Accordingly, the panel\\nheld that San Bernardino County did not have jurisdiction to\\nenforce California regulatory traffic laws within that area.\\n\\n\\n\\n *\\n The Honorable Edward R. Korman, United States District Judge\\nfor the Eastern District of New York, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 3\\n\\n The panel held that the individual plaintiffs, but not the\\nTribe, could challenge the citations under § 1983. The panel\\nheld that because § 1983 was designed to secure private\\nrights against government encroachment, tribal members\\ncould use it to vindicate their individual rights, but not the\\ntribe’s communal rights. The panel therefore vacated the\\ndistrict court’s judgment dismissing the complaint as to the\\nindividuals, but affirmed the judgment as to the Tribe.\\n\\n\\n COUNSEL\\n\\nLester John Marston (argued), Rapport and Marston, Ukiah,\\nCalifornia, for Plaintiffs-Appellants.\\n\\nShaun M. Murphy (argued) and Katelyn K. Empey, Slovak\\nBaron Empey Murphy & Pinkey LLP, Palm Springs,\\nCalifornia; Michelle Blakemore and Miles Kowalski, San\\nBernardino County Counsel, San Bernardino, California; for\\nDefendants-Appellees.\\n\\n\\n OPINION\\n\\nHURWITZ, Circuit Judge:\\n\\n In 2015, San Bernardino County Sheriff’s Deputies cited\\nfour enrolled members of the Chemehuevi Indian Tribe for\\nviolating California regulatory traffic laws. Two of the\\nTribe’s members were cited on Section 36 of Township\\n5 North, Range 24 East (“Section 36”), a one square mile\\nplot the Tribe claims is part of its Reservation; two were\\ncited elsewhere on the Reservation.\\n\\f4 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\n It is undisputed that the Sheriff cannot enforce regulatory\\ntraffic laws in “Indian country.” See 18 U.S.C. § 1162;\\n28 U.S.C. § 1360. “Indian country” includes, but is not\\nlimited to, land within the boundaries of a reservation.\\n18 U.S.C. § 1151. The issues for decision today are\\n(1) whether the individual Tribe members and the Tribe can\\nchallenge the citations through a 42 U.S.C. § 1983 action;\\nand, if so, (2) whether Section 36 is Indian country. We hold\\nthat the individual plaintiffs, but not the Tribe, can challenge\\nthe citations under § 1983. And, we conclude that all the\\ncitations occurred within Indian country. We therefore\\nvacate the district court’s judgment dismissing the complaint\\nas to the individuals but affirm the judgment as to the Tribe.\\n\\n I. Background.\\n\\nA. Facts.\\n\\n Chelsea Lynn Bunim, Jasmine Sansoucie, Tommie\\nRobert Ochoa, and Naomi Lopez are enrolled members of\\nthe Chemehuevi Tribe. Each was stopped and cited by a San\\nBernardino County Sheriff’s Deputy for violating a\\nCalifornia regulatory traffic law. Deputy Sheriff Ronald\\nSindelar stopped and cited Bunim on Section 36 for driving\\nwithout a valid registration. Sindelar impounded Bunim’s\\ncar, leaving her alone on the roadside. Deputy Sindelar also\\nstopped Sansoucie on Section 36, citing her for driving with\\na suspended license.\\n\\n Deputy Sindelar cited Ochoa for driving without a valid\\nregistration and failing to provide evidence of financial\\nresponsibility. Sindelar had Ochoa’s car towed, leaving him\\nalone on the roadside. Deputy Sheriff J. Wagner cited Lopez\\nfor driving without a valid registration. Both of these\\ncitations were issued at locations that the parties agree are\\ninside the boundaries of the Chemehuevi Reservation.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 5\\n\\n Bunim, Sansoucie, Ochoa, Lopez, and the Tribe sued the\\nSheriff and the Deputies under 42 U.S.C. § 1983, alleging\\nviolations of various federal statutory and constitutional\\nrights. The complaint sought monetary damages, a\\ndeclaratory judgment, and injunctive relief. The defendants\\nargued that none of the plaintiffs’ claims was cognizable\\nunder § 1983. In addition, in response to the claims raised\\nby Bunim and Sansoucie, the defendants argued that Section\\n36 was outside the Reservation boundaries, and therefore\\nwithin the County’s regulatory jurisdiction.\\n\\nB. Procedural History.\\n\\n The district court initially entered a preliminary\\ninjunction prohibiting the defendants from “citing, arresting,\\nimpounding the vehicles of, and prosecuting Chemehuevi\\ntribal members for on-reservation violations” of California\\nregulatory vehicle laws, including violations occurring on\\nSection 36. The court determined there were “at least\\nserious questions going to the merits” of whether Section 36\\nwas “Indian country.”\\n\\n But, the court later granted summary judgment to the\\ndefendants, concluding that Section 36 was not part of the\\nChemehuevi Reservation and therefore not Indian country\\nunder 18 U.S.C. § 1151(a). In a motion to amend the\\njudgment, Ochoa and Lopez noted that they were not\\nticketed in Section 36. The district court denied the motion,\\nholding that even if the plaintiffs were cited on the\\nReservation, they failed to allege “a well-established\\nconstitutional violation for purposes of their section 1983\\nclaim.” The court reasoned that § 1983 “is concerned with\\nthe relationship between individuals and the state, not the\\ndistribution of power between state, federal, or tribal\\ngovernments,” and therefore neither the “right to be free of\\n\\f6 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nstate regulation” nor “the right to tribal government” is\\n“within the scope of section 1983.” 1\\n\\n The Tribe and the individual plaintiffs timely appealed.\\nWe have jurisdiction of that appeal under 28 U.S.C. § 1291\\nand review the summary judgment de novo. Parravano v.\\nBabbitt, 70 F.3d 539, 543 (9th Cir. 1995).\\n\\n II. Discussion.\\n\\nA. Is Section 36 Indian Country?\\n\\n We turn first to the question whether Section 36 is in the\\nChemehuevi Reservation, and thus Indian country under\\n18 U.S.C. § 1151(a). But it is important also to note at the\\noutset what issues are not before us. We need not—and do\\nnot—decide today who holds title to Section 36. Indian\\ncountry includes “all land within the limits of any Indian\\nreservation . . . notwithstanding the issuance of any patent.”\\n18 U.S.C. § 1151(a). “[A]djudicating reservation\\nboundaries is conceptually quite distinct from adjudicating\\ntitle to the same lands. One inquiry does not necessarily\\nhave anything in common with the other, as title and\\nreservation status are not congruent concepts in Indian law.”\\nNavajo Tribe of Indians v. New Mexico, 809 F.2d 1455, 1475\\n(10th Cir. 1987) (internal quotation marks and footnote\\nomitted); see Solem v. Bartlett, 465 U.S. 463, 466–68\\n(1984).\\n\\n\\n 1\\n The complaint also alleged that the citations “constitute[d] racial\\ndiscrimination in direct violation” of the Equal Protection Clause. The\\ndistrict court held that “the specter of racial animus” was not “sufficient\\nto create a triable issue of fact that Defendants violated” the Fourteenth\\nAmendment. The plaintiffs do not challenge that ruling on appeal.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 7\\n\\n Our inquiry as to the reservation status of Section 36\\nbegins in 1853. After California gained statehood, Congress\\nordered a survey of its public lands and granted the State title\\nto sections 16 and 36 of each township. 2 Act of March 3,\\n1853, ch. 145, 10 Stat. 244, 245–46. But, the 1853 Act\\nspecifically excluded from that grant any land “in the\\noccupation or possession of any Indian tribe.” Id. at 246–47.\\nThe Surveyor General approved a survey of the land at issue\\nin this case in 1895.\\n\\n While that survey was being conducted, Congress\\nordered the Secretary of the Interior “to select a reservation”\\nfor each California Mission Indian tribe. Mission Indian\\nRelief Act, ch. 65, 26 Stat. 712, 712 (1891). The\\nreservations were to “include, as far as practicable, the lands\\nand villages which have been in the actual occupation and\\npossession of said Indians.” Id. Although the Secretary was\\nalso instructed to “cause a patent to issue for each”\\nreservation, and thus transfer title to the land to the United\\nStates as trustee for the tribes, the Act provided that the\\nreservations would be “valid when approved by the\\n[Executive Branch].” Id. In 1905, Congress authorized the\\nSecretary “to investigate through an inspector . . . existing\\nconditions of the California Indians and to report to\\nCongress at the next session some plan to improve the\\nsame.” Act of March 3, 1905, ch. 1479, 33 Stat. 1048, 1058.\\n\\n Special Agent C.E. Kelsey was then dispatched to visit\\nthe Chemehuevi Tribe and identify territory for a\\nreservation. In 1907, Kelsey issued a report to the\\n\\n 2\\n See U.S. Geological Survey, The Public Land Survey System, The\\nNational Map Small Scale (Jan. 18, 2018),\\nhttps://nationalmap.gov/small_scale/a_plss.html (explaining township,\\nrange, and section designations).\\n\\f8 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nCommissioner of Indian Affairs, identifying land to be\\nincluded in the reservation. He specifically recommended\\nthat the reservation include the eastern half of Township\\n5 North, Range 24 East (“E. 1/2 of T. 5 N., R. 24 E.”)—\\nwhich contains Section 36. Kelsey noted that this land was\\nthe “present location” of the tribal members and that “there\\nis no question but they have occupied this land since\\nprimeval times.” The Commissioner forwarded Kelsey’s\\nrecommendation to the Secretary of the Interior.\\n\\n In an executive order (the “1907 Order”), the Secretary\\nthen “direct[ed] that the lands referred to” by Kelsey and the\\nCommissioner “be withdrawn from all form of settlement,”\\nand created the Chemehuevi Reservation. The Secretary\\nalso asked Congress “to authorize the addition of certain\\nlands to the Mission Indian Reservations.” Although\\nCongress did not act upon this proposed legislation, it\\nsubsequently recognized the existence of the Chemehuevi\\nReservation in the Parker Dam Act, ch. 522, 54 Stat. 744\\n(1940).\\n\\n It is clear that a Chemehuevi Reservation was validly\\nestablished by the Secretary’s 1907 Order, notwithstanding\\nthe absence of subsequent Congressional approval. Indeed,\\nthe Supreme Court has expressly so recognized:\\n\\n Congress and the Executive have ever since\\n recognized these as Indian Reservations. . . .\\n They have been uniformly and universally\\n treated as reservations by map makers,\\n surveyors, and the public. We can give but\\n short shrift at this late date to the argument\\n that the reservations . . . are invalid because\\n they were originally set apart by the\\n Executive.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 9\\n\\nArizona v. California, 373 U.S. 546, 598 (1963); see also id.\\nat 596 & n.100. 3\\n\\n The defendants argue that the 1907 Order was invalid ab\\ninitio because Section 36 had already been deeded to\\nCalifornia. The factual premise of that argument, however,\\nis subject to question. The 1853 Act excluded any land “in\\nthe occupation or possession of any Indian tribe,” 10 Stat.\\nat 246–47, and the Kelsey survey, adopted by the Secretary,\\ndocuments that Section 36 falls in that exception. The\\ndistrict court erred in excluding the Kelsey report as hearsay.\\nIt is plainly admissible as an ancient document, Fed. R. Evid.\\n803(16), which may contain multiple levels of hearsay.\\n30B Charles Alan Wright & Arthur R. Miller, Federal\\nPractice & Procedure § 6935 (2018 ed.) (“[E]xclusion of\\nstatements in qualifying ancient documents on the grounds\\nthat the author lacked firsthand knowledge, or (relatedly)\\nthat the document contains hearsay-within-hearsay should\\nbe rare.”). Review of historical documents is typical—\\nindeed often necessary—in cases involving the boundaries\\nof Indian reservations. See, e.g., Idaho v. United States,\\n533 U.S. 262, 265–71 (2001).\\n\\n But, as noted, we need not today decide the extent of the\\n1853 land grant. “[E]xecutive orders must be liberally\\nconstrued in favor of establishing Indian rights,”\\nConfederated Tribes of Chehalis Indian Reservation v.\\n\\n 3\\n The defendants cite Arizona v. California, 460 U.S. 605, 636 n.26\\n(1983), for the proposition that “the secretarial orders do not constitute\\n‘final determinations.’” But, “the secretarial orders” at issue in that\\nlitigation were issued in the 1960s and 1970s. See id. at 631–32. In 1919\\nand 1927, Congress “prohibited future changes in Indian reservations by\\nexecutive order.” United States v. S. Pac. Transp. Co., 543 F.2d 676,\\n686 & n.15 (9th Cir. 1976). That prohibition plainly does not affect the\\n1907 Order.\\n\\f10 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nWashington, 96 F.3d 334, 340 (9th Cir. 1996), and are\\n“interpreted as the Indians would have understood them,”\\nParravano, 70 F.3d at 544. Given the language of the 1853\\nAct, the Kelsey report identifying Section 36 as land\\noccupied historically by Indians, and the express inclusion\\nof Section 36 in the 1907 Order, the Chemehuevi Tribe (and\\nindeed, the Secretary of the Interior) surely understood\\nSection 36 to be within the Reservation.\\n\\n Nor can we conclude that the boundaries of the\\nReservation as established in the 1907 Order were later\\ndiminished. “We do not lightly infer diminishment of\\nreservations.” Confederated Tribes of Chehalis, 96 F.3d\\nat 343–44. After 1927, Congress prohibited any change to\\nthe boundaries of existing executive-order reservations\\nexcept by Congressional act. Act of March 3, 1927, ch. 299,\\n§ 4, 44 Stat. 1347 (codified at 25 U.S.C. § 398d); see S. Pac.\\nTransp., 543 F.2d at 686 & n.15. There is no such act\\nremoving Section 36 from the Chemehuevi Reservation.\\n\\n The defendants also rely on a patent issued to the Tribe\\nby the Bureau of Land Management in 2010, which excluded\\n“[t]hose lands granted to the State of California . . . on July\\n10, 1895”—the date on which the government survey was\\nfinalized—“located in . . . sec. 36, T. 5 N., R. 24 E.” But, as\\nnoted above, we do not today adjudicate title. More\\nimportantly, because the 2010 patent was issued over a\\ncentury after the Reservation was established, it provides no\\nevidence of the intent of the Executive or the understanding\\nof the Tribe in 1907. Nor can it, nor does it purport to,\\ndiminish the Chemehuevi Reservation. The patent cites the\\n1907 Order, then grants some of the land covered by that\\norder to the Tribe in trust—an issue of ownership. It is silent\\nas to the reservation status of any land excluded from the\\npatent.\\n\\f CHEMEHUEVI INDIAN TRIBE V. MCMAHON 11\\n\\n We therefore conclude that Section 36 is within the\\nChemehuevi Reservation and hence “Indian country” under\\n18 U.S.C. § 1151(a).\\n\\nB. Can the Plaintiffs Sue Under § 1983?\\n\\n California cannot enforce state law that regulates—but\\ndoes not prohibit—tribal members’ conduct inside a\\nreservation. Confederated Tribes of Colville Reservation v.\\nWashington, 938 F.2d 146, 147 (9th Cir. 1991) (citing\\n18 U.S.C. § 1162; 28 U.S.C. § 1360). The defendants\\nconcede that the citations at issue involved regulatory laws\\nand therefore could not be issued against enrolled members\\nof the Tribe within the boundaries of the Reservation. See\\nid. at 148. But, they argue that even such citations cannot be\\nthe subject of a § 1983 action.\\n\\n We disagree. Section 1983 allows any “person” to sue\\nfor the “deprivation of any rights, privileges, or immunities\\nsecured by the Constitution and laws.” 42 U.S.C. § 1983.\\nBecause § 1983 “was designed to secure private rights\\nagainst government encroachment,” Inyo Cty. v. Paiute-\\nShoshone Indians of the Bishop Cmty. of the Bishop Colony,\\n538 U.S. 701, 712 (2003), tribal members can use it to\\nvindicate their “individual rights,” but not the tribe’s\\n“communal rights,” Skokomish Indian Tribe v. United\\nStates, 410 F.3d 506, 515–16 (9th Cir. 2005) (en banc). And,\\n“traditional section 1983 suit[s]”—for example, those\\nchallenging an arrest on tribal land—seek to vindicate an\\n“individual right.” See id. at 516 n.8 (citing Romero v.\\nKitsap Cty., 931 F.2d 624, 627 n.5 (9th Cir. 1991)).\\n\\n Bunim, Sansoucie, Ochoa, and Lopez’s claims are\\n“traditional” § 1983 suits. Each was stopped and detained\\nby a San Bernardino County Deputy; some had their vehicles\\nseized. They contend that their detentions and citations\\n\\f12 CHEMEHUEVI INDIAN TRIBE V. MCMAHON\\n\\nviolated the Constitution and federal statutes. They have a\\ncause of action under § 1983 against the defendants.\\n\\n The Tribe, however, does not have a § 1983 claim. An\\nIndian tribe “may not sue under § 1983 to vindicate” a\\n“sovereign right,” such as its right to be free of state\\nregulation and control. Inyo Cty., 538 U.S. at 712. Nor can\\nthe Tribe assert its members’ individual rights as parens\\npatriae in a § 1983 action. To assert parens patriae\\nstanding, the Tribe would have to “articulate an interest apart\\nfrom the interests of particular private parties,” i.e., “be more\\nthan a nominal party,” and “express a quasi-sovereign\\ninterest.” Missouri ex rel. Koster v. Harris, 847 F.3d 646,\\n651 (9th Cir. 2017). That requirement is inconsistent with a\\n§ 1983 action: quasi-sovereign interests are not individual\\nrights.\\n\\n III. Conclusion.\\n\\n The Chemehuevi Reservation, as established by the 1907\\nOrder, includes Section 36. Section 36 is therefore Indian\\ncountry, and San Bernardino County does not have\\njurisdiction to enforce California regulatory laws within it.\\nThe individual plaintiffs may bring § 1983 claims against the\\ndefendants. The Tribe, however, cannot assert its sovereign\\nrights under that statute. 4\\n\\n AFFIRMED in part, VACATED and REMANDED\\nin part. Each party shall bear its own costs.\\n\\n\\n\\n\\n 4\\n We take no position on any defenses, including immunity, the\\ndefendants might have to the claims raised by the individuals.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4082,"cells":{"id":{"kind":"number","value":4649767,"string":"4,649,767"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Michael Murray v. Mayo Clinic"},"decision_date":{"kind":"string","value":"2019-08-20"},"docket_number":{"kind":"string","value":"17-16803"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Ronald M. Gould and Sandra S. Ikuta, Circuit Judges, and Benita Y. Pearson, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PEARSON, District Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMICHAEL J. MURRAY, M.D. - a No. 17-16803\\nmarried man,\\n Plaintiff-Appellant, D.C. No.\\n 2:14-cv-01314-\\n v. SPL\\n\\nMAYO CLINIC, a Minnesota nonprofit\\ncorporation; MAYO CLINIC ARIZONA, OPINION\\nan Arizona nonprofit corporation;\\nWYATT DECKER, M.D. - husband;\\nGEORGIANNA DECKER, wife; LOIS\\nKRAHN, M.D. - wife; ERIC GORDON,\\nM.D. - husband; TERRENCE\\nTRENTMAN, M.D. - husband;\\nLARALEE TRENTMAN, wife; WILLIAM\\nSTONE, M.D. - husband; MAREE\\nSTONE, wife; DAVID ROSENFELD,\\nM.D. - husband; MELISSA\\nROSENFELD, M.D. - wife; ROSHANAK\\nDIDEHBAN, a single woman,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Steven Paul Logan, District Judge, Presiding\\n\\n Argued and Submitted June 13, 2019\\n San Francisco, California\\n\\f2 MURRAY V. MAYO CLINIC\\n\\n Filed August 20, 2019\\n\\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\\n Judges, and Benita Y. Pearson, * District Judge.\\n\\n Opinion by Judge Pearson\\n\\n\\n SUMMARY **\\n\\n\\n Employment Discrimination\\n\\n The panel affirmed the district court’s judgment, after a\\njury trial, in favor of the defendants in an employment\\ndiscrimination action under Title I of the Americans with\\nDisabilities Act.\\n\\n The panel held that the district court correctly instructed\\nthe jury to apply a but for causation standard, rather than a\\nmotivating factor standard. The panel concluded that Head\\nv. Glacier Northwest, Inc., 413 F.3d 1053 (9th Cir. 2005),\\nholding that ADA discrimination claims are evaluated under\\na motivating factor causation standard, is no longer good law\\nbecause its reasoning is clearly irreconcilable with the\\nSupreme Court’s rulings in Gross v. FBL Fin. Servs., Inc.,\\n557 U.S. 167 (2009), and Univ. of Texas Southwestern Med.\\nCtr. v. Nassar, 570 U.S. 338 (2013). Agreeing with other\\ncircuits, the panel held that an ADA discrimination plaintiff\\n\\n *\\n The Honorable Benita Y. Pearson, United States District Judge for\\nthe Northern District of Ohio, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f MURRAY V. MAYO CLINIC 3\\n\\nbringing a claim under 42 U.S.C. § 12112 must show that\\nthe adverse employment action would not have occurred but\\nfor the disability.\\n\\n The panel addressed other issues in a simultaneously\\nfiled memorandum disposition.\\n\\n\\n COUNSEL\\n\\nRoger L. Cohen (argued), Jaburg Wilk P.C., Phoenix,\\nArizona; Scott A. Blaney, Blaney Law PLLC, Phoenix,\\nArizona; for Plaintiff-Appellant.\\n\\nJohn F. Lomax, Jr. (argued) and Kelly Kszywienski, Snell &\\nWilmer L.L.P., Phoenix, Arizona, for Defendants-\\nAppellees.\\n\\n\\n OPINION\\n\\nPEARSON, District Judge:\\n\\n Plaintiff Michael Murray appeals the district court’s\\ninstruction to the jury on his claim under the Americans with\\nDisabilities Act (“ADA”), requiring him to prove that he was\\ndischarged because of his disability. Murray claims that our\\ndecision in Head v. Glacier Northwest, Inc., 413 F.3d 1053\\n(9th Cir. 2005), holding ADA discrimination claims are\\n\\f4 MURRAY V. MAYO CLINIC\\n\\nevaluated under a motivating factor causation standard,\\nremains good law. Because it is not, we affirm. 1\\n\\n I.\\n\\n Dr. Murray filed suit against Mayo Clinic, Mayo Clinic\\nArizona, Drs. Wyatt Decker, Lois Krahn, Terrence\\nTrentman, William Stone, and David Rosenfeld, and\\nOperations Administrator Roshanak Didehban. In\\nanticipation of trial, the parties submitted joint proposed jury\\ninstructions. The parties disagreed whether Murray’s ADA\\ndiscrimination claim should be tried under a but-for\\ncausation standard or a motivating factor causation standard.\\nMurray argued that our decision in Head required him to\\nshow only that the defendants’ belief that he had a disability\\nwas a motivating factor in their adverse employment\\ndecision. He accordingly requested the following\\ninstruction:\\n\\n As to Dr. Murray’s claim that his disability\\n was the reason for Mayo Clinic Arizona’s\\n decision to discharge him, Dr. Murray has the\\n burden of proving the following evidence by\\n a preponderance of the evidence:\\n\\n ...\\n\\n 3. Dr. Murray was discharged because\\n Defendants regarded him as disabled, which\\n means that Defendants’ belief that Plaintiff\\n\\n\\n 1\\n In a memorandum disposition filed simultaneously with this\\nopinion, we affirm the district court as to all other issues raised by\\nMurray.\\n\\f MURRAY V. MAYO CLINIC 5\\n\\n had a disability was a motivating factor in\\n Defendants’ decision to terminate him.\\n\\nThe district court instead instructed the jury to apply a but-\\nfor causation standard to Murray’s ADA claim. The\\ninstruction provided that Murray must prove he was\\ndischarged because of his disability:\\n\\n As to Dr. Murray’s claim that his disability\\n was the reason for Mayo Clinic Arizona’s\\n decision to discharge him, Dr. Murray has the\\n burden of proving the following evidence by\\n a preponderance of the evidence:\\n\\n ...\\n\\n 3. Dr. Murray was discharged because of his\\n disability.\\n\\nIn denying Murray’s motion for reconsideration, the district\\ncourt found that the Supreme Court’s rulings in Gross v. FBL\\nFinancial Services, Inc., 557 U.S. 167 (2009), and\\nUniversity of Texas Southwestern Medical Center v. Nassar,\\n570 U.S. 338 (2013), abrogated our reasoning in Head. The\\ndistrict court concluded that the but-for causation standard\\napplied.\\n\\n At trial, the jury returned a verdict for Defendants on all\\nclaims. Following entry of judgment, Murray timely filed a\\nnotice of appeal.\\n\\n II.\\n\\n “A district court’s formulation of the jury instructions is\\nreviewed for ‘abuse of discretion.’ If, however, ‘the\\ninstructions are challenged as a misstatement of the law, they\\n\\f6 MURRAY V. MAYO CLINIC\\n\\nare then reviewed de novo.’” Duran v. City of Maywood,\\n221 F.3d 1127, 1130 (9th Cir. 2000) (per curiam) (citation\\nomitted) (quoting Gilbrook v. City of Westminster, 177 F.3d\\n839, 860 (9th Cir. 1999), as amended on denial of reh’g\\n(July 15, 1999)). Jury instructions must fairly and\\nadequately cover the issues presented and must not be\\nmisleading. Gantt v. City of Los Angeles, 717 F.3d 702, 706\\n(9th Cir. 2013).\\n\\n A.\\n\\n Title I of the ADA provides that “[n]o covered entity\\nshall discriminate against a qualified individual on the basis\\nof disability in regard to job application procedures, the\\nhiring, advancement, or discharge of employees, employee\\ncompensation, job training, and other terms, conditions, and\\nprivileges of employment.” 42 U.S.C. § 12112(a) (emphasis\\nadded).\\n\\n Title I of the ADA also contains an enforcement\\nprovision, which cross-references specific portions of\\nTitle VII:\\n\\n The powers, remedies, and procedures set\\n forth in sections 2000e-4, 2000e-5, 2000e-6,\\n 2000e-8, and 2000e-9 of [Title VII] shall be\\n the powers, remedies, and procedures this\\n subchapter provides to . . . any person\\n alleging discrimination on the basis of\\n disability in violation of any provision of this\\n chapter . . . .\\n\\n42 U.S.C. § 12117(a). Of the cross-referenced sections, only\\n§ 2000e-5 references a causation standard. Specifically, that\\nsection provides: “[o]n a claim in which an individual proves\\na violation under section 2000e-2(m) of this title and a\\n\\f MURRAY V. MAYO CLINIC 7\\n\\nrespondent demonstrates that the respondent would have\\ntaken the same action in the absence of the impermissible\\nmotivating factor,” the court may award only limited relief.\\n42 U.S.C. § 2000e-5(g)(2)(B). Section 2000e-2(m), in turn,\\nprovides that “an unlawful employment practice is\\nestablished when the complaining party demonstrates that\\nrace, color, religion, sex, or national origin was a motivating\\nfactor for any employment practice, even though other\\nfactors also motivated the practice.”\\n\\n B.\\n\\n We first analyzed the standard for causation in a Title I\\nADA discrimination action in Head v. Glacier Northwest,\\nInc. In that case, we addressed “whether the ADA’s use of\\nthe causal language ‘because of,’ ‘by reason of,’ and\\n‘because’ means that discriminatory and retaliatory conduct\\nis proscribed only if it was solely because of, solely by reason\\nof, or solely because an employee was disabled or requested\\nan accommodation.” Head, 413 F.3d at 1063–64. We\\nconcluded that, under the “plain language of the ADA . . .\\n‘solely’ [was] not the appropriate causal standard under any\\nof the ADA’s liability provisions.” Id. at 1065.\\n\\n Considering whether the ADA instead requires but-for\\ncausation, or merely a showing that the disability was a\\nmotivating factor of the discrimination, we joined seven\\nother circuits in concluding that “a ‘motivating factor’\\nstandard [was] most consistent with the plain language of the\\nstatute and the purposes of the ADA.” Id. We thus held “the\\nADA outlaws adverse employment decisions motivated,\\neven in part, by animus based on a plaintiff’s disability or\\nrequest for an accommodation—a motivating factor\\nstandard.” Id.\\n\\f8 MURRAY V. MAYO CLINIC\\n\\n In so holding, we relied in part on the reasoning of our\\nsister circuits. See id. at 1065 n.63. The Fifth and Eighth\\nCircuits had held the motivating factor standard applied to\\nthe ADA by virtue of the ADA’s incorporation in § 12117\\nof Title VII’s remedies in § 2000e-5. 2 See Buchanan v. City\\nof San Antonio, 85 F.3d 196, 200 (5th Cir. 1996); Pedigo v.\\nP.A.M. Transp., Inc., 60 F.3d 1300, 1301 (8th Cir. 1995). 3\\n\\n The Second and Seventh Circuits had concluded that\\nADA discrimination claims, like Title VII discrimination\\nclaims, only required a showing that discrimination\\nmotivated an employer’s adverse employment action. This\\nis because ADA and Title VII, at the time, both used the\\nwords “because of” to indicate causation, suggesting\\nCongress intended the statutes to employ the same causation\\nstandard. See Parker v. Columbia Pictures Indus., 204 F.3d\\n326, 337 (2d Cir. 2000), overruled by Natofsky v. City of\\nNew York, 921 F.3d 337 (2d Cir. 2019); Foster v. Arthur\\nAndersen, LLP, 168 F.3d 1029, 1033 (7th Cir. 1999),\\noverruled by Serwatka v. Rockwell Automation Inc.,\\n591 F.3d 957 (7th Cir. 2010). Both courts acknowledged\\nthat, although § 2000e-2(m) is not, by its terms, specifically\\napplicable to ADA cases, Congress intended the mixed-\\nmotive framework to apply to ADA claims. Parker,\\n204 F.3d at 337; Foster, 168 F.3d at 1033.\\n\\n\\n 2\\n Similarly, the Fourth Circuit had found the motivating factor\\nstandard applied in a Title II ADA case through the ADA’s incorporation\\nin 42 U.S.C. § 12133 of the remedies set forth in 29 U.S.C. § 794a,\\nwhich, in turn, incorporated the remedies in 42 U.S.C. § 2000e-5. Baird\\nex rel. Baird v. Rose, 192 F.3d 462, 470 (4th Cir. 1999).\\n 3\\n The First Circuit relied on Pedigo, without additional analysis, in\\napplying the motivating factor standard. Katz v. City Metal Co., 87 F.3d\\n26, 33 (1st Cir. 1996).\\n\\f MURRAY V. MAYO CLINIC 9\\n\\n C.\\n\\n Four years after our decision in Head, the Supreme Court\\ndecided Gross v. FBL Financial Services, Inc. The Court\\nheld that the Age Discrimination in Employment Act\\n(“ADEA”)—which makes it unlawful for an employer to\\ndischarge or discriminate against any individual “because of\\nsuch an individual’s age”—requires the plaintiff to “prove\\nthat age was the ‘but-for’ cause of the employer’s adverse\\ndecision.” Gross, 557 U.S. at 177–78. In so doing, the Court\\ndeclined to extend 4 the “motivating factor” standard of\\ncausation to employment discrimination cases brought under\\nthe ADEA. Id. Four years after Gross, the Supreme Court\\nin Nassar again declined to extend the motivating factor\\nstandard, this time to Title VII retaliation claims. 570 U.S.\\nat 362–63.\\n\\n Against this backdrop, “circuits have retreated from the\\nmotivating factor standard of causation in ADA cases.”\\nBukiri v. Lynch, 648 F. App’x 729, 731 n.1 (9th Cir. 2016)\\n(collecting cases). We have not yet decided whether Gross\\nand Nassar have “eroded Head’s vitality.” Mendoza v.\\nRoman Catholic Archbishop of L.A., 824 F.3d 1148, 1150\\nn.1 (9th Cir. 2016) (per curiam). We do so now.\\n\\n D.\\n\\n Murray contends that the motivating factor standard\\napplies because we are bound by our decision in Head. We\\ndisagree.\\n\\n\\n 4\\n The Supreme Court recognized the “motivating standard” of\\ncausation as the appropriate standard for employment discrimination\\nactions brought under Title VII of the Civil Rights Act of 1964. Price\\nWaterhouse v. Hopkins, 490 U.S. 228, 258 (1989) (plurality opinion).\\n\\f10 MURRAY V. MAYO CLINIC\\n\\n Generally, a three-judge panel may not overrule a prior\\ndecision of the court. Miller v. Gammie, 335 F.3d 889, 899\\n(9th Cir. 2003) (en banc). If, however, “an intervening\\nSupreme Court decision undermines an existing precedent\\nof the Ninth Circuit, and both cases are closely on point,” the\\nthree-judge panel may then overrule prior circuit authority.\\nId. (quoting Galbraith v. Cty. of Santa Clara, 307 F.3d 1119,\\n1123 (9th Cir. 2002)). The issue decided by the higher court\\nneed not be identical. Id. at 900. The appropriate test is\\nwhether the higher court “undercut the theory or reasoning\\nunderlying the prior circuit precedent in such a way that the\\ncases are clearly irreconcilable.” Id.\\n\\n “The ‘clearly irreconcilable’ requirement is ‘a high\\nstandard.’” United States v. Robertson, 875 F.3d 1281, 1291\\n(9th Cir. 2017) (quoting Rodriguez v. AT & T Mobility Servs.\\nLLC, 728 F.3d 975, 979 (9th Cir. 2013)). “It is not enough\\nfor there to be ‘some tension’ between the intervening higher\\nauthority and prior circuit precedent, or for the intervening\\nhigher authority to ‘cast doubt’ on the prior circuit\\nprecedent.” Lair v. Bullock, 697 F.3d 1200, 1207 (9th Cir.\\n2012) (internal citation omitted) (quoting United States v.\\nOrm Hieng, 679 F.3d 1131, 1140–41 (9th Cir. 2012), and\\nUnited States v. Delgado-Ramos, 635 F.3d 1237, 1239 (9th\\nCir. 2011) (per curiam)). If the court can apply prior circuit\\nprecedent without running afoul of the intervening authority,\\nit must do so. Id.\\n\\n Because Head’s reasoning is clearly irreconcilable with\\nGross and Nassar, we overrule Head’s holding that a\\nplaintiff bringing a discrimination claim under Title I of the\\nADA need show only that a disability was a motivating\\nfactor of the adverse employment action. We hold instead\\nthat an ADA discrimination plaintiff bringing a claim under\\n\\f MURRAY V. MAYO CLINIC 11\\n\\n42 U.S.C. § 12112 must show that the adverse employment\\naction would not have occurred but for the disability.\\n\\n In Head, we relied on the reasoning of our sister circuits\\nand our existing precedent in finding that a motivating factor\\nwas most consistent with the ADA’s plain language and\\npurpose. Head, 413 F.3d at 1065 & nn.63–64. Our prior\\nprecedent, however, provides no further analysis of the text\\nor purpose of the ADA in support of applying a motivating\\nfactor causation standard. See Hernandez v. Hughes Missile\\nSys. Co., 362 F.3d 564, 568 (9th Cir. 2004); Snead v. Metro.\\nProp. & Cas. Ins. Co., 237 F.3d 1080, 1094 (9th Cir. 2001). 5\\nAdditionally, Gross and Nassar undercut the reasoning set\\nforth by our sister circuits.\\n\\n Gross held that the ADEA, which also used “because of”\\nto indicate causation, did not permit mixed-motive claims\\nbecause “the ADEA’s text does not provide that a plaintiff\\nmay establish discrimination by showing that age was\\nsimply a motivating factor.” 557 U.S. at 174. The Court\\nrejected the argument that Title VII decisions governed\\ninterpretation of the ADEA on the basis that the two statutes\\nwere distinguishable. Id. (“[W]e ‘must be careful not to\\napply the rules applicable under one statute to a different\\nstatute without careful and critical examination.’” (quoting\\n\\n\\n 5\\n Head noted Hernandez’s characterization of the burden on an\\nADA plaintiff as “proving that ‘disability actually played a role in the\\nemployer’s decisionmaking process and had a determinative influence\\non the outcome.’” Head, 413 F.3d at 1065 (quoting Hernandez, 362 F.3d\\nat 568 (emphasis in Head)). Head also observed Snead’s statement that\\na plaintiff must demonstrate that “a discriminatory reason more likely\\nmotivated the employer.” Head, 413 F.3d at 1065 (quoting Snead,\\n237 F.3d at 1094 (emphasis in Head)). Neither statement requires a\\nmotivating factor standard.\\n\\f12 MURRAY V. MAYO CLINIC\\n\\nFed. Exp. Corp. v. Holowecki, 552 U.S. 389, 393 (2008)).\\nThe Court explained,\\n\\n Unlike Title VII, which has been amended to\\n explicitly authorize discrimination claims\\n where an improper consideration was a\\n ‘motivating factor’ for the adverse action, the\\n ADEA does not provide that a plaintiff may\\n establish discrimination by showing that age\\n was simply a motivating factor. Moreover,\\n Congress neglected to add such a provision to\\n the ADEA when it added §§ 2000e-2(m) and\\n 2000e-5(g)(2)(B) to Title VII, even though it\\n contemporaneously amended the ADEA in\\n several ways.\\n\\nId. (citations omitted).\\n\\n Gross’s reasoning directly contradicts the textual\\nreasoning Head and other courts applied to conclude that\\nTitle VII’s motivating factor standard applied to ADA\\nclaims. See Parker, 204 F.3d at 337; Foster, 168 F.3d\\nat 1033. Like the ADEA, and unlike Title VII, the ADA\\ndoes not contain any explicit “motivating factor” language.\\nSee Gross, 557 U.S. at 174. Rather, Title I of the ADA\\nprovides that a plaintiff must show discrimination “on the\\nbasis of disability.” 42 U.S.C. § 12112(a). Under Gross, the\\nphrase “on the basis of disability” indicates but-for\\ncausation. Gross, 557 U.S. at 176; see also Nassar, 570 U.S.\\nat 350 (explaining Gross’s holding that “because of,” “by\\n\\f MURRAY V. MAYO CLINIC 13\\n\\nreason of,” “on account of,” and “based on” all indicate a\\nbut-for causal relationship). 6\\n\\n Nassar’s reasoning likewise directly undercuts the\\nreasoning of courts that relied on the ADA’s incorporation\\nin § 12117 of § 2000e-5. See Buchanan, 85 F.3d at 200;\\nPedigo, 60 F.3d at 1301; cf. Baird ex rel. Baird v. Rose,\\n192 F.3d 462, 470 (4th Cir. 1999) (ADA Title II’s\\nincorporation of § 2000e-5). Nassar rejected the argument\\nthat § 2000e-2(m), Title VII’s motivating factor causation\\nprovision, applies to Title VII retaliation claims. Nassar,\\n570 U.S. at 353. The Court emphasized that “the text of the\\nmotivating-factor provision, while it begins by referring to\\n‘unlawful employment practices,’ then proceeds to address\\nonly five of the seven prohibited discriminatory actions—\\nactions based on the employee’s status, i.e., race, color,\\nreligion, sex, and national origin.” Id. According to Nassar,\\nthe plain language of § 2000e-2(m) barred its application to\\nretaliation claims, and “it would be improper to conclude\\nthat what Congress omitted from the statute is nevertheless\\nwithin its scope.” Id.\\n\\n The same logic applies to Title I ADA discrimination\\nclaims. Relief under § 2000e-5(g)(2)(B) is available only if\\n\\n 6\\n Title I of the ADA was amended in 2008 to prohibit discrimination\\n“on the basis of” disability, rather than “because of” disability. We find\\nno meaningful textual difference in the two phrases with respect to\\ncausation. The Second and Fourth Circuits likewise found no\\nmeaningful textual difference between the two standards and found\\nnothing in the legislative history suggesting Congress intended to modify\\nthe ADA’s standard for causation. Natofsky v. City of New York,\\n921 F.3d 337, 349–50 (2d Cir. 2019); Gentry v. E.W. Partners Club\\nMgmt. Co., 816 F.3d 228, 234 (4th Cir. 2016) (noting that the ADA\\namendment was enacted before Gross, and was therefore not in response\\nto Gross’s causation analysis).\\n\\f14 MURRAY V. MAYO CLINIC\\n\\nthe plaintiff proves a violation under § 2000e-2(m). Section\\n2000e-2(m) narrowly prohibits the consideration of race,\\ncolor, religion, sex, or national origin as a motivating factor\\nfor any employment practice. It does not prohibit the\\nconsideration of disability. Congress’s express listing of\\nthese status-based considerations under § 2000e-2(m) is best\\nunderstood as an exclusion of all other considerations. See,\\ne.g., Silvers v. Sony Pictures Entm’t, Inc., 402 F.3d 881, 885\\n(9th Cir. 2005) (“The doctrine of expressio unius est exclusio\\nalterius ‘as applied to statutory interpretation creates a\\npresumption that when a statute designates certain persons,\\nthings, or manners of operation, all omissions should be\\nunderstood as exclusions.’” (quoting Boudette v. Barnette,\\n923 F.2d 754, 756–57 (9th Cir. 1991))). Section 2000e-\\n2(m), by its plain language, is inapplicable to claims of\\ndisability discrimination.\\n\\n Because Head’s reasoning—whether based on the\\nADA’s cross-reference to § 2000e-5(g)(2)(B) or on the\\nADA’s text—is irreconcilable with subsequent Supreme\\nCourt precedent, it cannot stand.\\n\\n III.\\n\\n Our decision comports with the decisions of all of our\\nsister circuits that have considered this question after Gross\\nand Nassar. The Second, Fourth, and Seventh Circuits\\nfound the Supreme Court’s intervening jurisprudence to be\\ndispositive of the issue. See Natofsky, 921 F.3d at 348\\n(“Gross and Nassar dictate our decision here.”); Gentry v.\\nE.W. Partners Club Mgmt. Co., 816 F.3d 228, 234 (4th Cir.\\n2016) (“The Supreme Court’s analysis in Gross dictates the\\noutcome here.”); Serwatka, 591 F.3d at 963 (“But in view of\\nthe Court’s intervening decision in Gross, it is clear that the\\ndistrict court’s decision . . . cannot be sustained.”). The\\nSixth Circuit, following en banc review, similarly held that\\n\\f MURRAY V. MAYO CLINIC 15\\n\\nGross’s reasoning was controlling. Lewis v. Humboldt\\nAcquisition Corp., 681 F.3d 312, 318 (6th Cir. 2012) (en\\nbanc) (“[Gross’s] rationale applies with equal force to the\\nADA.”).\\n\\n We agree. Gross and Nassar undermine Head’s\\nreasoning such that the cases are clearly irreconcilable. We\\njoin our sister circuits in holding that ADA discrimination\\nclaims under Title I must be evaluated under a but-for\\ncausation standard.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4427020/\", \"author_raw\": \"PEARSON, District Judge:\"}]}"},"judge1":{"kind":"string","value":"RONALD M GOULD"},"judge2":{"kind":"string","value":"SANDRA S IKUTA"},"judge3":{"kind":"string","value":"BENITA Y PEARSON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649767/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Michael Murray v. Mayo Clinic\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-20\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMICHAEL J. MURRAY, M.D. - a No. 17-16803\nmarried man,\n Plaintiff-Appellant, D.C. No.\n 2:14-cv-01314-\n v. SPL\n\nMAYO CLINIC, a Minnesota nonprofit\ncorporation; MAYO CLINIC ARIZONA, OPINION\nan Arizona nonprofit corporation;\nWYATT DECKER, M.D. - husband;\nGEORGIANNA DECKER, wife; LOIS\nKRAHN, M.D. - wife; ERIC GORDON,\nM.D. - husband; TERRENCE\nTRENTMAN, M.D. - husband;\nLARALEE TRENTMAN, wife; WILLIAM\nSTONE, M.D. - husband; MAREE\nSTONE, wife; DAVID ROSENFELD,\nM.D. - husband; MELISSA\nROSENFELD, M.D. - wife; ROSHANAK\nDIDEHBAN, a single woman,\n Defendants-Appellees.\n\n Appeal from the United States District Court\n for the District of Arizona\n Steven Paul Logan, District Judge, Presiding\n\n Argued and Submitted June 13, 2019\n San Francisco, California\n\f2 MURRAY V. MAYO CLINIC\n\n Filed August 20, 2019\n\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\n Judges, and Benita Y. Pearson, * District Judge.\n\n Opinion by Judge Pearson\n\n\n SUMMARY **\n\n\n Employment Discrimination\n\n The panel affirmed the district court’s judgment, after a\njury trial, in favor of the defendants in an employment\ndiscrimination action under Title I of the Americans with\nDisabilities Act.\n\n The panel held that the district court correctly instructed\nthe jury to apply a but for causation standard, rather than a\nmotivating factor standard. The panel concluded that Head\nv. Glacier Northwest, Inc., 413 F.3d 1053 (9th Cir. 2005),\nholding that ADA discrimination claims are evaluated under\na motivating factor causation standard, is no longer good law\nbecause its reasoning is clearly irreconcilable with the\nSupreme Court’s rulings in Gross v. FBL Fin. Servs., Inc.,\n557 U.S. 167 (2009), and Univ. of Texas Southwestern Med.\nCtr. v. Nassar, 570 U.S. 338 (2013). Agreeing with other\ncircuits, the panel held that an ADA discrimination plaintiff\n\n *\n The Honorable Benita Y. Pearson, United States District Judge for\nthe Northern District of Ohio, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f MURRAY V. MAYO CLINIC 3\n\nbringing a claim under 42 U.S.C. § 12112 must show that\nthe adverse employment action would not have occurred but\nfor the disability.\n\n The panel addressed other issues in a simultaneously\nfiled memorandum disposition.\n\n\n COUNSEL\n\nRoger L. Cohen (argued), Jaburg Wilk P.C., Phoenix,\nArizona; Scott A. Blaney, Blaney Law PLLC, Phoenix,\nArizona; for Plaintiff-Appellant.\n\nJohn F. Lomax, Jr. (argued) and Kelly Kszywienski, Snell &\nWilmer L.L.P., Phoenix, Arizona, for Defendants-\nAppellees.\n\n\n OPINION\n\nPEARSON, District Judge:\n\n Plaintiff Michael Murray appeals the district court’s\ninstruction to the jury on his claim under the Americans with\nDisabilities Act (“ADA”), requiring him to prove that he was\ndischarged because of his disability. Murray claims that our\ndecision in Head v. Glacier Northwest, Inc., 413 F.3d 1053\n(9th Cir. 2005), holding ADA discrimination claims are\n\f4 MURRAY V. MAYO CLINIC\n\nevaluated under a motivating factor causation standard,\nremains good law. Because it is not, we affirm. 1\n\n I.\n\n Dr. Murray filed suit against Mayo Clinic, Mayo Clinic\nArizona, Drs. Wyatt Decker, Lois Krahn, Terrence\nTrentman, William Stone, and David Rosenfeld, and\nOperations Administrator Roshanak Didehban. In\nanticipation of trial, the parties submitted joint proposed jury\ninstructions. The parties disagreed whether Murray’s ADA\ndiscrimination claim should be tried under a but-for\ncausation standard or a motivating factor causation standard.\nMurray argued that our decision in Head required him to\nshow only that the defendants’ belief that he had a disability\nwas a motivating factor in their adverse employment\ndecision. He accordingly requested the following\ninstruction:\n\n As to Dr. Murray’s claim that his disability\n was the reason for Mayo Clinic Arizona’s\n decision to discharge him, Dr. Murray has the\n burden of proving the following evidence by\n a preponderance of the evidence:\n\n ...\n\n 3. Dr. Murray was discharged because\n Defendants regarded him as disabled, which\n means that Defendants’ belief that Plaintiff\n\n\n 1\n In a memorandum disposition filed simultaneously with this\nopinion, we affirm the district court as to all other issues raised by\nMurray.\n\f MURRAY V. MAYO CLINIC 5\n\n had a disability was a motivating factor in\n Defendants’ decision to terminate him.\n\nThe district court instead instructed the jury to apply a but-\nfor causation standard to Murray’s ADA claim. The\ninstruction provided that Murray must prove he was\ndischarged because of his disability:\n\n As to Dr. Murray’s claim that his disability\n was the reason for Mayo Clinic Arizona’s\n decision to discharge him, Dr. Murray has the\n burden of proving the following evidence by\n a preponderance of the evidence:\n\n ...\n\n 3. Dr. Murray was discharged because of his\n disability.\n\nIn denying Murray’s motion for reconsideration, the district\ncourt found that the Supreme Court’s rulings in Gross v. FBL\nFinancial Services, Inc., 557 U.S. 167 (2009), and\nUniversity of Texas Southwestern Medical Center v. Nassar,\n570 U.S. 338 (2013), abrogated our reasoning in Head. The\ndistrict court concluded that the but-for causation standard\napplied.\n\n At trial, the jury returned a verdict for Defendants on all\nclaims. Following entry of judgment, Murray timely filed a\nnotice of appeal.\n\n II.\n\n “A district court’s formulation of the jury instructions is\nreviewed for ‘abuse of discretion.’ If, however, ‘the\ninstructions are challenged as a misstatement of the law, they\n\f6 MURRAY V. MAYO CLINIC\n\nare then reviewed de novo.’” Duran v. City of Maywood,\n221 F.3d 1127, 1130 (9th Cir. 2000) (per curiam) (citation\nomitted) (quoting Gilbrook v. City of Westminster, 177 F.3d\n839, 860 (9th Cir. 1999), as amended on denial of reh’g\n(July 15, 1999)). Jury instructions must fairly and\nadequately cover the issues presented and must not be\nmisleading. Gantt v. City of Los Angeles, 717 F.3d 702, 706\n(9th Cir. 2013).\n\n A.\n\n Title I of the ADA provides that “[n]o covered entity\nshall discriminate against a qualified individual on the basis\nof disability in regard to job application procedures, the\nhiring, advancement, or discharge of employees, employee\ncompensation, job training, and other terms, conditions, and\nprivileges of employment.” 42 U.S.C. § 12112(a) (emphasis\nadded).\n\n Title I of the ADA also contains an enforcement\nprovision, which cross-references specific portions of\nTitle VII:\n\n The powers, remedies, and procedures set\n forth in sections 2000e-4, 2000e-5, 2000e-6,\n 2000e-8, and 2000e-9 of [Title VII] shall be\n the powers, remedies, and procedures this\n subchapter provides to . . . any person\n alleging discrimination on the basis of\n disability in violation of any provision of this\n chapter . . . .\n\n42 U.S.C. § 12117(a). Of the cross-referenced sections, only\n§ 2000e-5 references a causation standard. Specifically, that\nsection provides: “[o]n a claim in which an individual proves\na violation under section 2000e-2(m) of this title and a\n\f MURRAY V. MAYO CLINIC 7\n\nrespondent demonstrates that the respondent would have\ntaken the same action in the absence of the impermissible\nmotivating factor,” the court may award only limited relief.\n42 U.S.C. § 2000e-5(g)(2)(B). Section 2000e-2(m), in turn,\nprovides that “an unlawful employment practice is\nestablished when the complaining party demonstrates that\nrace, color, religion, sex, or national origin was a motivating\nfactor for any employment practice, even though other\nfactors also motivated the practice.”\n\n B.\n\n We first analyzed the standard for causation in a Title I\nADA discrimination action in Head v. Glacier Northwest,\nInc. In that case, we addressed “whether the ADA’s use of\nthe causal language ‘because of,’ ‘by reason of,’ and\n‘because’ means that discriminatory and retaliatory conduct\nis proscribed only if it was solely because of, solely by reason\nof, or solely because an employee was disabled or requested\nan accommodation.” Head, 413 F.3d at 1063–64. We\nconcluded that, under the “plain language of the ADA . . .\n‘solely’ [was] not the appropriate causal standard under any\nof the ADA’s liability provisions.” Id. at 1065.\n\n Considering whether the ADA instead requires but-for\ncausation, or merely a showing that the disability was a\nmotivating factor of the discrimination, we joined seven\nother circuits in concluding that “a ‘motivating factor’\nstandard [was] most consistent with the plain language of the\nstatute and the purposes of the ADA.” Id. We thus held “the\nADA outlaws adverse employment decisions motivated,\neven in part, by animus based on a plaintiff’s disability or\nrequest for an accommodation—a motivating factor\nstandard.” Id.\n\f8 MURRAY V. MAYO CLINIC\n\n In so holding, we relied in part on the reasoning of our\nsister circuits. See id. at 1065 n.63. The Fifth and Eighth\nCircuits had held the motivating factor standard applied to\nthe ADA by virtue of the ADA’s incorporation in § 12117\nof Title VII’s remedies in § 2000e-5. 2 See Buchanan v. City\nof San Antonio, 85 F.3d 196, 200 (5th Cir. 1996); Pedigo v.\nP.A.M. Transp., Inc., 60 F.3d 1300, 1301 (8th Cir. 1995). 3\n\n The Second and Seventh Circuits had concluded that\nADA discrimination claims, like Title VII discrimination\nclaims, only required a showing that discrimination\nmotivated an employer’s adverse employment action. This\nis because ADA and Title VII, at the time, both used the\nwords “because of” to indicate causation, suggesting\nCongress intended the statutes to employ the same causation\nstandard. See Parker v. Columbia Pictures Indus., 204 F.3d\n326, 337 (2d Cir. 2000), overruled by Natofsky v. City of\nNew York, 921 F.3d 337 (2d Cir. 2019); Foster v. Arthur\nAndersen, LLP, 168 F.3d 1029, 1033 (7th Cir. 1999),\noverruled by Serwatka v. Rockwell Automation Inc.,\n591 F.3d 957 (7th Cir. 2010). Both courts acknowledged\nthat, although § 2000e-2(m) is not, by its terms, specifically\napplicable to ADA cases, Congress intended the mixed-\nmotive framework to apply to ADA claims. Parker,\n204 F.3d at 337; Foster, 168 F.3d at 1033.\n\n\n 2\n Similarly, the Fourth Circuit had found the motivating factor\nstandard applied in a Title II ADA case through the ADA’s incorporation\nin 42 U.S.C. § 12133 of the remedies set forth in 29 U.S.C. § 794a,\nwhich, in turn, incorporated the remedies in 42 U.S.C. § 2000e-5. Baird\nex rel. Baird v. Rose, 192 F.3d 462, 470 (4th Cir. 1999).\n 3\n The First Circuit relied on Pedigo, without additional analysis, in\napplying the motivating factor standard. Katz v. City Metal Co., 87 F.3d\n26, 33 (1st Cir. 1996).\n\f MURRAY V. MAYO CLINIC 9\n\n C.\n\n Four years after our decision in Head, the Supreme Court\ndecided Gross v. FBL Financial Services, Inc. The Court\nheld that the Age Discrimination in Employment Act\n(“ADEA”)—which makes it unlawful for an employer to\ndischarge or discriminate against any individual “because of\nsuch an individual’s age”—requires the plaintiff to “prove\nthat age was the ‘but-for’ cause of the employer’s adverse\ndecision.” Gross, 557 U.S. at 177–78. In so doing, the Court\ndeclined to extend 4 the “motivating factor” standard of\ncausation to employment discrimination cases brought under\nthe ADEA. Id. Four years after Gross, the Supreme Court\nin Nassar again declined to extend the motivating factor\nstandard, this time to Title VII retaliation claims. 570 U.S.\nat 362–63.\n\n Against this backdrop, “circuits have retreated from the\nmotivating factor standard of causation in ADA cases.”\nBukiri v. Lynch, 648 F. App’x 729, 731 n.1 (9th Cir. 2016)\n(collecting cases). We have not yet decided whether Gross\nand Nassar have “eroded Head’s vitality.” Mendoza v.\nRoman Catholic Archbishop of L.A., 824 F.3d 1148, 1150\nn.1 (9th Cir. 2016) (per curiam). We do so now.\n\n D.\n\n Murray contends that the motivating factor standard\napplies because we are bound by our decision in Head. We\ndisagree.\n\n\n 4\n The Supreme Court recognized the “motivating standard” of\ncausation as the appropriate standard for employment discrimination\nactions brought under Title VII of the Civil Rights Act of 1964. Price\nWaterhouse v. Hopkins, 490 U.S. 228, 258 (1989) (plurality opinion).\n\f10 MURRAY V. MAYO CLINIC\n\n Generally, a three-judge panel may not overrule a prior\ndecision of the court. Miller v. Gammie, 335 F.3d 889, 899\n(9th Cir. 2003) (en banc). If, however, “an intervening\nSupreme Court decision undermines an existing precedent\nof the Ninth Circuit, and both cases are closely on point,” the\nthree-judge panel may then overrule prior circuit authority.\nId. (quoting Galbraith v. Cty. of Santa Clara, 307 F.3d 1119,\n1123 (9th Cir. 2002)). The issue decided by the higher court\nneed not be identical. Id. at 900. The appropriate test is\nwhether the higher court “undercut the theory or reasoning\nunderlying the prior circuit precedent in such a way that the\ncases are clearly irreconcilable.” Id.\n\n “The ‘clearly irreconcilable’ requirement is ‘a high\nstandard.’” United States v. Robertson, 875 F.3d 1281, 1291\n(9th Cir. 2017) (quoting Rodriguez v. AT & T Mobility Servs.\nLLC, 728 F.3d 975, 979 (9th Cir. 2013)). “It is not enough\nfor there to be ‘some tension’ between the intervening higher\nauthority and prior circuit precedent, or for the intervening\nhigher authority to ‘cast doubt’ on the prior circuit\nprecedent.” Lair v. Bullock, 697 F.3d 1200, 1207 (9th Cir.\n2012) (internal citation omitted) (quoting United States v.\nOrm Hieng, 679 F.3d 1131, 1140–41 (9th Cir. 2012), and\nUnited States v. Delgado-Ramos, 635 F.3d 1237, 1239 (9th\nCir. 2011) (per curiam)). If the court can apply prior circuit\nprecedent without running afoul of the intervening authority,\nit must do so. Id.\n\n Because Head’s reasoning is clearly irreconcilable with\nGross and Nassar, we overrule Head’s holding that a\nplaintiff bringing a discrimination claim under Title I of the\nADA need show only that a disability was a motivating\nfactor of the adverse employment action. We hold instead\nthat an ADA discrimination plaintiff bringing a claim under\n\f MURRAY V. MAYO CLINIC 11\n\n42 U.S.C. § 12112 must show that the adverse employment\naction would not have occurred but for the disability.\n\n In Head, we relied on the reasoning of our sister circuits\nand our existing precedent in finding that a motivating factor\nwas most consistent with the ADA’s plain language and\npurpose. Head, 413 F.3d at 1065 & nn.63–64. Our prior\nprecedent, however, provides no further analysis of the text\nor purpose of the ADA in support of applying a motivating\nfactor causation standard. See Hernandez v. Hughes Missile\nSys. Co., 362 F.3d 564, 568 (9th Cir. 2004); Snead v. Metro.\nProp. & Cas. Ins. Co., 237 F.3d 1080, 1094 (9th Cir. 2001). 5\nAdditionally, Gross and Nassar undercut the reasoning set\nforth by our sister circuits.\n\n Gross held that the ADEA, which also used “because of”\nto indicate causation, did not permit mixed-motive claims\nbecause “the ADEA’s text does not provide that a plaintiff\nmay establish discrimination by showing that age was\nsimply a motivating factor.” 557 U.S. at 174. The Court\nrejected the argument that Title VII decisions governed\ninterpretation of the ADEA on the basis that the two statutes\nwere distinguishable. Id. (“[W]e ‘must be careful not to\napply the rules applicable under one statute to a different\nstatute without careful and critical examination.’” (quoting\n\n\n 5\n Head noted Hernandez’s characterization of the burden on an\nADA plaintiff as “proving that ‘disability actually played a role in the\nemployer’s decisionmaking process and had a determinative influence\non the outcome.’” Head, 413 F.3d at 1065 (quoting Hernandez, 362 F.3d\nat 568 (emphasis in Head)). Head also observed Snead’s statement that\na plaintiff must demonstrate that “a discriminatory reason more likely\nmotivated the employer.” Head, 413 F.3d at 1065 (quoting Snead,\n237 F.3d at 1094 (emphasis in Head)). Neither statement requires a\nmotivating factor standard.\n\f12 MURRAY V. MAYO CLINIC\n\nFed. Exp. Corp. v. Holowecki, 552 U.S. 389, 393 (2008)).\nThe Court explained,\n\n Unlike Title VII, which has been amended to\n explicitly authorize discrimination claims\n where an improper consideration was a\n ‘motivating factor’ for the adverse action, the\n ADEA does not provide that a plaintiff may\n establish discrimination by showing that age\n was simply a motivating factor. Moreover,\n Congress neglected to add such a provision to\n the ADEA when it added §§ 2000e-2(m) and\n 2000e-5(g)(2)(B) to Title VII, even though it\n contemporaneously amended the ADEA in\n several ways.\n\nId. (citations omitted).\n\n Gross’s reasoning directly contradicts the textual\nreasoning Head and other courts applied to conclude that\nTitle VII’s motivating factor standard applied to ADA\nclaims. See Parker, 204 F.3d at 337; Foster, 168 F.3d\nat 1033. Like the ADEA, and unlike Title VII, the ADA\ndoes not contain any explicit “motivating factor” language.\nSee Gross, 557 U.S. at 174. Rather, Title I of the ADA\nprovides that a plaintiff must show discrimination “on the\nbasis of disability.” 42 U.S.C. § 12112(a). Under Gross, the\nphrase “on the basis of disability” indicates but-for\ncausation. Gross, 557 U.S. at 176; see also Nassar, 570 U.S.\nat 350 (explaining Gross’s holding that “because of,” “by\n\f MURRAY V. MAYO CLINIC 13\n\nreason of,” “on account of,” and “based on” all indicate a\nbut-for causal relationship). 6\n\n Nassar’s reasoning likewise directly undercuts the\nreasoning of courts that relied on the ADA’s incorporation\nin § 12117 of § 2000e-5. See Buchanan, 85 F.3d at 200;\nPedigo, 60 F.3d at 1301; cf. Baird ex rel. Baird v. Rose,\n192 F.3d 462, 470 (4th Cir. 1999) (ADA Title II’s\nincorporation of § 2000e-5). Nassar rejected the argument\nthat § 2000e-2(m), Title VII’s motivating factor causation\nprovision, applies to Title VII retaliation claims. Nassar,\n570 U.S. at 353. The Court emphasized that “the text of the\nmotivating-factor provision, while it begins by referring to\n‘unlawful employment practices,’ then proceeds to address\nonly five of the seven prohibited discriminatory actions—\nactions based on the employee’s status, i.e., race, color,\nreligion, sex, and national origin.” Id. According to Nassar,\nthe plain language of § 2000e-2(m) barred its application to\nretaliation claims, and “it would be improper to conclude\nthat what Congress omitted from the statute is nevertheless\nwithin its scope.” Id.\n\n The same logic applies to Title I ADA discrimination\nclaims. Relief under § 2000e-5(g)(2)(B) is available only if\n\n 6\n Title I of the ADA was amended in 2008 to prohibit discrimination\n“on the basis of” disability, rather than “because of” disability. We find\nno meaningful textual difference in the two phrases with respect to\ncausation. The Second and Fourth Circuits likewise found no\nmeaningful textual difference between the two standards and found\nnothing in the legislative history suggesting Congress intended to modify\nthe ADA’s standard for causation. Natofsky v. City of New York,\n921 F.3d 337, 349–50 (2d Cir. 2019); Gentry v. E.W. Partners Club\nMgmt. Co., 816 F.3d 228, 234 (4th Cir. 2016) (noting that the ADA\namendment was enacted before Gross, and was therefore not in response\nto Gross’s causation analysis).\n\f14 MURRAY V. MAYO CLINIC\n\nthe plaintiff proves a violation under § 2000e-2(m). Section\n2000e-2(m) narrowly prohibits the consideration of race,\ncolor, religion, sex, or national origin as a motivating factor\nfor any employment practice. It does not prohibit the\nconsideration of disability. Congress’s express listing of\nthese status-based considerations under § 2000e-2(m) is best\nunderstood as an exclusion of all other considerations. See,\ne.g., Silvers v. Sony Pictures Entm’t, Inc., 402 F.3d 881, 885\n(9th Cir. 2005) (“The doctrine of expressio unius est exclusio\nalterius ‘as applied to statutory interpretation creates a\npresumption that when a statute designates certain persons,\nthings, or manners of operation, all omissions should be\nunderstood as exclusions.’” (quoting Boudette v. Barnette,\n923 F.2d 754, 756–57 (9th Cir. 1991))). Section 2000e-\n2(m), by its plain language, is inapplicable to claims of\ndisability discrimination.\n\n Because Head’s reasoning—whether based on the\nADA’s cross-reference to § 2000e-5(g)(2)(B) or on the\nADA’s text—is irreconcilable with subsequent Supreme\nCourt precedent, it cannot stand.\n\n III.\n\n Our decision comports with the decisions of all of our\nsister circuits that have considered this question after Gross\nand Nassar. The Second, Fourth, and Seventh Circuits\nfound the Supreme Court’s intervening jurisprudence to be\ndispositive of the issue. See Natofsky, 921 F.3d at 348\n(“Gross and Nassar dictate our decision here.”); Gentry v.\nE.W. Partners Club Mgmt. Co., 816 F.3d 228, 234 (4th Cir.\n2016) (“The Supreme Court’s analysis in Gross dictates the\noutcome here.”); Serwatka, 591 F.3d at 963 (“But in view of\nthe Court’s intervening decision in Gross, it is clear that the\ndistrict court’s decision . . . cannot be sustained.”). The\nSixth Circuit, following en banc review, similarly held that\n\f MURRAY V. MAYO CLINIC 15\n\nGross’s reasoning was controlling. Lewis v. Humboldt\nAcquisition Corp., 681 F.3d 312, 318 (6th Cir. 2012) (en\nbanc) (“[Gross’s] rationale applies with equal force to the\nADA.”).\n\n We agree. Gross and Nassar undermine Head’s\nreasoning such that the cases are clearly irreconcilable. We\njoin our sister circuits in holding that ADA discrimination\nclaims under Title I must be evaluated under a but-for\ncausation standard.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Michael Murray v. Mayo Clinic\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-20\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMICHAEL J. MURRAY, M.D. - a No. 17-16803\\nmarried man,\\n Plaintiff-Appellant, D.C. No.\\n 2:14-cv-01314-\\n v. SPL\\n\\nMAYO CLINIC, a Minnesota nonprofit\\ncorporation; MAYO CLINIC ARIZONA, OPINION\\nan Arizona nonprofit corporation;\\nWYATT DECKER, M.D. - husband;\\nGEORGIANNA DECKER, wife; LOIS\\nKRAHN, M.D. - wife; ERIC GORDON,\\nM.D. - husband; TERRENCE\\nTRENTMAN, M.D. - husband;\\nLARALEE TRENTMAN, wife; WILLIAM\\nSTONE, M.D. - husband; MAREE\\nSTONE, wife; DAVID ROSENFELD,\\nM.D. - husband; MELISSA\\nROSENFELD, M.D. - wife; ROSHANAK\\nDIDEHBAN, a single woman,\\n Defendants-Appellees.\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Steven Paul Logan, District Judge, Presiding\\n\\n Argued and Submitted June 13, 2019\\n San Francisco, California\\n\\f2 MURRAY V. MAYO CLINIC\\n\\n Filed August 20, 2019\\n\\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\\n Judges, and Benita Y. Pearson, * District Judge.\\n\\n Opinion by Judge Pearson\\n\\n\\n SUMMARY **\\n\\n\\n Employment Discrimination\\n\\n The panel affirmed the district court’s judgment, after a\\njury trial, in favor of the defendants in an employment\\ndiscrimination action under Title I of the Americans with\\nDisabilities Act.\\n\\n The panel held that the district court correctly instructed\\nthe jury to apply a but for causation standard, rather than a\\nmotivating factor standard. The panel concluded that Head\\nv. Glacier Northwest, Inc., 413 F.3d 1053 (9th Cir. 2005),\\nholding that ADA discrimination claims are evaluated under\\na motivating factor causation standard, is no longer good law\\nbecause its reasoning is clearly irreconcilable with the\\nSupreme Court’s rulings in Gross v. FBL Fin. Servs., Inc.,\\n557 U.S. 167 (2009), and Univ. of Texas Southwestern Med.\\nCtr. v. Nassar, 570 U.S. 338 (2013). Agreeing with other\\ncircuits, the panel held that an ADA discrimination plaintiff\\n\\n *\\n The Honorable Benita Y. Pearson, United States District Judge for\\nthe Northern District of Ohio, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f MURRAY V. MAYO CLINIC 3\\n\\nbringing a claim under 42 U.S.C. § 12112 must show that\\nthe adverse employment action would not have occurred but\\nfor the disability.\\n\\n The panel addressed other issues in a simultaneously\\nfiled memorandum disposition.\\n\\n\\n COUNSEL\\n\\nRoger L. Cohen (argued), Jaburg Wilk P.C., Phoenix,\\nArizona; Scott A. Blaney, Blaney Law PLLC, Phoenix,\\nArizona; for Plaintiff-Appellant.\\n\\nJohn F. Lomax, Jr. (argued) and Kelly Kszywienski, Snell &\\nWilmer L.L.P., Phoenix, Arizona, for Defendants-\\nAppellees.\\n\\n\\n OPINION\\n\\nPEARSON, District Judge:\\n\\n Plaintiff Michael Murray appeals the district court’s\\ninstruction to the jury on his claim under the Americans with\\nDisabilities Act (“ADA”), requiring him to prove that he was\\ndischarged because of his disability. Murray claims that our\\ndecision in Head v. Glacier Northwest, Inc., 413 F.3d 1053\\n(9th Cir. 2005), holding ADA discrimination claims are\\n\\f4 MURRAY V. MAYO CLINIC\\n\\nevaluated under a motivating factor causation standard,\\nremains good law. Because it is not, we affirm. 1\\n\\n I.\\n\\n Dr. Murray filed suit against Mayo Clinic, Mayo Clinic\\nArizona, Drs. Wyatt Decker, Lois Krahn, Terrence\\nTrentman, William Stone, and David Rosenfeld, and\\nOperations Administrator Roshanak Didehban. In\\nanticipation of trial, the parties submitted joint proposed jury\\ninstructions. The parties disagreed whether Murray’s ADA\\ndiscrimination claim should be tried under a but-for\\ncausation standard or a motivating factor causation standard.\\nMurray argued that our decision in Head required him to\\nshow only that the defendants’ belief that he had a disability\\nwas a motivating factor in their adverse employment\\ndecision. He accordingly requested the following\\ninstruction:\\n\\n As to Dr. Murray’s claim that his disability\\n was the reason for Mayo Clinic Arizona’s\\n decision to discharge him, Dr. Murray has the\\n burden of proving the following evidence by\\n a preponderance of the evidence:\\n\\n ...\\n\\n 3. Dr. Murray was discharged because\\n Defendants regarded him as disabled, which\\n means that Defendants’ belief that Plaintiff\\n\\n\\n 1\\n In a memorandum disposition filed simultaneously with this\\nopinion, we affirm the district court as to all other issues raised by\\nMurray.\\n\\f MURRAY V. MAYO CLINIC 5\\n\\n had a disability was a motivating factor in\\n Defendants’ decision to terminate him.\\n\\nThe district court instead instructed the jury to apply a but-\\nfor causation standard to Murray’s ADA claim. The\\ninstruction provided that Murray must prove he was\\ndischarged because of his disability:\\n\\n As to Dr. Murray’s claim that his disability\\n was the reason for Mayo Clinic Arizona’s\\n decision to discharge him, Dr. Murray has the\\n burden of proving the following evidence by\\n a preponderance of the evidence:\\n\\n ...\\n\\n 3. Dr. Murray was discharged because of his\\n disability.\\n\\nIn denying Murray’s motion for reconsideration, the district\\ncourt found that the Supreme Court’s rulings in Gross v. FBL\\nFinancial Services, Inc., 557 U.S. 167 (2009), and\\nUniversity of Texas Southwestern Medical Center v. Nassar,\\n570 U.S. 338 (2013), abrogated our reasoning in Head. The\\ndistrict court concluded that the but-for causation standard\\napplied.\\n\\n At trial, the jury returned a verdict for Defendants on all\\nclaims. Following entry of judgment, Murray timely filed a\\nnotice of appeal.\\n\\n II.\\n\\n “A district court’s formulation of the jury instructions is\\nreviewed for ‘abuse of discretion.’ If, however, ‘the\\ninstructions are challenged as a misstatement of the law, they\\n\\f6 MURRAY V. MAYO CLINIC\\n\\nare then reviewed de novo.’” Duran v. City of Maywood,\\n221 F.3d 1127, 1130 (9th Cir. 2000) (per curiam) (citation\\nomitted) (quoting Gilbrook v. City of Westminster, 177 F.3d\\n839, 860 (9th Cir. 1999), as amended on denial of reh’g\\n(July 15, 1999)). Jury instructions must fairly and\\nadequately cover the issues presented and must not be\\nmisleading. Gantt v. City of Los Angeles, 717 F.3d 702, 706\\n(9th Cir. 2013).\\n\\n A.\\n\\n Title I of the ADA provides that “[n]o covered entity\\nshall discriminate against a qualified individual on the basis\\nof disability in regard to job application procedures, the\\nhiring, advancement, or discharge of employees, employee\\ncompensation, job training, and other terms, conditions, and\\nprivileges of employment.” 42 U.S.C. § 12112(a) (emphasis\\nadded).\\n\\n Title I of the ADA also contains an enforcement\\nprovision, which cross-references specific portions of\\nTitle VII:\\n\\n The powers, remedies, and procedures set\\n forth in sections 2000e-4, 2000e-5, 2000e-6,\\n 2000e-8, and 2000e-9 of [Title VII] shall be\\n the powers, remedies, and procedures this\\n subchapter provides to . . . any person\\n alleging discrimination on the basis of\\n disability in violation of any provision of this\\n chapter . . . .\\n\\n42 U.S.C. § 12117(a). Of the cross-referenced sections, only\\n§ 2000e-5 references a causation standard. Specifically, that\\nsection provides: “[o]n a claim in which an individual proves\\na violation under section 2000e-2(m) of this title and a\\n\\f MURRAY V. MAYO CLINIC 7\\n\\nrespondent demonstrates that the respondent would have\\ntaken the same action in the absence of the impermissible\\nmotivating factor,” the court may award only limited relief.\\n42 U.S.C. § 2000e-5(g)(2)(B). Section 2000e-2(m), in turn,\\nprovides that “an unlawful employment practice is\\nestablished when the complaining party demonstrates that\\nrace, color, religion, sex, or national origin was a motivating\\nfactor for any employment practice, even though other\\nfactors also motivated the practice.”\\n\\n B.\\n\\n We first analyzed the standard for causation in a Title I\\nADA discrimination action in Head v. Glacier Northwest,\\nInc. In that case, we addressed “whether the ADA’s use of\\nthe causal language ‘because of,’ ‘by reason of,’ and\\n‘because’ means that discriminatory and retaliatory conduct\\nis proscribed only if it was solely because of, solely by reason\\nof, or solely because an employee was disabled or requested\\nan accommodation.” Head, 413 F.3d at 1063–64. We\\nconcluded that, under the “plain language of the ADA . . .\\n‘solely’ [was] not the appropriate causal standard under any\\nof the ADA’s liability provisions.” Id. at 1065.\\n\\n Considering whether the ADA instead requires but-for\\ncausation, or merely a showing that the disability was a\\nmotivating factor of the discrimination, we joined seven\\nother circuits in concluding that “a ‘motivating factor’\\nstandard [was] most consistent with the plain language of the\\nstatute and the purposes of the ADA.” Id. We thus held “the\\nADA outlaws adverse employment decisions motivated,\\neven in part, by animus based on a plaintiff’s disability or\\nrequest for an accommodation—a motivating factor\\nstandard.” Id.\\n\\f8 MURRAY V. MAYO CLINIC\\n\\n In so holding, we relied in part on the reasoning of our\\nsister circuits. See id. at 1065 n.63. The Fifth and Eighth\\nCircuits had held the motivating factor standard applied to\\nthe ADA by virtue of the ADA’s incorporation in § 12117\\nof Title VII’s remedies in § 2000e-5. 2 See Buchanan v. City\\nof San Antonio, 85 F.3d 196, 200 (5th Cir. 1996); Pedigo v.\\nP.A.M. Transp., Inc., 60 F.3d 1300, 1301 (8th Cir. 1995). 3\\n\\n The Second and Seventh Circuits had concluded that\\nADA discrimination claims, like Title VII discrimination\\nclaims, only required a showing that discrimination\\nmotivated an employer’s adverse employment action. This\\nis because ADA and Title VII, at the time, both used the\\nwords “because of” to indicate causation, suggesting\\nCongress intended the statutes to employ the same causation\\nstandard. See Parker v. Columbia Pictures Indus., 204 F.3d\\n326, 337 (2d Cir. 2000), overruled by Natofsky v. City of\\nNew York, 921 F.3d 337 (2d Cir. 2019); Foster v. Arthur\\nAndersen, LLP, 168 F.3d 1029, 1033 (7th Cir. 1999),\\noverruled by Serwatka v. Rockwell Automation Inc.,\\n591 F.3d 957 (7th Cir. 2010). Both courts acknowledged\\nthat, although § 2000e-2(m) is not, by its terms, specifically\\napplicable to ADA cases, Congress intended the mixed-\\nmotive framework to apply to ADA claims. Parker,\\n204 F.3d at 337; Foster, 168 F.3d at 1033.\\n\\n\\n 2\\n Similarly, the Fourth Circuit had found the motivating factor\\nstandard applied in a Title II ADA case through the ADA’s incorporation\\nin 42 U.S.C. § 12133 of the remedies set forth in 29 U.S.C. § 794a,\\nwhich, in turn, incorporated the remedies in 42 U.S.C. § 2000e-5. Baird\\nex rel. Baird v. Rose, 192 F.3d 462, 470 (4th Cir. 1999).\\n 3\\n The First Circuit relied on Pedigo, without additional analysis, in\\napplying the motivating factor standard. Katz v. City Metal Co., 87 F.3d\\n26, 33 (1st Cir. 1996).\\n\\f MURRAY V. MAYO CLINIC 9\\n\\n C.\\n\\n Four years after our decision in Head, the Supreme Court\\ndecided Gross v. FBL Financial Services, Inc. The Court\\nheld that the Age Discrimination in Employment Act\\n(“ADEA”)—which makes it unlawful for an employer to\\ndischarge or discriminate against any individual “because of\\nsuch an individual’s age”—requires the plaintiff to “prove\\nthat age was the ‘but-for’ cause of the employer’s adverse\\ndecision.” Gross, 557 U.S. at 177–78. In so doing, the Court\\ndeclined to extend 4 the “motivating factor” standard of\\ncausation to employment discrimination cases brought under\\nthe ADEA. Id. Four years after Gross, the Supreme Court\\nin Nassar again declined to extend the motivating factor\\nstandard, this time to Title VII retaliation claims. 570 U.S.\\nat 362–63.\\n\\n Against this backdrop, “circuits have retreated from the\\nmotivating factor standard of causation in ADA cases.”\\nBukiri v. Lynch, 648 F. App’x 729, 731 n.1 (9th Cir. 2016)\\n(collecting cases). We have not yet decided whether Gross\\nand Nassar have “eroded Head’s vitality.” Mendoza v.\\nRoman Catholic Archbishop of L.A., 824 F.3d 1148, 1150\\nn.1 (9th Cir. 2016) (per curiam). We do so now.\\n\\n D.\\n\\n Murray contends that the motivating factor standard\\napplies because we are bound by our decision in Head. We\\ndisagree.\\n\\n\\n 4\\n The Supreme Court recognized the “motivating standard” of\\ncausation as the appropriate standard for employment discrimination\\nactions brought under Title VII of the Civil Rights Act of 1964. Price\\nWaterhouse v. Hopkins, 490 U.S. 228, 258 (1989) (plurality opinion).\\n\\f10 MURRAY V. MAYO CLINIC\\n\\n Generally, a three-judge panel may not overrule a prior\\ndecision of the court. Miller v. Gammie, 335 F.3d 889, 899\\n(9th Cir. 2003) (en banc). If, however, “an intervening\\nSupreme Court decision undermines an existing precedent\\nof the Ninth Circuit, and both cases are closely on point,” the\\nthree-judge panel may then overrule prior circuit authority.\\nId. (quoting Galbraith v. Cty. of Santa Clara, 307 F.3d 1119,\\n1123 (9th Cir. 2002)). The issue decided by the higher court\\nneed not be identical. Id. at 900. The appropriate test is\\nwhether the higher court “undercut the theory or reasoning\\nunderlying the prior circuit precedent in such a way that the\\ncases are clearly irreconcilable.” Id.\\n\\n “The ‘clearly irreconcilable’ requirement is ‘a high\\nstandard.’” United States v. Robertson, 875 F.3d 1281, 1291\\n(9th Cir. 2017) (quoting Rodriguez v. AT & T Mobility Servs.\\nLLC, 728 F.3d 975, 979 (9th Cir. 2013)). “It is not enough\\nfor there to be ‘some tension’ between the intervening higher\\nauthority and prior circuit precedent, or for the intervening\\nhigher authority to ‘cast doubt’ on the prior circuit\\nprecedent.” Lair v. Bullock, 697 F.3d 1200, 1207 (9th Cir.\\n2012) (internal citation omitted) (quoting United States v.\\nOrm Hieng, 679 F.3d 1131, 1140–41 (9th Cir. 2012), and\\nUnited States v. Delgado-Ramos, 635 F.3d 1237, 1239 (9th\\nCir. 2011) (per curiam)). If the court can apply prior circuit\\nprecedent without running afoul of the intervening authority,\\nit must do so. Id.\\n\\n Because Head’s reasoning is clearly irreconcilable with\\nGross and Nassar, we overrule Head’s holding that a\\nplaintiff bringing a discrimination claim under Title I of the\\nADA need show only that a disability was a motivating\\nfactor of the adverse employment action. We hold instead\\nthat an ADA discrimination plaintiff bringing a claim under\\n\\f MURRAY V. MAYO CLINIC 11\\n\\n42 U.S.C. § 12112 must show that the adverse employment\\naction would not have occurred but for the disability.\\n\\n In Head, we relied on the reasoning of our sister circuits\\nand our existing precedent in finding that a motivating factor\\nwas most consistent with the ADA’s plain language and\\npurpose. Head, 413 F.3d at 1065 & nn.63–64. Our prior\\nprecedent, however, provides no further analysis of the text\\nor purpose of the ADA in support of applying a motivating\\nfactor causation standard. See Hernandez v. Hughes Missile\\nSys. Co., 362 F.3d 564, 568 (9th Cir. 2004); Snead v. Metro.\\nProp. & Cas. Ins. Co., 237 F.3d 1080, 1094 (9th Cir. 2001). 5\\nAdditionally, Gross and Nassar undercut the reasoning set\\nforth by our sister circuits.\\n\\n Gross held that the ADEA, which also used “because of”\\nto indicate causation, did not permit mixed-motive claims\\nbecause “the ADEA’s text does not provide that a plaintiff\\nmay establish discrimination by showing that age was\\nsimply a motivating factor.” 557 U.S. at 174. The Court\\nrejected the argument that Title VII decisions governed\\ninterpretation of the ADEA on the basis that the two statutes\\nwere distinguishable. Id. (“[W]e ‘must be careful not to\\napply the rules applicable under one statute to a different\\nstatute without careful and critical examination.’” (quoting\\n\\n\\n 5\\n Head noted Hernandez’s characterization of the burden on an\\nADA plaintiff as “proving that ‘disability actually played a role in the\\nemployer’s decisionmaking process and had a determinative influence\\non the outcome.’” Head, 413 F.3d at 1065 (quoting Hernandez, 362 F.3d\\nat 568 (emphasis in Head)). Head also observed Snead’s statement that\\na plaintiff must demonstrate that “a discriminatory reason more likely\\nmotivated the employer.” Head, 413 F.3d at 1065 (quoting Snead,\\n237 F.3d at 1094 (emphasis in Head)). Neither statement requires a\\nmotivating factor standard.\\n\\f12 MURRAY V. MAYO CLINIC\\n\\nFed. Exp. Corp. v. Holowecki, 552 U.S. 389, 393 (2008)).\\nThe Court explained,\\n\\n Unlike Title VII, which has been amended to\\n explicitly authorize discrimination claims\\n where an improper consideration was a\\n ‘motivating factor’ for the adverse action, the\\n ADEA does not provide that a plaintiff may\\n establish discrimination by showing that age\\n was simply a motivating factor. Moreover,\\n Congress neglected to add such a provision to\\n the ADEA when it added §§ 2000e-2(m) and\\n 2000e-5(g)(2)(B) to Title VII, even though it\\n contemporaneously amended the ADEA in\\n several ways.\\n\\nId. (citations omitted).\\n\\n Gross’s reasoning directly contradicts the textual\\nreasoning Head and other courts applied to conclude that\\nTitle VII’s motivating factor standard applied to ADA\\nclaims. See Parker, 204 F.3d at 337; Foster, 168 F.3d\\nat 1033. Like the ADEA, and unlike Title VII, the ADA\\ndoes not contain any explicit “motivating factor” language.\\nSee Gross, 557 U.S. at 174. Rather, Title I of the ADA\\nprovides that a plaintiff must show discrimination “on the\\nbasis of disability.” 42 U.S.C. § 12112(a). Under Gross, the\\nphrase “on the basis of disability” indicates but-for\\ncausation. Gross, 557 U.S. at 176; see also Nassar, 570 U.S.\\nat 350 (explaining Gross’s holding that “because of,” “by\\n\\f MURRAY V. MAYO CLINIC 13\\n\\nreason of,” “on account of,” and “based on” all indicate a\\nbut-for causal relationship). 6\\n\\n Nassar’s reasoning likewise directly undercuts the\\nreasoning of courts that relied on the ADA’s incorporation\\nin § 12117 of § 2000e-5. See Buchanan, 85 F.3d at 200;\\nPedigo, 60 F.3d at 1301; cf. Baird ex rel. Baird v. Rose,\\n192 F.3d 462, 470 (4th Cir. 1999) (ADA Title II’s\\nincorporation of § 2000e-5). Nassar rejected the argument\\nthat § 2000e-2(m), Title VII’s motivating factor causation\\nprovision, applies to Title VII retaliation claims. Nassar,\\n570 U.S. at 353. The Court emphasized that “the text of the\\nmotivating-factor provision, while it begins by referring to\\n‘unlawful employment practices,’ then proceeds to address\\nonly five of the seven prohibited discriminatory actions—\\nactions based on the employee’s status, i.e., race, color,\\nreligion, sex, and national origin.” Id. According to Nassar,\\nthe plain language of § 2000e-2(m) barred its application to\\nretaliation claims, and “it would be improper to conclude\\nthat what Congress omitted from the statute is nevertheless\\nwithin its scope.” Id.\\n\\n The same logic applies to Title I ADA discrimination\\nclaims. Relief under § 2000e-5(g)(2)(B) is available only if\\n\\n 6\\n Title I of the ADA was amended in 2008 to prohibit discrimination\\n“on the basis of” disability, rather than “because of” disability. We find\\nno meaningful textual difference in the two phrases with respect to\\ncausation. The Second and Fourth Circuits likewise found no\\nmeaningful textual difference between the two standards and found\\nnothing in the legislative history suggesting Congress intended to modify\\nthe ADA’s standard for causation. Natofsky v. City of New York,\\n921 F.3d 337, 349–50 (2d Cir. 2019); Gentry v. E.W. Partners Club\\nMgmt. Co., 816 F.3d 228, 234 (4th Cir. 2016) (noting that the ADA\\namendment was enacted before Gross, and was therefore not in response\\nto Gross’s causation analysis).\\n\\f14 MURRAY V. MAYO CLINIC\\n\\nthe plaintiff proves a violation under § 2000e-2(m). Section\\n2000e-2(m) narrowly prohibits the consideration of race,\\ncolor, religion, sex, or national origin as a motivating factor\\nfor any employment practice. It does not prohibit the\\nconsideration of disability. Congress’s express listing of\\nthese status-based considerations under § 2000e-2(m) is best\\nunderstood as an exclusion of all other considerations. See,\\ne.g., Silvers v. Sony Pictures Entm’t, Inc., 402 F.3d 881, 885\\n(9th Cir. 2005) (“The doctrine of expressio unius est exclusio\\nalterius ‘as applied to statutory interpretation creates a\\npresumption that when a statute designates certain persons,\\nthings, or manners of operation, all omissions should be\\nunderstood as exclusions.’” (quoting Boudette v. Barnette,\\n923 F.2d 754, 756–57 (9th Cir. 1991))). Section 2000e-\\n2(m), by its plain language, is inapplicable to claims of\\ndisability discrimination.\\n\\n Because Head’s reasoning—whether based on the\\nADA’s cross-reference to § 2000e-5(g)(2)(B) or on the\\nADA’s text—is irreconcilable with subsequent Supreme\\nCourt precedent, it cannot stand.\\n\\n III.\\n\\n Our decision comports with the decisions of all of our\\nsister circuits that have considered this question after Gross\\nand Nassar. The Second, Fourth, and Seventh Circuits\\nfound the Supreme Court’s intervening jurisprudence to be\\ndispositive of the issue. See Natofsky, 921 F.3d at 348\\n(“Gross and Nassar dictate our decision here.”); Gentry v.\\nE.W. Partners Club Mgmt. Co., 816 F.3d 228, 234 (4th Cir.\\n2016) (“The Supreme Court’s analysis in Gross dictates the\\noutcome here.”); Serwatka, 591 F.3d at 963 (“But in view of\\nthe Court’s intervening decision in Gross, it is clear that the\\ndistrict court’s decision . . . cannot be sustained.”). The\\nSixth Circuit, following en banc review, similarly held that\\n\\f MURRAY V. MAYO CLINIC 15\\n\\nGross’s reasoning was controlling. Lewis v. Humboldt\\nAcquisition Corp., 681 F.3d 312, 318 (6th Cir. 2012) (en\\nbanc) (“[Gross’s] rationale applies with equal force to the\\nADA.”).\\n\\n We agree. Gross and Nassar undermine Head’s\\nreasoning such that the cases are clearly irreconcilable. We\\njoin our sister circuits in holding that ADA discrimination\\nclaims under Title I must be evaluated under a but-for\\ncausation standard.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4083,"cells":{"id":{"kind":"number","value":4649768,"string":"4,649,768"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Michael Dorman v. the Charles Schwab Corporation"},"decision_date":{"kind":"string","value":"2019-08-20"},"docket_number":{"kind":"string","value":"18-15281"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Ronald M. Gould and Sandra S. Ikuta, Circuit Judges, and Benita Y. Pearson, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PEARSON, District Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMICHAEL F. DORMAN, individually No. 18-15281\\nas a participant in the SCHWAB PLAN\\nRETIREMENT SAVINGS AND D.C. No.\\nINVESTMENT PLAN and on behalf of a 4:17-cv-00285-\\nclass of all those similarly situated, CW\\n Plaintiff-Appellee,\\n\\n v. OPINION\\n\\nTHE CHARLES SCHWAB\\nCORPORATION; CHARLES SCHWAB &\\nCO., INC.; SCHWAB RETIREMENT\\nPLAN SERVICES, INC.; CHARLES\\nSCHWAB BANK; CHARLES SCHWAB\\nINVESTMENT MANAGEMENT, INC.;\\nWALTER W. BETTINGER III;\\nCHARLES R. SCHWAB; JOSEPH R.\\nMARTINETTO; MARTHA TUMA; JAY\\nALLEN; DAVE CALLAHAN; JOHN C.\\nCLARK,\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n Claudia Wilken, District Judge, Presiding\\n\\n Argued and Submitted June 14, 2019\\n San Francisco, California\\n\\f2 DORMAN V. CHARLES SCHWAB CORP.\\n\\n Filed August 20, 2019\\n\\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\\n Judges, and Benita Y. Pearson, * District Judge.\\n\\n Opinion by Judge Pearson\\n\\n\\n SUMMARY **\\n\\n\\n ERISA / Arbitration\\n\\n The panel reversed the district court’s order denying\\ndefendants’ motion to compel arbitration of claims and\\nremanded in a class action suit brought by a former\\nparticipant in an ERISA retirement plan, alleging that\\ndefendants violated ERISA and breached their fiduciary\\nduties by including certain investment funds in the plan.\\n\\n The panel concluded that Amaro v. Continental Can Co.,\\n724 F.2d 747 (9th Cir. 1984), which held that ERISA claims\\nare not arbitrable, is no longer good law in light of\\nintervening Supreme Court case law, including American\\nExpress Co. v. Italian Colors Restaurant, 570 U.S. 228\\n(2013).\\n\\n\\n\\n\\n *\\n The Honorable Benita Y. Pearson, United States District Judge for\\nthe Northern District of Ohio, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f DORMAN V. CHARLES SCHWAB CORP. 3\\n\\n The panel addressed other issues in a concurrently filed\\nmemorandum disposition.\\n\\n\\n COUNSEL\\n\\nHoward Shapiro (argued), Stacey C.S. Cerrone, and Tulio D.\\nChirinos, Proskauer Rose LLP, New Orleans, Louisiana;\\nMyron D. Rumeld, Proskauer Rose LLP, New York, New\\nYork; John E. Roberts, Proskauer Rose LLP, Boston,\\nMassachusetts; for Defendants-Appellants.\\n\\nJames Bloom (argued), Todd M. Schneider, and Kyle G.\\nBates, Schneider Wallace Cottrell Konecky Wotkyns LLP,\\nEmeryville, California; Todd S. Collins, Eric Lechtzin,\\nShanon J. Carson, and Ellen T. Noteware, Berger Montague\\nPC, Philadelphia, Pennsylvania; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nPEARSON, District Judge:\\n\\n Defendants appeal the district court’s denial of their\\nmotion to compel individual arbitration in an ERISA action\\nfiled by Michael Dorman, a former participant in the Schwab\\nRetirement Savings and Investment Plan (the “Plan”).\\nDorman participated in a defined contribution 401(k)\\nretirement plan through his employment with Charles\\nSchwab & Co., Inc. (“Schwab”). In 2017, Dorman filed a\\nclass action suit in district court alleging that Defendants\\nviolated ERISA and breached their fiduciary duties by\\nincluding Schwab-affiliated investment funds in the Plan—\\ndespite the funds’ poor performance—to generate fees for\\nSchwab and its affiliates. Defendants moved to compel\\n\\f4 DORMAN V. CHARLES SCHWAB CORP.\\n\\narbitration pursuant to an arbitration agreement in the Plan.\\nThe district court denied the motion, and this interlocutory\\nappeal followed.\\n\\n On appeal, Defendants contend that the district court\\nerred by not enforcing the Plan’s arbitration agreement. We\\naddress these arguments in a concurrently filed\\nmemorandum disposition. But before we can reach the\\nparties’ specific contentions, we must first address the\\nthreshold question of whether ERISA claims can be subject\\nto mandatory arbitration. In so doing, we must revisit our\\nholding in Amaro v. Continental Can Co., 724 F.2d 747 (9th\\nCir. 1984), in which we held that ERISA claims were not\\narbitrable. In light of intervening Supreme Court case law,\\nincluding American Express Co. v. Italian Colors\\nRestaurant, 570 U.S. 228 (2013), we conclude that our\\nholding in Amaro is no longer good law.\\n\\n I\\n\\n Michael Dorman was employed at Schwab from\\nFebruary 17, 2009, until October 8, 2015. Through his\\nemployment, Dorman joined the Plan in 2009, and he\\nvoluntarily contributed to his retirement account through\\npayroll deductions until he left his employment with\\nSchwab. Dorman withdrew his full account balance on\\nDecember 18, 2015, and ceased participating in the Plan. 1\\n\\n In this defined contribution 401(k) retirement plan, Plan\\nparticipants are given the choice to allocate their earnings\\namong a menu of investment funds, and they may alter their\\n 1\\n In a separate memorandum filed concurrently with this opinion,\\nwe address Defendants’ challenges to the district court’s denial of their\\nmotions to compel individual arbitration and for leave to move for\\nreconsideration and reverse.\\n\\f DORMAN V. CHARLES SCHWAB CORP. 5\\n\\ninvestment allocations at any time. During the relevant\\nperiod, the Plan offered as many as 17 different funds in\\nwhich participants could choose to invest, including both\\nSchwab-affiliated and unaffiliated funds.\\n\\n In December 2014, the Plan was amended to add an\\narbitration provision. That provision took effect on\\nJanuary 1, 2015, nine months before Dorman ended his\\nemployment at Schwab and nearly a year before he\\nterminated his participation in the Plan. The Plan document\\nstates that “[a]ny claim, dispute or breach arising out of or in\\nany way related to the Plan shall be settled by binding\\narbitration . . . .” The arbitration provision includes a waiver\\nof class or collective action that requires individual\\narbitrations, even if absent the waiver Dorman could have\\nrepresented the interests of other Plan participants. It states\\nthat any arbitration would be conducted “on an individual\\nbasis only, and not on a class, collective or representative\\nbasis,” and that Plan participants waive the right to be part\\nof any class action. If that waiver of collective action were\\nto be held unenforceable, the arbitration provision mandates\\nthat “any claim on a class, collective or representative basis\\nshall be filed and adjudicated in a court of competent\\njurisdiction, and not in arbitration.”\\n\\n In 2014, Dorman was promoted to financial consultant,\\nand he enrolled in the Schwab Investor Financial Consultant\\nCompensation Plan (the “Compensation Plan”). By\\nenrolling in the Compensation Plan, Dorman agreed to\\narbitrate “[a]ny controversy, dispute, or claim arising out of\\nor relating to [his] employment . . . .” By its terms, the\\narbitration provision encompassed claims that arise out of\\n“federal, state, or local law.” The arbitration agreement\\ncarved out “claims for benefits” under the Plan and provided\\nthat such “claims for benefits” would be resolved pursuant\\n\\f6 DORMAN V. CHARLES SCHWAB CORP.\\n\\nto the procedures prescribed by the Plan. The Compensation\\nPlan further contains a “Class Action Waiver.”\\n\\n In June 2017, Dorman filed his First Amended Class\\nAction Complaint against Defendants, asserting claims\\nunder § 502(a)(2) and (3) of the Employee Retirement\\nIncome Security Act of 1974 (“ERISA”), 29 U.S.C.\\n§ 1132(a)(2) and (3), and seeking plan-wide relief on behalf\\nof a class comprising all participants in, and beneficiaries of,\\nthe Plan at any time within six years of the filing of the\\nComplaint. The Complaint alleges that various Defendants\\nbreached their fiduciary duties of loyalty and prudence and\\nviolated ERISA’s prohibited transaction rules by selecting\\nfor inclusion in the Plan investment funds that are affiliated\\nwith Schwab. According to the Complaint, the Schwab-\\naffiliated funds performed poorly but were kept in the Plan\\nsolely to generate fees for Schwab and its affiliates. The\\nComplaint also alleges that members of the Board of\\nDirectors of Charles Schwab & Co. breached their duty to\\nmonitor the Plan fiduciaries who selected the investment\\nfunds for inclusion in the Plan. The Complaint further\\nasserts claims for co-fiduciary breach and knowing\\nparticipation in a breach against various Defendants.\\n\\n In response to the Complaint, Defendants moved to\\ncompel individual arbitration of the asserted claims pursuant\\nto the arbitration agreements in the Plan and the\\nCompensation Plan.\\n\\n On January 18, 2018, the district court denied\\nDefendants’ motion, holding that neither agreement required\\nthe claims asserted in the Complaint to be arbitrated. In the\\ndistrict court’s view, neither of the two arbitration provisions\\napplied to the claims in the Complaint. The district court\\nerroneously held that the arbitration provision in the Plan\\ndocument was inapplicable because the provision was\\n\\f DORMAN V. CHARLES SCHWAB CORP. 7\\n\\nenacted after Dorman’s participation in the Plan ended, and\\nit thus did not bind him. With respect to the Compensation\\nPlan’s arbitration agreement, the district court concluded\\nthat “it is not clear” that the asserted ERISA claims arose out\\nof Dorman’s employment at Schwab as required by that\\nagreement. It also held that the claims were “claims for\\nbenefits” that were expressly carved out of the arbitration\\nagreement in the Compensation Plan.\\n\\n As an alternative basis for denying Defendants’ motion,\\nthe district court held that even if the claims asserted in the\\nComplaint did fall within the scope of one or more of the\\narbitration agreements, the agreements would be\\nunenforceable on two grounds. According to the district\\ncourt, Dorman’s claims were brought “on behalf of the\\nPlan”—not on his own behalf—and without the Plan’s\\nconsent he “cannot waive rights that belong to the Plan, such\\nas the right to file this action in court.” (citing Bowles v.\\nReade, 198 F.3d 752, 760 (9th Cir. 1999) (holding that a plan\\nparticipant cannot settle an ERISA § 502(a)(2) claim without\\nthe plan’s consent). The district court acknowledged that the\\nPlan did consent to arbitration “by virtue of its Plan\\nDocument’s arbitration provision,” but it erroneously held\\nthat consent invalid because the Plan fiduciaries added the\\narbitration provision to the Plan document after they were\\nsued. 2 The district court cited Johnson v. Couturier,\\n572 F.3d 1067 (9th Cir. 2009), which held that plan\\n\\n 2\\n After filing their Notice of Appeal, Defendants sought leave to file\\na motion for partial reconsideration of the district court’s order.\\nDefendants asked the district court to reconsider its ruling that the Plan\\ndocument’s arbitration provision did not take effect until after Dorman\\nceased participating in the Plan. Defendants submitted evidence that the\\narbitration provision took effect while Dorman was still a Plan\\nparticipant; however, the district court denied leave to move for\\nreconsideration.\\n\\f8 DORMAN V. CHARLES SCHWAB CORP.\\n\\nfiduciaries cannot insulate themselves from fiduciary\\nresponsibility by amending a plan document. Id. at 1080. In\\nthe district court’s view, allowing Plan fiduciaries to amend\\nthe Plan document to consent to arbitration would “in a\\nsense, be allowing the fox to guard the henhouse.” (quoting\\nMunro v. Univ. of S. Cal., No. CV 16-6191-VAP (CFEx),\\n2017 WL 1654075, at *6 (C.D. Cal. Mar. 23, 2017)).\\n\\n Finally, the district court held that this court’s decision\\nin Morris v. Ernst & Young, LLP, 834 F.3d 975 (9th Cir.\\n2016), reversed by Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612,\\n1621 (2018), further precludes arbitration. According to the\\ndistrict court, Morris holds that class-action waivers that are\\nrequired as a condition of employment violate the National\\nLabor Relations Act (“NLRA”) and are, therefore,\\nunenforceable. This interlocutory appeal followed.\\n\\n II\\n\\n Over 35 years ago, in Amaro v. Continental Can Co. we\\nwrote that ERISA mandated “minimum standards [for]\\nassuring the equitable character of [ERISA] plans” that\\ncould not be satisfied by arbitral proceedings. 724 F.2d\\nat 752. We reasoned that “[a]rbitrators, many of whom are\\nnot lawyers, lack the competence of courts to interpret and\\napply statutes as Congress intended.” Id. at 750 (internal\\ncitation omitted). In Comer v. Micor, Inc., 436 F.3d 1098\\n(9th Cir. 2006), we acknowledged in dicta that our past\\n“expressed skepticism about the arbitrability of ERISA\\nclaims . . . seem[ed] to have been put to rest by the Supreme\\nCourt’s opinions.” Id. at 1100 (citing Shearson/Am. Express\\nInc. v. McMahon, 482 U.S. 220, 226(1987)).\\n\\n Since Amaro, the Supreme Court has ruled that\\narbitrators are competent to interpret and apply federal\\nstatutes. See, e.g., Am. Express Co., 570 U.S. at 233 (holding\\n\\f DORMAN V. CHARLES SCHWAB CORP. 9\\n\\nthat there is nothing unfair about arbitration—even\\narbitration on an individual basis—as long as individuals can\\nvindicate their statutory rights in the arbitral forum).\\nRecently, in Munro v. Univ. of S. Cal., 896 F.3d 1088 (9th\\nCir. 2018), we noted that “there is considerable force” to the\\nargument that Amaro has been overruled. Id. at 1094 n.1.3\\nWe agree.\\n\\n Generally, a three-judge panel may not overrule a prior\\ndecision of the court. Miller v. Gammie, 335 F.3d 889, 899\\n(9th Cir. 2003) (en banc). If, however, “an intervening\\nSupreme Court decision undermines an existing precedent\\nof the Ninth Circuit, and both cases are closely on point[,]”\\nthe three-judge panel may then overrule prior circuit\\nauthority. Id. The issue decided by the higher court need\\nnot be identical. Id. at 900. The appropriate test is whether\\nthe higher court “undercut the theory or reasoning\\nunderlying the prior circuit precedent in such a way that the\\ncases are clearly irreconcilable.” Id.\\n\\n “[W]here the reasoning or theory of our prior circuit\\nauthority is clearly irreconcilable with the reasoning or\\ntheory of intervening higher authority, a three-judge panel\\nshould consider itself bound by the later and controlling\\nauthority, and should reject the prior circuit opinion as\\nhaving been effectively overruled.” Miller, 335 F.3d at 893.\\nThe holding in American Express Co. that federal statutory\\nclaims are generally arbitrable and arbitrators can\\ncompetently interpret and apply federal statutes, 570 U.S.\\nat 233, constitutes intervening Supreme Court authority that\\n\\n 3\\n Even before Munro, we had begun to question the force of Amaro.\\nSee Comer at 1101(“Curiously, however, we have echoed the doubts\\nexpressed in Amaro without taking account of the intervening Supreme\\nCourt cases.” (collecting cases)).\\n\\f10 DORMAN V. CHARLES SCHWAB CORP.\\n\\nis irreconcilable with Amaro. Amaro, therefore, is no longer\\nbinding precedent.\\n\\n REVERSED and REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4427021/\", \"author_raw\": \"PEARSON, District Judge:\"}]}"},"judge1":{"kind":"string","value":"RONALD M GOULD"},"judge2":{"kind":"string","value":"SANDRA S IKUTA"},"judge3":{"kind":"string","value":"BENITA Y PEARSON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649768/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Michael Dorman v. the Charles Schwab Corporation\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-20\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMICHAEL F. DORMAN, individually No. 18-15281\nas a participant in the SCHWAB PLAN\nRETIREMENT SAVINGS AND D.C. No.\nINVESTMENT PLAN and on behalf of a 4:17-cv-00285-\nclass of all those similarly situated, CW\n Plaintiff-Appellee,\n\n v. OPINION\n\nTHE CHARLES SCHWAB\nCORPORATION; CHARLES SCHWAB &\nCO., INC.; SCHWAB RETIREMENT\nPLAN SERVICES, INC.; CHARLES\nSCHWAB BANK; CHARLES SCHWAB\nINVESTMENT MANAGEMENT, INC.;\nWALTER W. BETTINGER III;\nCHARLES R. SCHWAB; JOSEPH R.\nMARTINETTO; MARTHA TUMA; JAY\nALLEN; DAVE CALLAHAN; JOHN C.\nCLARK,\n Defendants-Appellants.\n\n Appeal from the United States District Court\n for the Northern District of California\n Claudia Wilken, District Judge, Presiding\n\n Argued and Submitted June 14, 2019\n San Francisco, California\n\f2 DORMAN V. CHARLES SCHWAB CORP.\n\n Filed August 20, 2019\n\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\n Judges, and Benita Y. Pearson, * District Judge.\n\n Opinion by Judge Pearson\n\n\n SUMMARY **\n\n\n ERISA / Arbitration\n\n The panel reversed the district court’s order denying\ndefendants’ motion to compel arbitration of claims and\nremanded in a class action suit brought by a former\nparticipant in an ERISA retirement plan, alleging that\ndefendants violated ERISA and breached their fiduciary\nduties by including certain investment funds in the plan.\n\n The panel concluded that Amaro v. Continental Can Co.,\n724 F.2d 747 (9th Cir. 1984), which held that ERISA claims\nare not arbitrable, is no longer good law in light of\nintervening Supreme Court case law, including American\nExpress Co. v. Italian Colors Restaurant, 570 U.S. 228\n(2013).\n\n\n\n\n *\n The Honorable Benita Y. Pearson, United States District Judge for\nthe Northern District of Ohio, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f DORMAN V. CHARLES SCHWAB CORP. 3\n\n The panel addressed other issues in a concurrently filed\nmemorandum disposition.\n\n\n COUNSEL\n\nHoward Shapiro (argued), Stacey C.S. Cerrone, and Tulio D.\nChirinos, Proskauer Rose LLP, New Orleans, Louisiana;\nMyron D. Rumeld, Proskauer Rose LLP, New York, New\nYork; John E. Roberts, Proskauer Rose LLP, Boston,\nMassachusetts; for Defendants-Appellants.\n\nJames Bloom (argued), Todd M. Schneider, and Kyle G.\nBates, Schneider Wallace Cottrell Konecky Wotkyns LLP,\nEmeryville, California; Todd S. Collins, Eric Lechtzin,\nShanon J. Carson, and Ellen T. Noteware, Berger Montague\nPC, Philadelphia, Pennsylvania; for Plaintiff-Appellee.\n\n\n OPINION\n\nPEARSON, District Judge:\n\n Defendants appeal the district court’s denial of their\nmotion to compel individual arbitration in an ERISA action\nfiled by Michael Dorman, a former participant in the Schwab\nRetirement Savings and Investment Plan (the “Plan”).\nDorman participated in a defined contribution 401(k)\nretirement plan through his employment with Charles\nSchwab & Co., Inc. (“Schwab”). In 2017, Dorman filed a\nclass action suit in district court alleging that Defendants\nviolated ERISA and breached their fiduciary duties by\nincluding Schwab-affiliated investment funds in the Plan—\ndespite the funds’ poor performance—to generate fees for\nSchwab and its affiliates. Defendants moved to compel\n\f4 DORMAN V. CHARLES SCHWAB CORP.\n\narbitration pursuant to an arbitration agreement in the Plan.\nThe district court denied the motion, and this interlocutory\nappeal followed.\n\n On appeal, Defendants contend that the district court\nerred by not enforcing the Plan’s arbitration agreement. We\naddress these arguments in a concurrently filed\nmemorandum disposition. But before we can reach the\nparties’ specific contentions, we must first address the\nthreshold question of whether ERISA claims can be subject\nto mandatory arbitration. In so doing, we must revisit our\nholding in Amaro v. Continental Can Co., 724 F.2d 747 (9th\nCir. 1984), in which we held that ERISA claims were not\narbitrable. In light of intervening Supreme Court case law,\nincluding American Express Co. v. Italian Colors\nRestaurant, 570 U.S. 228 (2013), we conclude that our\nholding in Amaro is no longer good law.\n\n I\n\n Michael Dorman was employed at Schwab from\nFebruary 17, 2009, until October 8, 2015. Through his\nemployment, Dorman joined the Plan in 2009, and he\nvoluntarily contributed to his retirement account through\npayroll deductions until he left his employment with\nSchwab. Dorman withdrew his full account balance on\nDecember 18, 2015, and ceased participating in the Plan. 1\n\n In this defined contribution 401(k) retirement plan, Plan\nparticipants are given the choice to allocate their earnings\namong a menu of investment funds, and they may alter their\n 1\n In a separate memorandum filed concurrently with this opinion,\nwe address Defendants’ challenges to the district court’s denial of their\nmotions to compel individual arbitration and for leave to move for\nreconsideration and reverse.\n\f DORMAN V. CHARLES SCHWAB CORP. 5\n\ninvestment allocations at any time. During the relevant\nperiod, the Plan offered as many as 17 different funds in\nwhich participants could choose to invest, including both\nSchwab-affiliated and unaffiliated funds.\n\n In December 2014, the Plan was amended to add an\narbitration provision. That provision took effect on\nJanuary 1, 2015, nine months before Dorman ended his\nemployment at Schwab and nearly a year before he\nterminated his participation in the Plan. The Plan document\nstates that “[a]ny claim, dispute or breach arising out of or in\nany way related to the Plan shall be settled by binding\narbitration . . . .” The arbitration provision includes a waiver\nof class or collective action that requires individual\narbitrations, even if absent the waiver Dorman could have\nrepresented the interests of other Plan participants. It states\nthat any arbitration would be conducted “on an individual\nbasis only, and not on a class, collective or representative\nbasis,” and that Plan participants waive the right to be part\nof any class action. If that waiver of collective action were\nto be held unenforceable, the arbitration provision mandates\nthat “any claim on a class, collective or representative basis\nshall be filed and adjudicated in a court of competent\njurisdiction, and not in arbitration.”\n\n In 2014, Dorman was promoted to financial consultant,\nand he enrolled in the Schwab Investor Financial Consultant\nCompensation Plan (the “Compensation Plan”). By\nenrolling in the Compensation Plan, Dorman agreed to\narbitrate “[a]ny controversy, dispute, or claim arising out of\nor relating to [his] employment . . . .” By its terms, the\narbitration provision encompassed claims that arise out of\n“federal, state, or local law.” The arbitration agreement\ncarved out “claims for benefits” under the Plan and provided\nthat such “claims for benefits” would be resolved pursuant\n\f6 DORMAN V. CHARLES SCHWAB CORP.\n\nto the procedures prescribed by the Plan. The Compensation\nPlan further contains a “Class Action Waiver.”\n\n In June 2017, Dorman filed his First Amended Class\nAction Complaint against Defendants, asserting claims\nunder § 502(a)(2) and (3) of the Employee Retirement\nIncome Security Act of 1974 (“ERISA”), 29 U.S.C.\n§ 1132(a)(2) and (3), and seeking plan-wide relief on behalf\nof a class comprising all participants in, and beneficiaries of,\nthe Plan at any time within six years of the filing of the\nComplaint. The Complaint alleges that various Defendants\nbreached their fiduciary duties of loyalty and prudence and\nviolated ERISA’s prohibited transaction rules by selecting\nfor inclusion in the Plan investment funds that are affiliated\nwith Schwab. According to the Complaint, the Schwab-\naffiliated funds performed poorly but were kept in the Plan\nsolely to generate fees for Schwab and its affiliates. The\nComplaint also alleges that members of the Board of\nDirectors of Charles Schwab & Co. breached their duty to\nmonitor the Plan fiduciaries who selected the investment\nfunds for inclusion in the Plan. The Complaint further\nasserts claims for co-fiduciary breach and knowing\nparticipation in a breach against various Defendants.\n\n In response to the Complaint, Defendants moved to\ncompel individual arbitration of the asserted claims pursuant\nto the arbitration agreements in the Plan and the\nCompensation Plan.\n\n On January 18, 2018, the district court denied\nDefendants’ motion, holding that neither agreement required\nthe claims asserted in the Complaint to be arbitrated. In the\ndistrict court’s view, neither of the two arbitration provisions\napplied to the claims in the Complaint. The district court\nerroneously held that the arbitration provision in the Plan\ndocument was inapplicable because the provision was\n\f DORMAN V. CHARLES SCHWAB CORP. 7\n\nenacted after Dorman’s participation in the Plan ended, and\nit thus did not bind him. With respect to the Compensation\nPlan’s arbitration agreement, the district court concluded\nthat “it is not clear” that the asserted ERISA claims arose out\nof Dorman’s employment at Schwab as required by that\nagreement. It also held that the claims were “claims for\nbenefits” that were expressly carved out of the arbitration\nagreement in the Compensation Plan.\n\n As an alternative basis for denying Defendants’ motion,\nthe district court held that even if the claims asserted in the\nComplaint did fall within the scope of one or more of the\narbitration agreements, the agreements would be\nunenforceable on two grounds. According to the district\ncourt, Dorman’s claims were brought “on behalf of the\nPlan”—not on his own behalf—and without the Plan’s\nconsent he “cannot waive rights that belong to the Plan, such\nas the right to file this action in court.” (citing Bowles v.\nReade, 198 F.3d 752, 760 (9th Cir. 1999) (holding that a plan\nparticipant cannot settle an ERISA § 502(a)(2) claim without\nthe plan’s consent). The district court acknowledged that the\nPlan did consent to arbitration “by virtue of its Plan\nDocument’s arbitration provision,” but it erroneously held\nthat consent invalid because the Plan fiduciaries added the\narbitration provision to the Plan document after they were\nsued. 2 The district court cited Johnson v. Couturier,\n572 F.3d 1067 (9th Cir. 2009), which held that plan\n\n 2\n After filing their Notice of Appeal, Defendants sought leave to file\na motion for partial reconsideration of the district court’s order.\nDefendants asked the district court to reconsider its ruling that the Plan\ndocument’s arbitration provision did not take effect until after Dorman\nceased participating in the Plan. Defendants submitted evidence that the\narbitration provision took effect while Dorman was still a Plan\nparticipant; however, the district court denied leave to move for\nreconsideration.\n\f8 DORMAN V. CHARLES SCHWAB CORP.\n\nfiduciaries cannot insulate themselves from fiduciary\nresponsibility by amending a plan document. Id. at 1080. In\nthe district court’s view, allowing Plan fiduciaries to amend\nthe Plan document to consent to arbitration would “in a\nsense, be allowing the fox to guard the henhouse.” (quoting\nMunro v. Univ. of S. Cal., No. CV 16-6191-VAP (CFEx),\n2017 WL 1654075, at *6 (C.D. Cal. Mar. 23, 2017)).\n\n Finally, the district court held that this court’s decision\nin Morris v. Ernst & Young, LLP, 834 F.3d 975 (9th Cir.\n2016), reversed by Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612,\n1621 (2018), further precludes arbitration. According to the\ndistrict court, Morris holds that class-action waivers that are\nrequired as a condition of employment violate the National\nLabor Relations Act (“NLRA”) and are, therefore,\nunenforceable. This interlocutory appeal followed.\n\n II\n\n Over 35 years ago, in Amaro v. Continental Can Co. we\nwrote that ERISA mandated “minimum standards [for]\nassuring the equitable character of [ERISA] plans” that\ncould not be satisfied by arbitral proceedings. 724 F.2d\nat 752. We reasoned that “[a]rbitrators, many of whom are\nnot lawyers, lack the competence of courts to interpret and\napply statutes as Congress intended.” Id. at 750 (internal\ncitation omitted). In Comer v. Micor, Inc., 436 F.3d 1098\n(9th Cir. 2006), we acknowledged in dicta that our past\n“expressed skepticism about the arbitrability of ERISA\nclaims . . . seem[ed] to have been put to rest by the Supreme\nCourt’s opinions.” Id. at 1100 (citing Shearson/Am. Express\nInc. v. McMahon, 482 U.S. 220, 226(1987)).\n\n Since Amaro, the Supreme Court has ruled that\narbitrators are competent to interpret and apply federal\nstatutes. See, e.g., Am. Express Co., 570 U.S. at 233 (holding\n\f DORMAN V. CHARLES SCHWAB CORP. 9\n\nthat there is nothing unfair about arbitration—even\narbitration on an individual basis—as long as individuals can\nvindicate their statutory rights in the arbitral forum).\nRecently, in Munro v. Univ. of S. Cal., 896 F.3d 1088 (9th\nCir. 2018), we noted that “there is considerable force” to the\nargument that Amaro has been overruled. Id. at 1094 n.1.3\nWe agree.\n\n Generally, a three-judge panel may not overrule a prior\ndecision of the court. Miller v. Gammie, 335 F.3d 889, 899\n(9th Cir. 2003) (en banc). If, however, “an intervening\nSupreme Court decision undermines an existing precedent\nof the Ninth Circuit, and both cases are closely on point[,]”\nthe three-judge panel may then overrule prior circuit\nauthority. Id. The issue decided by the higher court need\nnot be identical. Id. at 900. The appropriate test is whether\nthe higher court “undercut the theory or reasoning\nunderlying the prior circuit precedent in such a way that the\ncases are clearly irreconcilable.” Id.\n\n “[W]here the reasoning or theory of our prior circuit\nauthority is clearly irreconcilable with the reasoning or\ntheory of intervening higher authority, a three-judge panel\nshould consider itself bound by the later and controlling\nauthority, and should reject the prior circuit opinion as\nhaving been effectively overruled.” Miller, 335 F.3d at 893.\nThe holding in American Express Co. that federal statutory\nclaims are generally arbitrable and arbitrators can\ncompetently interpret and apply federal statutes, 570 U.S.\nat 233, constitutes intervening Supreme Court authority that\n\n 3\n Even before Munro, we had begun to question the force of Amaro.\nSee Comer at 1101(“Curiously, however, we have echoed the doubts\nexpressed in Amaro without taking account of the intervening Supreme\nCourt cases.” (collecting cases)).\n\f10 DORMAN V. CHARLES SCHWAB CORP.\n\nis irreconcilable with Amaro. Amaro, therefore, is no longer\nbinding precedent.\n\n REVERSED and REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Michael Dorman v. the Charles Schwab Corporation\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-20\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nMICHAEL F. DORMAN, individually No. 18-15281\\nas a participant in the SCHWAB PLAN\\nRETIREMENT SAVINGS AND D.C. No.\\nINVESTMENT PLAN and on behalf of a 4:17-cv-00285-\\nclass of all those similarly situated, CW\\n Plaintiff-Appellee,\\n\\n v. OPINION\\n\\nTHE CHARLES SCHWAB\\nCORPORATION; CHARLES SCHWAB &\\nCO., INC.; SCHWAB RETIREMENT\\nPLAN SERVICES, INC.; CHARLES\\nSCHWAB BANK; CHARLES SCHWAB\\nINVESTMENT MANAGEMENT, INC.;\\nWALTER W. BETTINGER III;\\nCHARLES R. SCHWAB; JOSEPH R.\\nMARTINETTO; MARTHA TUMA; JAY\\nALLEN; DAVE CALLAHAN; JOHN C.\\nCLARK,\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n Claudia Wilken, District Judge, Presiding\\n\\n Argued and Submitted June 14, 2019\\n San Francisco, California\\n\\f2 DORMAN V. CHARLES SCHWAB CORP.\\n\\n Filed August 20, 2019\\n\\n Before: Ronald M. Gould and Sandra S. Ikuta, Circuit\\n Judges, and Benita Y. Pearson, * District Judge.\\n\\n Opinion by Judge Pearson\\n\\n\\n SUMMARY **\\n\\n\\n ERISA / Arbitration\\n\\n The panel reversed the district court’s order denying\\ndefendants’ motion to compel arbitration of claims and\\nremanded in a class action suit brought by a former\\nparticipant in an ERISA retirement plan, alleging that\\ndefendants violated ERISA and breached their fiduciary\\nduties by including certain investment funds in the plan.\\n\\n The panel concluded that Amaro v. Continental Can Co.,\\n724 F.2d 747 (9th Cir. 1984), which held that ERISA claims\\nare not arbitrable, is no longer good law in light of\\nintervening Supreme Court case law, including American\\nExpress Co. v. Italian Colors Restaurant, 570 U.S. 228\\n(2013).\\n\\n\\n\\n\\n *\\n The Honorable Benita Y. Pearson, United States District Judge for\\nthe Northern District of Ohio, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f DORMAN V. CHARLES SCHWAB CORP. 3\\n\\n The panel addressed other issues in a concurrently filed\\nmemorandum disposition.\\n\\n\\n COUNSEL\\n\\nHoward Shapiro (argued), Stacey C.S. Cerrone, and Tulio D.\\nChirinos, Proskauer Rose LLP, New Orleans, Louisiana;\\nMyron D. Rumeld, Proskauer Rose LLP, New York, New\\nYork; John E. Roberts, Proskauer Rose LLP, Boston,\\nMassachusetts; for Defendants-Appellants.\\n\\nJames Bloom (argued), Todd M. Schneider, and Kyle G.\\nBates, Schneider Wallace Cottrell Konecky Wotkyns LLP,\\nEmeryville, California; Todd S. Collins, Eric Lechtzin,\\nShanon J. Carson, and Ellen T. Noteware, Berger Montague\\nPC, Philadelphia, Pennsylvania; for Plaintiff-Appellee.\\n\\n\\n OPINION\\n\\nPEARSON, District Judge:\\n\\n Defendants appeal the district court’s denial of their\\nmotion to compel individual arbitration in an ERISA action\\nfiled by Michael Dorman, a former participant in the Schwab\\nRetirement Savings and Investment Plan (the “Plan”).\\nDorman participated in a defined contribution 401(k)\\nretirement plan through his employment with Charles\\nSchwab & Co., Inc. (“Schwab”). In 2017, Dorman filed a\\nclass action suit in district court alleging that Defendants\\nviolated ERISA and breached their fiduciary duties by\\nincluding Schwab-affiliated investment funds in the Plan—\\ndespite the funds’ poor performance—to generate fees for\\nSchwab and its affiliates. Defendants moved to compel\\n\\f4 DORMAN V. CHARLES SCHWAB CORP.\\n\\narbitration pursuant to an arbitration agreement in the Plan.\\nThe district court denied the motion, and this interlocutory\\nappeal followed.\\n\\n On appeal, Defendants contend that the district court\\nerred by not enforcing the Plan’s arbitration agreement. We\\naddress these arguments in a concurrently filed\\nmemorandum disposition. But before we can reach the\\nparties’ specific contentions, we must first address the\\nthreshold question of whether ERISA claims can be subject\\nto mandatory arbitration. In so doing, we must revisit our\\nholding in Amaro v. Continental Can Co., 724 F.2d 747 (9th\\nCir. 1984), in which we held that ERISA claims were not\\narbitrable. In light of intervening Supreme Court case law,\\nincluding American Express Co. v. Italian Colors\\nRestaurant, 570 U.S. 228 (2013), we conclude that our\\nholding in Amaro is no longer good law.\\n\\n I\\n\\n Michael Dorman was employed at Schwab from\\nFebruary 17, 2009, until October 8, 2015. Through his\\nemployment, Dorman joined the Plan in 2009, and he\\nvoluntarily contributed to his retirement account through\\npayroll deductions until he left his employment with\\nSchwab. Dorman withdrew his full account balance on\\nDecember 18, 2015, and ceased participating in the Plan. 1\\n\\n In this defined contribution 401(k) retirement plan, Plan\\nparticipants are given the choice to allocate their earnings\\namong a menu of investment funds, and they may alter their\\n 1\\n In a separate memorandum filed concurrently with this opinion,\\nwe address Defendants’ challenges to the district court’s denial of their\\nmotions to compel individual arbitration and for leave to move for\\nreconsideration and reverse.\\n\\f DORMAN V. CHARLES SCHWAB CORP. 5\\n\\ninvestment allocations at any time. During the relevant\\nperiod, the Plan offered as many as 17 different funds in\\nwhich participants could choose to invest, including both\\nSchwab-affiliated and unaffiliated funds.\\n\\n In December 2014, the Plan was amended to add an\\narbitration provision. That provision took effect on\\nJanuary 1, 2015, nine months before Dorman ended his\\nemployment at Schwab and nearly a year before he\\nterminated his participation in the Plan. The Plan document\\nstates that “[a]ny claim, dispute or breach arising out of or in\\nany way related to the Plan shall be settled by binding\\narbitration . . . .” The arbitration provision includes a waiver\\nof class or collective action that requires individual\\narbitrations, even if absent the waiver Dorman could have\\nrepresented the interests of other Plan participants. It states\\nthat any arbitration would be conducted “on an individual\\nbasis only, and not on a class, collective or representative\\nbasis,” and that Plan participants waive the right to be part\\nof any class action. If that waiver of collective action were\\nto be held unenforceable, the arbitration provision mandates\\nthat “any claim on a class, collective or representative basis\\nshall be filed and adjudicated in a court of competent\\njurisdiction, and not in arbitration.”\\n\\n In 2014, Dorman was promoted to financial consultant,\\nand he enrolled in the Schwab Investor Financial Consultant\\nCompensation Plan (the “Compensation Plan”). By\\nenrolling in the Compensation Plan, Dorman agreed to\\narbitrate “[a]ny controversy, dispute, or claim arising out of\\nor relating to [his] employment . . . .” By its terms, the\\narbitration provision encompassed claims that arise out of\\n“federal, state, or local law.” The arbitration agreement\\ncarved out “claims for benefits” under the Plan and provided\\nthat such “claims for benefits” would be resolved pursuant\\n\\f6 DORMAN V. CHARLES SCHWAB CORP.\\n\\nto the procedures prescribed by the Plan. The Compensation\\nPlan further contains a “Class Action Waiver.”\\n\\n In June 2017, Dorman filed his First Amended Class\\nAction Complaint against Defendants, asserting claims\\nunder § 502(a)(2) and (3) of the Employee Retirement\\nIncome Security Act of 1974 (“ERISA”), 29 U.S.C.\\n§ 1132(a)(2) and (3), and seeking plan-wide relief on behalf\\nof a class comprising all participants in, and beneficiaries of,\\nthe Plan at any time within six years of the filing of the\\nComplaint. The Complaint alleges that various Defendants\\nbreached their fiduciary duties of loyalty and prudence and\\nviolated ERISA’s prohibited transaction rules by selecting\\nfor inclusion in the Plan investment funds that are affiliated\\nwith Schwab. According to the Complaint, the Schwab-\\naffiliated funds performed poorly but were kept in the Plan\\nsolely to generate fees for Schwab and its affiliates. The\\nComplaint also alleges that members of the Board of\\nDirectors of Charles Schwab & Co. breached their duty to\\nmonitor the Plan fiduciaries who selected the investment\\nfunds for inclusion in the Plan. The Complaint further\\nasserts claims for co-fiduciary breach and knowing\\nparticipation in a breach against various Defendants.\\n\\n In response to the Complaint, Defendants moved to\\ncompel individual arbitration of the asserted claims pursuant\\nto the arbitration agreements in the Plan and the\\nCompensation Plan.\\n\\n On January 18, 2018, the district court denied\\nDefendants’ motion, holding that neither agreement required\\nthe claims asserted in the Complaint to be arbitrated. In the\\ndistrict court’s view, neither of the two arbitration provisions\\napplied to the claims in the Complaint. The district court\\nerroneously held that the arbitration provision in the Plan\\ndocument was inapplicable because the provision was\\n\\f DORMAN V. CHARLES SCHWAB CORP. 7\\n\\nenacted after Dorman’s participation in the Plan ended, and\\nit thus did not bind him. With respect to the Compensation\\nPlan’s arbitration agreement, the district court concluded\\nthat “it is not clear” that the asserted ERISA claims arose out\\nof Dorman’s employment at Schwab as required by that\\nagreement. It also held that the claims were “claims for\\nbenefits” that were expressly carved out of the arbitration\\nagreement in the Compensation Plan.\\n\\n As an alternative basis for denying Defendants’ motion,\\nthe district court held that even if the claims asserted in the\\nComplaint did fall within the scope of one or more of the\\narbitration agreements, the agreements would be\\nunenforceable on two grounds. According to the district\\ncourt, Dorman’s claims were brought “on behalf of the\\nPlan”—not on his own behalf—and without the Plan’s\\nconsent he “cannot waive rights that belong to the Plan, such\\nas the right to file this action in court.” (citing Bowles v.\\nReade, 198 F.3d 752, 760 (9th Cir. 1999) (holding that a plan\\nparticipant cannot settle an ERISA § 502(a)(2) claim without\\nthe plan’s consent). The district court acknowledged that the\\nPlan did consent to arbitration “by virtue of its Plan\\nDocument’s arbitration provision,” but it erroneously held\\nthat consent invalid because the Plan fiduciaries added the\\narbitration provision to the Plan document after they were\\nsued. 2 The district court cited Johnson v. Couturier,\\n572 F.3d 1067 (9th Cir. 2009), which held that plan\\n\\n 2\\n After filing their Notice of Appeal, Defendants sought leave to file\\na motion for partial reconsideration of the district court’s order.\\nDefendants asked the district court to reconsider its ruling that the Plan\\ndocument’s arbitration provision did not take effect until after Dorman\\nceased participating in the Plan. Defendants submitted evidence that the\\narbitration provision took effect while Dorman was still a Plan\\nparticipant; however, the district court denied leave to move for\\nreconsideration.\\n\\f8 DORMAN V. CHARLES SCHWAB CORP.\\n\\nfiduciaries cannot insulate themselves from fiduciary\\nresponsibility by amending a plan document. Id. at 1080. In\\nthe district court’s view, allowing Plan fiduciaries to amend\\nthe Plan document to consent to arbitration would “in a\\nsense, be allowing the fox to guard the henhouse.” (quoting\\nMunro v. Univ. of S. Cal., No. CV 16-6191-VAP (CFEx),\\n2017 WL 1654075, at *6 (C.D. Cal. Mar. 23, 2017)).\\n\\n Finally, the district court held that this court’s decision\\nin Morris v. Ernst & Young, LLP, 834 F.3d 975 (9th Cir.\\n2016), reversed by Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612,\\n1621 (2018), further precludes arbitration. According to the\\ndistrict court, Morris holds that class-action waivers that are\\nrequired as a condition of employment violate the National\\nLabor Relations Act (“NLRA”) and are, therefore,\\nunenforceable. This interlocutory appeal followed.\\n\\n II\\n\\n Over 35 years ago, in Amaro v. Continental Can Co. we\\nwrote that ERISA mandated “minimum standards [for]\\nassuring the equitable character of [ERISA] plans” that\\ncould not be satisfied by arbitral proceedings. 724 F.2d\\nat 752. We reasoned that “[a]rbitrators, many of whom are\\nnot lawyers, lack the competence of courts to interpret and\\napply statutes as Congress intended.” Id. at 750 (internal\\ncitation omitted). In Comer v. Micor, Inc., 436 F.3d 1098\\n(9th Cir. 2006), we acknowledged in dicta that our past\\n“expressed skepticism about the arbitrability of ERISA\\nclaims . . . seem[ed] to have been put to rest by the Supreme\\nCourt’s opinions.” Id. at 1100 (citing Shearson/Am. Express\\nInc. v. McMahon, 482 U.S. 220, 226(1987)).\\n\\n Since Amaro, the Supreme Court has ruled that\\narbitrators are competent to interpret and apply federal\\nstatutes. See, e.g., Am. Express Co., 570 U.S. at 233 (holding\\n\\f DORMAN V. CHARLES SCHWAB CORP. 9\\n\\nthat there is nothing unfair about arbitration—even\\narbitration on an individual basis—as long as individuals can\\nvindicate their statutory rights in the arbitral forum).\\nRecently, in Munro v. Univ. of S. Cal., 896 F.3d 1088 (9th\\nCir. 2018), we noted that “there is considerable force” to the\\nargument that Amaro has been overruled. Id. at 1094 n.1.3\\nWe agree.\\n\\n Generally, a three-judge panel may not overrule a prior\\ndecision of the court. Miller v. Gammie, 335 F.3d 889, 899\\n(9th Cir. 2003) (en banc). If, however, “an intervening\\nSupreme Court decision undermines an existing precedent\\nof the Ninth Circuit, and both cases are closely on point[,]”\\nthe three-judge panel may then overrule prior circuit\\nauthority. Id. The issue decided by the higher court need\\nnot be identical. Id. at 900. The appropriate test is whether\\nthe higher court “undercut the theory or reasoning\\nunderlying the prior circuit precedent in such a way that the\\ncases are clearly irreconcilable.” Id.\\n\\n “[W]here the reasoning or theory of our prior circuit\\nauthority is clearly irreconcilable with the reasoning or\\ntheory of intervening higher authority, a three-judge panel\\nshould consider itself bound by the later and controlling\\nauthority, and should reject the prior circuit opinion as\\nhaving been effectively overruled.” Miller, 335 F.3d at 893.\\nThe holding in American Express Co. that federal statutory\\nclaims are generally arbitrable and arbitrators can\\ncompetently interpret and apply federal statutes, 570 U.S.\\nat 233, constitutes intervening Supreme Court authority that\\n\\n 3\\n Even before Munro, we had begun to question the force of Amaro.\\nSee Comer at 1101(“Curiously, however, we have echoed the doubts\\nexpressed in Amaro without taking account of the intervening Supreme\\nCourt cases.” (collecting cases)).\\n\\f10 DORMAN V. CHARLES SCHWAB CORP.\\n\\nis irreconcilable with Amaro. Amaro, therefore, is no longer\\nbinding precedent.\\n\\n REVERSED and REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4084,"cells":{"id":{"kind":"number","value":4649769,"string":"4,649,769"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Kristanalea Dyroff v. the Ultimate Software Group"},"decision_date":{"kind":"string","value":"2019-08-20"},"docket_number":{"kind":"string","value":"18-15175"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Dorothy W. Nelson, Johnnie B. Rawlinson, and Carlos T. Bea, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"D.W. NELSON, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nKRISTANALEA DYROFF, individually No. 18-15175\\nand on behalf of the estate of Wesley\\nGreer, deceased, D.C. No.\\n Plaintiff-Appellant, 3:17-cv-05359-\\n LB\\n v.\\n\\nTHE ULTIMATE SOFTWARE GROUP, OPINION\\nINC.,\\n Defendant-Appellee.\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n Laurel D. Beeler, Magistrate Judge, Presiding\\n\\n Argued and Submitted June 4, 2019\\n Seattle, Washington\\n\\n Filed August 20, 2019\\n\\n Before: Dorothy W. Nelson, Johnnie B. Rawlinson,\\n and Carlos T. Bea, Circuit Judges.\\n\\n Opinion by Judge D.W. Nelson\\n\\f2 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n SUMMARY *\\n\\n\\n Communications Decency Act\\n\\n The panel affirmed the district court’s dismissal, based\\non The Ultimate Software Group’s immunity from liability\\nunder Section 230 of the Communications Decency Act, of\\nan action alleging claims concerning Ultimate Software’s\\nrole in the death of plaintiff’s son.\\n\\n Ultimate Software was the operator of the Experience\\nProject website, which allegedly facilitated illegal drug\\nsales. Plaintiff’s son died of a drug overdose after buying\\ndrugs from a fellow Experience Project user.\\n\\n The panel held that Ultimate Software satisfied all three\\nprongs of the test for immunity under Section 230 of the\\nCommunications Decency Act. Specifically, the panel held\\nthat Ultimate Software was an interactive computer service\\nbecause it did not create or publish its own content under the\\nplain language of the statute. The panel also held that\\nplaintiff’s son treated Ultimate Software as a publisher or\\nspeaker of other’s information or content. Finally, the panel\\nheld that Ultimate Software published information/content\\nprovided by another information content provider where the\\ncontent at issue was created by plaintiff’s son and his drug\\ndealer. The panel rejected plaintiff’s argument that a website\\ndevelops content if it manipulates the content in a unique\\nway through content-neutral tools.\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 3\\n\\n The panel held that plaintiff did not plead sufficient facts\\nto show that Ultimate Software colluded with drug dealers\\non Experience Project. The panel held that plaintiff’s\\nallegation that user anonymity equaled promoting drug\\ntransactions was not plausible. The panel concluded that the\\ndistrict court was right to dismiss all claims related to this\\nsupposed theory of liability because Ultimate Software was\\nimmune under Section 230 of the Communications Decency\\nAct.\\n\\n The panel held that Ultimate Software did not owe a duty\\nof care to plaintiff’s son because Experience Project’s\\nfeatures amounted to content-neutral functions that did not\\ncreate a risk of harm. The panel rejected plaintiff’s claim\\nthat misfeasance by Ultimate Software created a duty to\\nplaintiff’s son.\\n\\n\\n COUNSEL\\n\\nDavid F. Slade (argued), Carney Bates & Pulliam PLLC,\\nLittle Rock, Arkansas; Sin-Ting Mary Liu, Aylstock Witkin\\nKreis & Overholtz PLLC, Alameda, California; for Plaintiff-\\nAppellant.\\n\\nJeffry A. Miller (argued) and Scott M. Schoenwald, Lewis\\nBrisbois Bisgaard & Smith LLP, San Diego, California;\\nShawn A. Tolliver, David E. Russo, and Justin S. Kim,\\nLewis Brisbois Bisgaard & Smith LLP, San Diego,\\nCalifornia; for Defendant-Appellee.\\n\\f4 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n OPINION\\n\\nD.W. NELSON, Circuit Judge:\\n\\n Plaintiff Kristanalea Dyroff appeals the district court’s\\ndismissal of her claims against Defendant The Ultimate\\nSoftware Group (“Ultimate Software”), operator of the\\nExperience Project website, for its alleged role in the death\\nof her son, Wesley Greer. While the circumstances and facts\\nof this case are no doubt tragic, we find that Ultimate\\nSoftware is immune from liability under Section 230 of the\\nCommunications Decency Act. We therefore affirm.\\n\\n BACKGROUND\\n\\n This being an appeal from a motion to dismiss, we\\ndescribe the case as Plaintiff presents it. We take her\\nplausible allegations as true and draw all reasonable\\ninferences in her favor.\\n\\n Experience Project was a social networking website\\nmade up of various online communities or groups where\\nusers anonymously shared their first-person experiences,\\nposted and answered questions, and interacted with other\\nusers about different topics. The site did not limit or promote\\nthe types of experiences users shared. The site’s “blank box”\\napproach to user content resulted in an array of topics and\\nforums ranging from “I like dogs” and “I am going to\\nStanford” to “I have lung cancer” and “I Love Heroin.”\\n\\n Users registered with the site anonymously; in other\\nwords, the site did not collect users’ identifying information,\\nincluding name, phone number, or mailing address. The\\nsite’s operator, Ultimate Software, believed that anonymity\\nwould promote users to share more personal and authentic\\nexperiences without inhibition. Experience Project’s\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 5\\n\\nfounder stated, “We don’t want to know [users’] real name,\\ntheir phone number, what town they’re from.” Id. “The\\nimpetus behind this policy [of anonymity] was to encourage\\nusers to share experiences with the least amount of inhibition\\npossible. The greater the anonymity, the more ‘honest’ the\\npost . . . .”\\n\\n Experience Project was live from 2007 until March\\n2016, during which its users shared 67 million experiences,\\nmade 15 million connections, and asked 5 million questions.\\nUsers could join groups and the site also recommended\\ngroups for users to join, based on the content of their posts\\nand other attributes, using machine-learning algorithms.\\nWhen a user posted content to a group, the site would send\\nan email notification to the other users active in that group.\\nThe site generated revenue through advertisements and the\\nsale of tokens that users used to post questions to other users\\nin their groups.\\n\\n Some of the site’s functions, including user anonymity\\nand grouping, facilitated illegal drug sales. Wesley Geer\\nwas involved in one such transaction, which turned fatal.\\nWesley suffered from drug addiction, which began when a\\ndoctor overprescribed him opioid pain killers after a serious\\nsports-related injury. After several unsuccessful\\nrehabilitation attempts, Wesley bought what he believed to\\nbe heroin from a fellow Experience Project user. Wesley\\nposted in a heroin-related group, “where can i [sic] score\\nheroin in jacksonville, fl.” The site sent him an email\\nnotification when another user, Hugo Margenat-Castro or\\n“Potheadjuice,” an Orlando-based drug dealer, posted in the\\nsame group. Wesley and Margenat-Castro connected off the\\nsite and Wesley bought heroin from Margenat-Castro on\\nAugust 18, 2015.\\n\\f6 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n Wesley died the next day from fentanyl toxicity. He did\\nnot know that the heroin Margenat-Castro sold him was\\nlaced with fentanyl. Margenat-Castro was ultimately\\narrested and prosecuted. He pleaded guilty in March 2017\\nadmitting that he sold heroin laced with fentanyl while active\\non Experience Project.\\n\\n In March 2016, Experience Project announced, in an\\nopen letter to its users, that it was shutting down. The letter\\nexpressed concern for the future of online privacy because\\nof government overreach. It stated that the site always\\nsupported proper law enforcement efforts but recognized\\nthat it did not have the resources to respond to increased\\ngovernment information requests. The site shut down on\\nApril 21, 2016.\\n\\n Plaintiff Kristanalea Dyroff, Wesley Greer’s mother,\\nfiled a complaint in San Francisco Superior Court. She\\nalleges that Ultimate Software: (1) allowed users to traffic\\nanonymously in illegal, deadly narcotics and to create\\ngroups dedicated to their sale and use; (2) steered users to\\nadditional groups dedicated to the sale and use of narcotics;\\n(3) sent users alerts to posts within groups that were\\ndedicated to the sale and use of narcotics; (4) permitted users\\nto remain active accountholders despite evidence that they\\nopenly engaged in drug trafficking and that law enforcement\\nhad undertaken related investigations; and (5) demonstrated\\nantipathy toward law enforcement efforts to stop illegal\\nactivity on Experience Project.\\n\\n Ultimate Software removed the action from state court\\nbased on diversity jurisdiction and filed a motion to dismiss\\nall claims under Federal Rule of Civil Procedure 12(b)(6).\\nThe district court granted the motion without prejudice.\\nDyroff filed a notice stating that she would not file an\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 7\\n\\namended complaint and asked the district court to enter\\njudgement. Dyroff timely appealed the judgment.\\n\\n STANDARD OF REVIEW\\n\\n We review de novo both a district court order dismissing\\na plaintiff’s claims pursuant to Federal Rule of Civil\\nProcedure 12(b)(6) and questions of statutory interpretation.\\nFields v. Twitter, Inc., 881 F.3d 739, 743 (9th Cir. 2018).\\nThe Court must “accept all factual allegations in the\\ncomplaint as true and construe the pleadings in the light most\\nfavorable to the nonmoving party.” Rowe v. Educ. Credit\\nMgmt. Corp., 559 F.3d 1028, 1029–30 (9th Cir. 2009). Only\\na complaint that states a plausible claim for relief may\\nsurvive a motion to dismiss. Ashcroft v. Iqbal, 556 U.S. 662,\\n678 (2009). Plausibility exists when a court may “draw the\\nreasonable inference that the defendant is liable for the\\nmisconduct alleged.” Id.\\n\\n DISCUSSION\\n\\n Plaintiff argues that in granting the motion to dismiss,\\nthe district court made three errors. First, she argues, the\\ndistrict court erred when it held that Communications\\nDecency Act (CDA) Section 230 immunizes Defendant\\nUltimate Software. Plaintiff reasons that Ultimate Software,\\nas the operator of the Experience Project website, was an\\ninformation content provider, as defined by the statute,\\nbecause its recommendation and notification functions were\\n“specifically designed to make subjective, editorial\\ndecisions about users based on their posts.” Second,\\naccording to Plaintiff, the district court erred when it found\\nthat her allegations of collusion between Ultimate Software\\nand drug dealers using Experience Project were not\\nplausible. Her third argument is that the district court erred\\nin finding that Ultimate Software owed no duty of care to her\\n\\f8 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\nson, Wesley Greer, an Experience Project user. We affirm\\nbecause the district court did not err in any of these respects.\\n\\nI. CDA Section 230 Immunizes Ultimate Software from\\n Plaintiff’s Claims\\n\\n The CDA provides that website operators are immune\\nfrom liability for third-party information (or content, like the\\nposts on Experience Project) unless the website operator “is\\nresponsible, in whole or in part, for the creation or\\ndevelopment of [the] information.” 47 U.S.C. §§ 230(c)(1)\\n& (f)(3). Ultimate Software did not create content on\\nExperience Project, in whole or in part. Accordingly,\\nUltimate Software, as the operator of Experience Project, is\\nimmune from liability under the CDA because its functions,\\nincluding recommendations and notifications, were content-\\nneutral tools used to facilitate communications. See Fair\\nHous. Council of San Fernando Valley v. Roommates.com,\\n521 F.3d 1157, 1167–69 (9th Cir 2008) (en banc).\\n\\n A. Scope of CDA Section 230 Immunity\\n\\n The CDA instructs us that “[n]o provider or user of an\\ninteractive computer service shall be treated as the publisher\\nor speaker of any information provided by another\\ninformation content provider.” 47 U.S.C. § 230(c)(1)\\n(emphasis added). The CDA defines an “interactive\\ncomputer service” as\\n\\n [A]ny information service, system, or access\\n software provider that provides or enables\\n computer access by multiple users to a\\n computer server, including specifically a\\n service or system that provides access to the\\n Internet and such systems operated or\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 9\\n\\n services offered by libraries or educational\\n institutions.\\n\\n47 U.S.C. § 230(f)(2).\\n\\n On the other hand, an “information content provider” is\\n\\n [A]ny person or entity that is responsible, in\\n whole or in part, for the creation or\\n development of information provided\\n through the Internet or any other interactive\\n computer service.\\n\\n47 U.S.C. § 230(f)(3).\\n\\n “The prototypical service qualifying for [CDA]\\nimmunity is an online messaging board (or bulletin board)\\non which Internet subscribers post comments and respond to\\ncomments posted by others.” Kimzey v. Yelp! Inc., 836 F.3d\\n1263, 1266 (9th Cir. 2016) (internal quotations omitted). In\\nother words, a website like Experience Project. Taking the\\nrelevant statutory definitions and case law in account, it\\nbecomes clear that, in general, Section 230(c)(1) “protects\\nwebsites from liability [under state or local law] for material\\nposted on the[ir] website[s] by someone else.” Doe v.\\nInternet Brands, Inc., 824 F.3d 846, 850 (9th Cir. 2016); see\\nalso 47 U.S.C. § 230(e)(3).\\n\\n Combining the above principles, in Barnes v. Yahoo!,\\nInc., we created three-prong test for Section 230 immunity.\\n570 F.3d 1096, 1100 (9th Cir. 2009). Immunity from liability\\nexists for “(1) a provider or user of an interactive computer\\nservice (2) whom a plaintiff seeks to treat, under a state law\\ncause of action, as a publisher or speaker (3) of information\\nprovided by another information content provider.” Id.\\nat 1100–01. When a plaintiff cannot allege enough facts to\\n\\f10 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\novercome Section 230 immunity, a plaintiff’s claims should\\nbe dismissed. See Kimzey, 836 F.3d at 1268–71. Ultimate\\nSoftware satisfies all three prongs of the test.\\n\\n B. Section 230 Immunity—The Barnes test\\n\\n 1. Defendant is an Interactive Computer Service\\n\\n We interpret the term “interactive computer service”\\nexpansively. Kimzey, 836 F.3d at 1268. Ultimate Software\\nwas an interactive computer service because it did not create\\nor publish its own content under the plain language of the\\nstatute. Rather, Ultimate Software published Experience\\nProject users’ posts and did not materially contribute to its\\nusers’ posts.\\n\\n Millions of users, including Plaintiff’s son, Wesley\\nGreer, set up accounts on Experience Project, a website, to\\ncommunicate with each other. Websites are the most\\ncommon interactive computer services. Kimzey, 836 F.3d at\\n1268; see also Roommates.com, 521 F.3d at 1162 n.6\\n(“[t]oday, the most common interactive computer services\\nare websites”).\\n\\n No binding legal authority supports Plaintiff’s\\ncontention that Ultimate Software became an information\\ncontent provider, losing its Section 230 immunity, by\\nfacilitating communication on Experience Project through\\ncontent-neutral website functions like group\\nrecommendations and post notifications. Ultimate Software,\\ntherefore, satisfies the first prong.\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 11\\n\\n 2. Plaintiff Treats Ultimate Software as a\\n Publisher or Speaker of Other’s\\n Information/Content\\n\\n An interactive computer service, like Ultimate Software,\\ncan also be an information content provider, but that is only\\nrelevant, for the purposes of Section 230 immunity, if the\\nwebsite it operates creates or develops the specific content\\nat issue. Carafano v. Metrosplash.com, Inc., 339 F.3d 1119,\\n1124 (9th Cir. 2003). Here, Ultimate Software was not an\\ninformation content provider because it did not create or\\ndevelop information (or content). 47 U.S.C. § 230(f)(3).\\nRather, it published information created or developed by\\nthird parties. Specifically, Experience Project did not create\\nor develop the posts that led to Greer’s death. Rather, it was\\nGreer, himself, who posted “where can i [sic] score heroin\\nin jacksonville, fl” on Experience Project. And it was the\\ndrug dealer, Margenat-Castro, who posted in response to\\nGreer’s post.\\n\\n It is true that Ultimate Software used features and\\nfunctions, including algorithms, to analyze user posts on\\nExperience Project and recommended other user groups.\\nThis includes the heroin-related discussion group to which\\nGreer posted and (through its emails and push notifications)\\nto the drug dealer who sold him the fentanyl-laced heroin.\\nPlaintiff, however, cannot plead around Section 230\\nimmunity by framing these website features as content. We\\nhave held that what matters is whether the claims “inherently\\nrequire[] the court to treat the defendant as the ‘publisher or\\nspeaker’ of content provided by another.” Barnes, 570 F.3d\\nat 1102. If they do, then Section 230(c)(1) provides\\nimmunity from liability. Id.\\n\\n By recommending user groups and sending email\\nnotifications, Ultimate Software, through its Experience\\n\\f12 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\nProject website, was acting as a publisher of others’ content.\\nThese functions—recommendations and notifications—are\\ntools meant to facilitate the communication and content of\\nothers. They are not content in and of themselves.\\n\\n Our recent decision, HomeAway.com, Inc. v. City of\\nSanta Monica, 918 F.3d 676 (9th Cir. 2019) is of no help to\\nPlaintiff. There, the City of Santa Monica required short-\\nterm vacation rentals to be licensed and imposed liability on\\nvacation rental hosting platforms—HomeAway.com and\\nAirbnb—that facilitated unlicensed short-term vacation\\nrentals. Id. at 680. The platforms sued, alleging, among other\\nthings, that Section 230 immunized them from liability. Id.\\nWe found that HomeAway.com and Airbnb did not meet the\\nsecond prong of the Barnes test because the Santa Monica\\nordinance did not “proscribe, mandate, or even discuss the\\ncontent of the [website] listings” and required only that the\\nwebsite’s transactions involve licensed properties. Id. at 683.\\nIn other words, the vacation rental platforms did not face\\nliability for the content of their listings; rather liability arose\\nfrom facilitating unlicensed booking transactions.\\n\\n Ultimate Software, therefore, satisfies the second prong\\nof the Barnes test.\\n\\n 3. Ultimate Software Published\\n Information/Content Provided by Another\\n Information Content Provider\\n\\n The third prong is also met because, as stated previously\\nand as detailed in Plaintiff’s complaint, the content at issue\\nwas created and developed by Greer and his drug dealer.\\nPlaintiff’s content “manipulation” theory is without support\\nin the statute and case law. First, Plaintiff misreads\\nRoommates.com when she argues it holds that a website\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 13\\n\\ndevelops content if it manipulates the content in a unique\\nway through content-neutral tools.\\n\\n The question in Roommates.com was whether Section\\n230 immunized a website, which matched people renting\\nrooms with people looking for somewhere to live, from\\nclaims that it violated federal and state housing anti-\\ndiscrimination laws by requiring subscribers to disclose,\\nusing dropdown menus and checkboxes, their sex, sexual\\norientation, and family status. See Roommates.com,\\n521 F.3d at 1161–2, 1165.\\n\\n We answered “no” to this question. We rested our\\ndecision, however, on the fact that Roommates.com\\naffirmatively required users to disclose information related\\nto protected classes through discriminatory questions and\\nanswer choices. As a result, this information, especially\\ninformation related to a user’s protected class, served as the\\nfocus of the registration process and, ultimately, became the\\ncornerstone of each user’s online profile. Moreover, the\\nwebsite designed its search function to guide users through\\nthe required discriminatory criteria. Id. at 1164, 1167. Under\\nthese set of facts, the website in Roommates.com was clearly\\nthe developer of the discriminatory content at issue. Id.\\nat 1170.\\n\\n In Roommates.com, we also identified the type of\\nconduct that does not constitute the “development” of\\ncontent under Section 230. Id. at 1169. For example, a\\nhousing website that lets users create their own criteria for\\nidentifying and choosing potential roommates (including\\ncriteria based on protected classes like race or sex) in a blank\\ntext box, does not become a developer of content if it does\\nnot require the use of that discriminatory criteria. Id. In other\\nwords, a website does not become a developer of content\\nwhen it provides neutral tools that a user exploits to create a\\n\\f14 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\nprofile or perform a search using criteria that constitutes a\\nprotected class. Id. We, furthermore, concluded that\\n“[w]here it is very clear that the website directly participates\\nin developing the alleged illegality—as it is clear here with\\nrespect to [Roommates.com’s] questions, answers and the\\nresulting profile pages—immunity will be lost.” However,\\n“in cases of enhancement by… inference—such as with\\nrespect to the ‘Additional Comments’ [on\\nRommates.com]—[S]ection 230 must be interpreted to\\nprotect websites not merely from ultimate liability, but from\\nhaving to fight costly and protracted legal battles.” Id.\\nat 1174–75.\\n\\n Here, Ultimate Software’s functions on Experience\\nProject most resemble the “Additional Comments” features\\nin Roommates.com in that Experience Project users,\\nincluding Wesley Greer, were not required to disclose that\\nthey were looking for heroin or other illegal drugs. Rather,\\nusers were given something along the lines of blank text\\nboxes in which they could post and share experiences,\\nquestions, and answers. The recommendation and\\nnotification functions helped facilitate this user-to-user\\ncommunication, but it did not materially contribute, as\\nPlaintiff argues, to the alleged unlawfulness of the content.\\nRoommates.com, 521 F.3d at 1175; see also Kimzey,\\n836 F.3d at 1269 n.4 (the material contribution test makes a\\n“‘crucial distinction between, on the one hand, taking actions\\n(traditional to publishers) that are necessary to the display of\\nunwelcome and actionable content and, on the other hand,\\nresponsibility for what makes the displayed content illegal\\nor actionable.’”).\\n\\n In summary, Plaintiff is unable to allege that Ultimate\\nSoftware materially contributed to the content posted on\\nExperience Project that led to Greer’s death. Plaintiff cannot\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 15\\n\\nand does not plead that Ultimate Software required users to\\npost specific content, made suggestions regarding the\\ncontent of potential user posts, or contributed to making\\nunlawful or objectionable user posts. Ultimate Software is\\nentitled to immunity under the plain terms of Section 230\\nand our case law as a publisher of third-party content.\\n\\nII. Plaintiff Does Not Plead Sufficient Facts to Show that\\n Ultimate Software Colluded with Drug Dealers on\\n Experience Project\\n\\n The complaint’s allegations as it relates to Plaintiff’s\\n“collusion” with bad actors does not establish an\\nindependent theory of liability. Rather, Plaintiff tries, again,\\nto circumvent Section 230 immunity by alleging that\\nUltimate Software knew or should have known that users\\nsold drugs on Experience Project, and it supported and\\nprotected these drug dealers through its anonymity policies.\\nThe district court characterized this claim well, stating “The\\nidea is that Ultimate Software is less Match.com and more\\nSilk Road (a notorious online platform for criminal\\nactivities, including selling illegal drugs).”\\n\\n To advance this collusion and inducement theory,\\nPlaintiff relies on a Washington Supreme Court decision,\\nJ.S. v. Village Voice Media Holdings, LLC, 184 Wash. 2d 95\\n(2015) (en banc). In Village Voice Media, plaintiffs, minors\\nfeatured in advertisements for sexual services, sued the\\noperators of the website Backpage.com alleging, among\\nother things, violations of state laws prohibiting the sexual\\nexploitation of children. Id. at 98. The court held that\\nplaintiffs sufficiently alleged that the website operators\\nhelped develop the illegal content and therefore were not\\nimmune from liability under Section 230. Id. at 103.\\n\\f16 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n Specifically, the court pointed to allegations that\\nBackpage.com required users to disclose certain information\\nwithin its “escorts” section that encouraged the sexual\\nexploitation of children. Id. at 102. One such allegation is\\nthat Backpage.com’s “content requirements [were]\\nspecifically designed to control the nature and context of\\n[escort] advertisements so that pimps can continue to use\\nBackpage.com to traffic in sex, including the trafficking of\\nchildren.” Id. at 102–03. In other words, the court found that\\nthe plaintiffs alleged enough facts such that it was plausible\\nto infer that Backpage.com’s content requirements—within\\nthe website’s escort section—were designed to facilitate the\\nprostitution of children.\\n\\n Here, Ultimate Software’s anonymity features along\\nwith its public statements expressing concern for internet\\nprivacy and detailing the burden of law enforcement\\ninformation requests are not facts whose inferences, viewed\\nin the light most favorable to Plaintiff, plausibly allege\\ncollusion with drug dealers or other bad actors. Today,\\nonline privacy is a ubiquitous public concern for both users\\nand technology companies. These statements do not\\nestablish, on the part of Ultimate Software, antipathy to law\\nenforcement, especially given the corresponding statements\\nabout always supporting “proper law enforcement requests.”\\n\\n Unlike the plaintiffs in Village Voice Media, Plaintiff\\nhere did not allege that Experience Project had a section for\\ndrug-related experiences on its website with specific content\\nposting requirements that facilitated illegal drug\\ntransactions. Plaintiff’s allegation that user anonymity\\nequals promoting drug transactions is not plausible. Iqbal,\\n556 U.S. at 678. The district court was right to dismiss all\\nclaims related to this supposed theory of liability because\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 17\\n\\nUltimate Software is, as reasoned above, immune under\\nSection 230.\\n\\nIII. Ultimate Software Did Not Owe a Duty to\\n Plaintiff’s Son\\n\\n Ultimate Software owed Greer no duty of care because\\nExperience Project’s features amounted to content-neutral\\nfunctions that did not create a risk of harm. Plaintiff rests her\\n“failure to warn claim” on a misguided premise that\\nmisfeasance by Ultimate Software created a duty to Greer.\\n\\n When analyzing a duty of care in the context of third-\\nparty acts, California courts distinguish between\\n“misfeasance” and “nonfeasance.” Melton v. Boustred,\\n183 Cal. App. 4th 521, 531 (2010). Misfeasance is when a\\ndefendant makes the plaintiff’s position worse while\\nnonfeasance is when a defendant does not help a plaintiff.\\nLugtu v. Cal. Highway Patrol, 26 Cal. 4th 703, 716 (2001).\\nMisfeasance, unlike nonfeasance, creates an ordinary duty\\nof care where none may have existed before. See id.\\n\\n Ultimate Software did not make Plaintiff’s son, Greer,\\nworse off because the functions Plaintiff references—\\nrecommendations and notifications—were used regardless\\nof the groups in which a user participated. No website could\\nfunction if a duty of care was created when a website\\nfacilitates communication, in a content-neutral fashion, of its\\nusers’ content. See e.g., Klayman v. Zuckerberg, 753 F.3d\\n1354, 1359–60 (D.C. Cir. 2014) (no special relationship\\nbetween Facebook and its users). We decline to create such\\na relationship. Accordingly, the district was correct to\\ndismiss Plaintiff’s duty to warn claim.\\n\\f18 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n CONCLUSION\\n\\n For the preceding reasons, we AFFIRM the district\\ncourt’s order granting Defendant Ultimate Software’s\\nmotion to dismiss.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4427022/\", \"author_raw\": \"D.W. NELSON, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"DOROTHY W NELSON"},"judge2":{"kind":"string","value":"JOHNNIE B RAWLINSON"},"judge3":{"kind":"string","value":"CARLOS T BEA"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649769/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Kristanalea Dyroff v. the Ultimate Software Group\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-20\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nKRISTANALEA DYROFF, individually No. 18-15175\nand on behalf of the estate of Wesley\nGreer, deceased, D.C. No.\n Plaintiff-Appellant, 3:17-cv-05359-\n LB\n v.\n\nTHE ULTIMATE SOFTWARE GROUP, OPINION\nINC.,\n Defendant-Appellee.\n\n Appeal from the United States District Court\n for the Northern District of California\n Laurel D. Beeler, Magistrate Judge, Presiding\n\n Argued and Submitted June 4, 2019\n Seattle, Washington\n\n Filed August 20, 2019\n\n Before: Dorothy W. Nelson, Johnnie B. Rawlinson,\n and Carlos T. Bea, Circuit Judges.\n\n Opinion by Judge D.W. Nelson\n\f2 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\n SUMMARY *\n\n\n Communications Decency Act\n\n The panel affirmed the district court’s dismissal, based\non The Ultimate Software Group’s immunity from liability\nunder Section 230 of the Communications Decency Act, of\nan action alleging claims concerning Ultimate Software’s\nrole in the death of plaintiff’s son.\n\n Ultimate Software was the operator of the Experience\nProject website, which allegedly facilitated illegal drug\nsales. Plaintiff’s son died of a drug overdose after buying\ndrugs from a fellow Experience Project user.\n\n The panel held that Ultimate Software satisfied all three\nprongs of the test for immunity under Section 230 of the\nCommunications Decency Act. Specifically, the panel held\nthat Ultimate Software was an interactive computer service\nbecause it did not create or publish its own content under the\nplain language of the statute. The panel also held that\nplaintiff’s son treated Ultimate Software as a publisher or\nspeaker of other’s information or content. Finally, the panel\nheld that Ultimate Software published information/content\nprovided by another information content provider where the\ncontent at issue was created by plaintiff’s son and his drug\ndealer. The panel rejected plaintiff’s argument that a website\ndevelops content if it manipulates the content in a unique\nway through content-neutral tools.\n\n\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 3\n\n The panel held that plaintiff did not plead sufficient facts\nto show that Ultimate Software colluded with drug dealers\non Experience Project. The panel held that plaintiff’s\nallegation that user anonymity equaled promoting drug\ntransactions was not plausible. The panel concluded that the\ndistrict court was right to dismiss all claims related to this\nsupposed theory of liability because Ultimate Software was\nimmune under Section 230 of the Communications Decency\nAct.\n\n The panel held that Ultimate Software did not owe a duty\nof care to plaintiff’s son because Experience Project’s\nfeatures amounted to content-neutral functions that did not\ncreate a risk of harm. The panel rejected plaintiff’s claim\nthat misfeasance by Ultimate Software created a duty to\nplaintiff’s son.\n\n\n COUNSEL\n\nDavid F. Slade (argued), Carney Bates & Pulliam PLLC,\nLittle Rock, Arkansas; Sin-Ting Mary Liu, Aylstock Witkin\nKreis & Overholtz PLLC, Alameda, California; for Plaintiff-\nAppellant.\n\nJeffry A. Miller (argued) and Scott M. Schoenwald, Lewis\nBrisbois Bisgaard & Smith LLP, San Diego, California;\nShawn A. Tolliver, David E. Russo, and Justin S. Kim,\nLewis Brisbois Bisgaard & Smith LLP, San Diego,\nCalifornia; for Defendant-Appellee.\n\f4 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\n OPINION\n\nD.W. NELSON, Circuit Judge:\n\n Plaintiff Kristanalea Dyroff appeals the district court’s\ndismissal of her claims against Defendant The Ultimate\nSoftware Group (“Ultimate Software”), operator of the\nExperience Project website, for its alleged role in the death\nof her son, Wesley Greer. While the circumstances and facts\nof this case are no doubt tragic, we find that Ultimate\nSoftware is immune from liability under Section 230 of the\nCommunications Decency Act. We therefore affirm.\n\n BACKGROUND\n\n This being an appeal from a motion to dismiss, we\ndescribe the case as Plaintiff presents it. We take her\nplausible allegations as true and draw all reasonable\ninferences in her favor.\n\n Experience Project was a social networking website\nmade up of various online communities or groups where\nusers anonymously shared their first-person experiences,\nposted and answered questions, and interacted with other\nusers about different topics. The site did not limit or promote\nthe types of experiences users shared. The site’s “blank box”\napproach to user content resulted in an array of topics and\nforums ranging from “I like dogs” and “I am going to\nStanford” to “I have lung cancer” and “I Love Heroin.”\n\n Users registered with the site anonymously; in other\nwords, the site did not collect users’ identifying information,\nincluding name, phone number, or mailing address. The\nsite’s operator, Ultimate Software, believed that anonymity\nwould promote users to share more personal and authentic\nexperiences without inhibition. Experience Project’s\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 5\n\nfounder stated, “We don’t want to know [users’] real name,\ntheir phone number, what town they’re from.” Id. “The\nimpetus behind this policy [of anonymity] was to encourage\nusers to share experiences with the least amount of inhibition\npossible. The greater the anonymity, the more ‘honest’ the\npost . . . .”\n\n Experience Project was live from 2007 until March\n2016, during which its users shared 67 million experiences,\nmade 15 million connections, and asked 5 million questions.\nUsers could join groups and the site also recommended\ngroups for users to join, based on the content of their posts\nand other attributes, using machine-learning algorithms.\nWhen a user posted content to a group, the site would send\nan email notification to the other users active in that group.\nThe site generated revenue through advertisements and the\nsale of tokens that users used to post questions to other users\nin their groups.\n\n Some of the site’s functions, including user anonymity\nand grouping, facilitated illegal drug sales. Wesley Geer\nwas involved in one such transaction, which turned fatal.\nWesley suffered from drug addiction, which began when a\ndoctor overprescribed him opioid pain killers after a serious\nsports-related injury. After several unsuccessful\nrehabilitation attempts, Wesley bought what he believed to\nbe heroin from a fellow Experience Project user. Wesley\nposted in a heroin-related group, “where can i [sic] score\nheroin in jacksonville, fl.” The site sent him an email\nnotification when another user, Hugo Margenat-Castro or\n“Potheadjuice,” an Orlando-based drug dealer, posted in the\nsame group. Wesley and Margenat-Castro connected off the\nsite and Wesley bought heroin from Margenat-Castro on\nAugust 18, 2015.\n\f6 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\n Wesley died the next day from fentanyl toxicity. He did\nnot know that the heroin Margenat-Castro sold him was\nlaced with fentanyl. Margenat-Castro was ultimately\narrested and prosecuted. He pleaded guilty in March 2017\nadmitting that he sold heroin laced with fentanyl while active\non Experience Project.\n\n In March 2016, Experience Project announced, in an\nopen letter to its users, that it was shutting down. The letter\nexpressed concern for the future of online privacy because\nof government overreach. It stated that the site always\nsupported proper law enforcement efforts but recognized\nthat it did not have the resources to respond to increased\ngovernment information requests. The site shut down on\nApril 21, 2016.\n\n Plaintiff Kristanalea Dyroff, Wesley Greer’s mother,\nfiled a complaint in San Francisco Superior Court. She\nalleges that Ultimate Software: (1) allowed users to traffic\nanonymously in illegal, deadly narcotics and to create\ngroups dedicated to their sale and use; (2) steered users to\nadditional groups dedicated to the sale and use of narcotics;\n(3) sent users alerts to posts within groups that were\ndedicated to the sale and use of narcotics; (4) permitted users\nto remain active accountholders despite evidence that they\nopenly engaged in drug trafficking and that law enforcement\nhad undertaken related investigations; and (5) demonstrated\nantipathy toward law enforcement efforts to stop illegal\nactivity on Experience Project.\n\n Ultimate Software removed the action from state court\nbased on diversity jurisdiction and filed a motion to dismiss\nall claims under Federal Rule of Civil Procedure 12(b)(6).\nThe district court granted the motion without prejudice.\nDyroff filed a notice stating that she would not file an\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 7\n\namended complaint and asked the district court to enter\njudgement. Dyroff timely appealed the judgment.\n\n STANDARD OF REVIEW\n\n We review de novo both a district court order dismissing\na plaintiff’s claims pursuant to Federal Rule of Civil\nProcedure 12(b)(6) and questions of statutory interpretation.\nFields v. Twitter, Inc., 881 F.3d 739, 743 (9th Cir. 2018).\nThe Court must “accept all factual allegations in the\ncomplaint as true and construe the pleadings in the light most\nfavorable to the nonmoving party.” Rowe v. Educ. Credit\nMgmt. Corp., 559 F.3d 1028, 1029–30 (9th Cir. 2009). Only\na complaint that states a plausible claim for relief may\nsurvive a motion to dismiss. Ashcroft v. Iqbal, 556 U.S. 662,\n678 (2009). Plausibility exists when a court may “draw the\nreasonable inference that the defendant is liable for the\nmisconduct alleged.” Id.\n\n DISCUSSION\n\n Plaintiff argues that in granting the motion to dismiss,\nthe district court made three errors. First, she argues, the\ndistrict court erred when it held that Communications\nDecency Act (CDA) Section 230 immunizes Defendant\nUltimate Software. Plaintiff reasons that Ultimate Software,\nas the operator of the Experience Project website, was an\ninformation content provider, as defined by the statute,\nbecause its recommendation and notification functions were\n“specifically designed to make subjective, editorial\ndecisions about users based on their posts.” Second,\naccording to Plaintiff, the district court erred when it found\nthat her allegations of collusion between Ultimate Software\nand drug dealers using Experience Project were not\nplausible. Her third argument is that the district court erred\nin finding that Ultimate Software owed no duty of care to her\n\f8 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\nson, Wesley Greer, an Experience Project user. We affirm\nbecause the district court did not err in any of these respects.\n\nI. CDA Section 230 Immunizes Ultimate Software from\n Plaintiff’s Claims\n\n The CDA provides that website operators are immune\nfrom liability for third-party information (or content, like the\nposts on Experience Project) unless the website operator “is\nresponsible, in whole or in part, for the creation or\ndevelopment of [the] information.” 47 U.S.C. §§ 230(c)(1)\n& (f)(3). Ultimate Software did not create content on\nExperience Project, in whole or in part. Accordingly,\nUltimate Software, as the operator of Experience Project, is\nimmune from liability under the CDA because its functions,\nincluding recommendations and notifications, were content-\nneutral tools used to facilitate communications. See Fair\nHous. Council of San Fernando Valley v. Roommates.com,\n521 F.3d 1157, 1167–69 (9th Cir 2008) (en banc).\n\n A. Scope of CDA Section 230 Immunity\n\n The CDA instructs us that “[n]o provider or user of an\ninteractive computer service shall be treated as the publisher\nor speaker of any information provided by another\ninformation content provider.” 47 U.S.C. § 230(c)(1)\n(emphasis added). The CDA defines an “interactive\ncomputer service” as\n\n [A]ny information service, system, or access\n software provider that provides or enables\n computer access by multiple users to a\n computer server, including specifically a\n service or system that provides access to the\n Internet and such systems operated or\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 9\n\n services offered by libraries or educational\n institutions.\n\n47 U.S.C. § 230(f)(2).\n\n On the other hand, an “information content provider” is\n\n [A]ny person or entity that is responsible, in\n whole or in part, for the creation or\n development of information provided\n through the Internet or any other interactive\n computer service.\n\n47 U.S.C. § 230(f)(3).\n\n “The prototypical service qualifying for [CDA]\nimmunity is an online messaging board (or bulletin board)\non which Internet subscribers post comments and respond to\ncomments posted by others.” Kimzey v. Yelp! Inc., 836 F.3d\n1263, 1266 (9th Cir. 2016) (internal quotations omitted). In\nother words, a website like Experience Project. Taking the\nrelevant statutory definitions and case law in account, it\nbecomes clear that, in general, Section 230(c)(1) “protects\nwebsites from liability [under state or local law] for material\nposted on the[ir] website[s] by someone else.” Doe v.\nInternet Brands, Inc., 824 F.3d 846, 850 (9th Cir. 2016); see\nalso 47 U.S.C. § 230(e)(3).\n\n Combining the above principles, in Barnes v. Yahoo!,\nInc., we created three-prong test for Section 230 immunity.\n570 F.3d 1096, 1100 (9th Cir. 2009). Immunity from liability\nexists for “(1) a provider or user of an interactive computer\nservice (2) whom a plaintiff seeks to treat, under a state law\ncause of action, as a publisher or speaker (3) of information\nprovided by another information content provider.” Id.\nat 1100–01. When a plaintiff cannot allege enough facts to\n\f10 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\novercome Section 230 immunity, a plaintiff’s claims should\nbe dismissed. See Kimzey, 836 F.3d at 1268–71. Ultimate\nSoftware satisfies all three prongs of the test.\n\n B. Section 230 Immunity—The Barnes test\n\n 1. Defendant is an Interactive Computer Service\n\n We interpret the term “interactive computer service”\nexpansively. Kimzey, 836 F.3d at 1268. Ultimate Software\nwas an interactive computer service because it did not create\nor publish its own content under the plain language of the\nstatute. Rather, Ultimate Software published Experience\nProject users’ posts and did not materially contribute to its\nusers’ posts.\n\n Millions of users, including Plaintiff’s son, Wesley\nGreer, set up accounts on Experience Project, a website, to\ncommunicate with each other. Websites are the most\ncommon interactive computer services. Kimzey, 836 F.3d at\n1268; see also Roommates.com, 521 F.3d at 1162 n.6\n(“[t]oday, the most common interactive computer services\nare websites”).\n\n No binding legal authority supports Plaintiff’s\ncontention that Ultimate Software became an information\ncontent provider, losing its Section 230 immunity, by\nfacilitating communication on Experience Project through\ncontent-neutral website functions like group\nrecommendations and post notifications. Ultimate Software,\ntherefore, satisfies the first prong.\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 11\n\n 2. Plaintiff Treats Ultimate Software as a\n Publisher or Speaker of Other’s\n Information/Content\n\n An interactive computer service, like Ultimate Software,\ncan also be an information content provider, but that is only\nrelevant, for the purposes of Section 230 immunity, if the\nwebsite it operates creates or develops the specific content\nat issue. Carafano v. Metrosplash.com, Inc., 339 F.3d 1119,\n1124 (9th Cir. 2003). Here, Ultimate Software was not an\ninformation content provider because it did not create or\ndevelop information (or content). 47 U.S.C. § 230(f)(3).\nRather, it published information created or developed by\nthird parties. Specifically, Experience Project did not create\nor develop the posts that led to Greer’s death. Rather, it was\nGreer, himself, who posted “where can i [sic] score heroin\nin jacksonville, fl” on Experience Project. And it was the\ndrug dealer, Margenat-Castro, who posted in response to\nGreer’s post.\n\n It is true that Ultimate Software used features and\nfunctions, including algorithms, to analyze user posts on\nExperience Project and recommended other user groups.\nThis includes the heroin-related discussion group to which\nGreer posted and (through its emails and push notifications)\nto the drug dealer who sold him the fentanyl-laced heroin.\nPlaintiff, however, cannot plead around Section 230\nimmunity by framing these website features as content. We\nhave held that what matters is whether the claims “inherently\nrequire[] the court to treat the defendant as the ‘publisher or\nspeaker’ of content provided by another.” Barnes, 570 F.3d\nat 1102. If they do, then Section 230(c)(1) provides\nimmunity from liability. Id.\n\n By recommending user groups and sending email\nnotifications, Ultimate Software, through its Experience\n\f12 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\nProject website, was acting as a publisher of others’ content.\nThese functions—recommendations and notifications—are\ntools meant to facilitate the communication and content of\nothers. They are not content in and of themselves.\n\n Our recent decision, HomeAway.com, Inc. v. City of\nSanta Monica, 918 F.3d 676 (9th Cir. 2019) is of no help to\nPlaintiff. There, the City of Santa Monica required short-\nterm vacation rentals to be licensed and imposed liability on\nvacation rental hosting platforms—HomeAway.com and\nAirbnb—that facilitated unlicensed short-term vacation\nrentals. Id. at 680. The platforms sued, alleging, among other\nthings, that Section 230 immunized them from liability. Id.\nWe found that HomeAway.com and Airbnb did not meet the\nsecond prong of the Barnes test because the Santa Monica\nordinance did not “proscribe, mandate, or even discuss the\ncontent of the [website] listings” and required only that the\nwebsite’s transactions involve licensed properties. Id. at 683.\nIn other words, the vacation rental platforms did not face\nliability for the content of their listings; rather liability arose\nfrom facilitating unlicensed booking transactions.\n\n Ultimate Software, therefore, satisfies the second prong\nof the Barnes test.\n\n 3. Ultimate Software Published\n Information/Content Provided by Another\n Information Content Provider\n\n The third prong is also met because, as stated previously\nand as detailed in Plaintiff’s complaint, the content at issue\nwas created and developed by Greer and his drug dealer.\nPlaintiff’s content “manipulation” theory is without support\nin the statute and case law. First, Plaintiff misreads\nRoommates.com when she argues it holds that a website\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 13\n\ndevelops content if it manipulates the content in a unique\nway through content-neutral tools.\n\n The question in Roommates.com was whether Section\n230 immunized a website, which matched people renting\nrooms with people looking for somewhere to live, from\nclaims that it violated federal and state housing anti-\ndiscrimination laws by requiring subscribers to disclose,\nusing dropdown menus and checkboxes, their sex, sexual\norientation, and family status. See Roommates.com,\n521 F.3d at 1161–2, 1165.\n\n We answered “no” to this question. We rested our\ndecision, however, on the fact that Roommates.com\naffirmatively required users to disclose information related\nto protected classes through discriminatory questions and\nanswer choices. As a result, this information, especially\ninformation related to a user’s protected class, served as the\nfocus of the registration process and, ultimately, became the\ncornerstone of each user’s online profile. Moreover, the\nwebsite designed its search function to guide users through\nthe required discriminatory criteria. Id. at 1164, 1167. Under\nthese set of facts, the website in Roommates.com was clearly\nthe developer of the discriminatory content at issue. Id.\nat 1170.\n\n In Roommates.com, we also identified the type of\nconduct that does not constitute the “development” of\ncontent under Section 230. Id. at 1169. For example, a\nhousing website that lets users create their own criteria for\nidentifying and choosing potential roommates (including\ncriteria based on protected classes like race or sex) in a blank\ntext box, does not become a developer of content if it does\nnot require the use of that discriminatory criteria. Id. In other\nwords, a website does not become a developer of content\nwhen it provides neutral tools that a user exploits to create a\n\f14 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\nprofile or perform a search using criteria that constitutes a\nprotected class. Id. We, furthermore, concluded that\n“[w]here it is very clear that the website directly participates\nin developing the alleged illegality—as it is clear here with\nrespect to [Roommates.com’s] questions, answers and the\nresulting profile pages—immunity will be lost.” However,\n“in cases of enhancement by… inference—such as with\nrespect to the ‘Additional Comments’ [on\nRommates.com]—[S]ection 230 must be interpreted to\nprotect websites not merely from ultimate liability, but from\nhaving to fight costly and protracted legal battles.” Id.\nat 1174–75.\n\n Here, Ultimate Software’s functions on Experience\nProject most resemble the “Additional Comments” features\nin Roommates.com in that Experience Project users,\nincluding Wesley Greer, were not required to disclose that\nthey were looking for heroin or other illegal drugs. Rather,\nusers were given something along the lines of blank text\nboxes in which they could post and share experiences,\nquestions, and answers. The recommendation and\nnotification functions helped facilitate this user-to-user\ncommunication, but it did not materially contribute, as\nPlaintiff argues, to the alleged unlawfulness of the content.\nRoommates.com, 521 F.3d at 1175; see also Kimzey,\n836 F.3d at 1269 n.4 (the material contribution test makes a\n“‘crucial distinction between, on the one hand, taking actions\n(traditional to publishers) that are necessary to the display of\nunwelcome and actionable content and, on the other hand,\nresponsibility for what makes the displayed content illegal\nor actionable.’”).\n\n In summary, Plaintiff is unable to allege that Ultimate\nSoftware materially contributed to the content posted on\nExperience Project that led to Greer’s death. Plaintiff cannot\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 15\n\nand does not plead that Ultimate Software required users to\npost specific content, made suggestions regarding the\ncontent of potential user posts, or contributed to making\nunlawful or objectionable user posts. Ultimate Software is\nentitled to immunity under the plain terms of Section 230\nand our case law as a publisher of third-party content.\n\nII. Plaintiff Does Not Plead Sufficient Facts to Show that\n Ultimate Software Colluded with Drug Dealers on\n Experience Project\n\n The complaint’s allegations as it relates to Plaintiff’s\n“collusion” with bad actors does not establish an\nindependent theory of liability. Rather, Plaintiff tries, again,\nto circumvent Section 230 immunity by alleging that\nUltimate Software knew or should have known that users\nsold drugs on Experience Project, and it supported and\nprotected these drug dealers through its anonymity policies.\nThe district court characterized this claim well, stating “The\nidea is that Ultimate Software is less Match.com and more\nSilk Road (a notorious online platform for criminal\nactivities, including selling illegal drugs).”\n\n To advance this collusion and inducement theory,\nPlaintiff relies on a Washington Supreme Court decision,\nJ.S. v. Village Voice Media Holdings, LLC, 184 Wash. 2d 95\n(2015) (en banc). In Village Voice Media, plaintiffs, minors\nfeatured in advertisements for sexual services, sued the\noperators of the website Backpage.com alleging, among\nother things, violations of state laws prohibiting the sexual\nexploitation of children. Id. at 98. The court held that\nplaintiffs sufficiently alleged that the website operators\nhelped develop the illegal content and therefore were not\nimmune from liability under Section 230. Id. at 103.\n\f16 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\n Specifically, the court pointed to allegations that\nBackpage.com required users to disclose certain information\nwithin its “escorts” section that encouraged the sexual\nexploitation of children. Id. at 102. One such allegation is\nthat Backpage.com’s “content requirements [were]\nspecifically designed to control the nature and context of\n[escort] advertisements so that pimps can continue to use\nBackpage.com to traffic in sex, including the trafficking of\nchildren.” Id. at 102–03. In other words, the court found that\nthe plaintiffs alleged enough facts such that it was plausible\nto infer that Backpage.com’s content requirements—within\nthe website’s escort section—were designed to facilitate the\nprostitution of children.\n\n Here, Ultimate Software’s anonymity features along\nwith its public statements expressing concern for internet\nprivacy and detailing the burden of law enforcement\ninformation requests are not facts whose inferences, viewed\nin the light most favorable to Plaintiff, plausibly allege\ncollusion with drug dealers or other bad actors. Today,\nonline privacy is a ubiquitous public concern for both users\nand technology companies. These statements do not\nestablish, on the part of Ultimate Software, antipathy to law\nenforcement, especially given the corresponding statements\nabout always supporting “proper law enforcement requests.”\n\n Unlike the plaintiffs in Village Voice Media, Plaintiff\nhere did not allege that Experience Project had a section for\ndrug-related experiences on its website with specific content\nposting requirements that facilitated illegal drug\ntransactions. Plaintiff’s allegation that user anonymity\nequals promoting drug transactions is not plausible. Iqbal,\n556 U.S. at 678. The district court was right to dismiss all\nclaims related to this supposed theory of liability because\n\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 17\n\nUltimate Software is, as reasoned above, immune under\nSection 230.\n\nIII. Ultimate Software Did Not Owe a Duty to\n Plaintiff’s Son\n\n Ultimate Software owed Greer no duty of care because\nExperience Project’s features amounted to content-neutral\nfunctions that did not create a risk of harm. Plaintiff rests her\n“failure to warn claim” on a misguided premise that\nmisfeasance by Ultimate Software created a duty to Greer.\n\n When analyzing a duty of care in the context of third-\nparty acts, California courts distinguish between\n“misfeasance” and “nonfeasance.” Melton v. Boustred,\n183 Cal. App. 4th 521, 531 (2010). Misfeasance is when a\ndefendant makes the plaintiff’s position worse while\nnonfeasance is when a defendant does not help a plaintiff.\nLugtu v. Cal. Highway Patrol, 26 Cal. 4th 703, 716 (2001).\nMisfeasance, unlike nonfeasance, creates an ordinary duty\nof care where none may have existed before. See id.\n\n Ultimate Software did not make Plaintiff’s son, Greer,\nworse off because the functions Plaintiff references—\nrecommendations and notifications—were used regardless\nof the groups in which a user participated. No website could\nfunction if a duty of care was created when a website\nfacilitates communication, in a content-neutral fashion, of its\nusers’ content. See e.g., Klayman v. Zuckerberg, 753 F.3d\n1354, 1359–60 (D.C. Cir. 2014) (no special relationship\nbetween Facebook and its users). We decline to create such\na relationship. Accordingly, the district was correct to\ndismiss Plaintiff’s duty to warn claim.\n\f18 DYROFF V. THE ULTIMATE SOFTWARE GROUP\n\n CONCLUSION\n\n For the preceding reasons, we AFFIRM the district\ncourt’s order granting Defendant Ultimate Software’s\nmotion to dismiss.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Kristanalea Dyroff v. the Ultimate Software Group\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-20\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nKRISTANALEA DYROFF, individually No. 18-15175\\nand on behalf of the estate of Wesley\\nGreer, deceased, D.C. No.\\n Plaintiff-Appellant, 3:17-cv-05359-\\n LB\\n v.\\n\\nTHE ULTIMATE SOFTWARE GROUP, OPINION\\nINC.,\\n Defendant-Appellee.\\n\\n Appeal from the United States District Court\\n for the Northern District of California\\n Laurel D. Beeler, Magistrate Judge, Presiding\\n\\n Argued and Submitted June 4, 2019\\n Seattle, Washington\\n\\n Filed August 20, 2019\\n\\n Before: Dorothy W. Nelson, Johnnie B. Rawlinson,\\n and Carlos T. Bea, Circuit Judges.\\n\\n Opinion by Judge D.W. Nelson\\n\\f2 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n SUMMARY *\\n\\n\\n Communications Decency Act\\n\\n The panel affirmed the district court’s dismissal, based\\non The Ultimate Software Group’s immunity from liability\\nunder Section 230 of the Communications Decency Act, of\\nan action alleging claims concerning Ultimate Software’s\\nrole in the death of plaintiff’s son.\\n\\n Ultimate Software was the operator of the Experience\\nProject website, which allegedly facilitated illegal drug\\nsales. Plaintiff’s son died of a drug overdose after buying\\ndrugs from a fellow Experience Project user.\\n\\n The panel held that Ultimate Software satisfied all three\\nprongs of the test for immunity under Section 230 of the\\nCommunications Decency Act. Specifically, the panel held\\nthat Ultimate Software was an interactive computer service\\nbecause it did not create or publish its own content under the\\nplain language of the statute. The panel also held that\\nplaintiff’s son treated Ultimate Software as a publisher or\\nspeaker of other’s information or content. Finally, the panel\\nheld that Ultimate Software published information/content\\nprovided by another information content provider where the\\ncontent at issue was created by plaintiff’s son and his drug\\ndealer. The panel rejected plaintiff’s argument that a website\\ndevelops content if it manipulates the content in a unique\\nway through content-neutral tools.\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 3\\n\\n The panel held that plaintiff did not plead sufficient facts\\nto show that Ultimate Software colluded with drug dealers\\non Experience Project. The panel held that plaintiff’s\\nallegation that user anonymity equaled promoting drug\\ntransactions was not plausible. The panel concluded that the\\ndistrict court was right to dismiss all claims related to this\\nsupposed theory of liability because Ultimate Software was\\nimmune under Section 230 of the Communications Decency\\nAct.\\n\\n The panel held that Ultimate Software did not owe a duty\\nof care to plaintiff’s son because Experience Project’s\\nfeatures amounted to content-neutral functions that did not\\ncreate a risk of harm. The panel rejected plaintiff’s claim\\nthat misfeasance by Ultimate Software created a duty to\\nplaintiff’s son.\\n\\n\\n COUNSEL\\n\\nDavid F. Slade (argued), Carney Bates & Pulliam PLLC,\\nLittle Rock, Arkansas; Sin-Ting Mary Liu, Aylstock Witkin\\nKreis & Overholtz PLLC, Alameda, California; for Plaintiff-\\nAppellant.\\n\\nJeffry A. Miller (argued) and Scott M. Schoenwald, Lewis\\nBrisbois Bisgaard & Smith LLP, San Diego, California;\\nShawn A. Tolliver, David E. Russo, and Justin S. Kim,\\nLewis Brisbois Bisgaard & Smith LLP, San Diego,\\nCalifornia; for Defendant-Appellee.\\n\\f4 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n OPINION\\n\\nD.W. NELSON, Circuit Judge:\\n\\n Plaintiff Kristanalea Dyroff appeals the district court’s\\ndismissal of her claims against Defendant The Ultimate\\nSoftware Group (“Ultimate Software”), operator of the\\nExperience Project website, for its alleged role in the death\\nof her son, Wesley Greer. While the circumstances and facts\\nof this case are no doubt tragic, we find that Ultimate\\nSoftware is immune from liability under Section 230 of the\\nCommunications Decency Act. We therefore affirm.\\n\\n BACKGROUND\\n\\n This being an appeal from a motion to dismiss, we\\ndescribe the case as Plaintiff presents it. We take her\\nplausible allegations as true and draw all reasonable\\ninferences in her favor.\\n\\n Experience Project was a social networking website\\nmade up of various online communities or groups where\\nusers anonymously shared their first-person experiences,\\nposted and answered questions, and interacted with other\\nusers about different topics. The site did not limit or promote\\nthe types of experiences users shared. The site’s “blank box”\\napproach to user content resulted in an array of topics and\\nforums ranging from “I like dogs” and “I am going to\\nStanford” to “I have lung cancer” and “I Love Heroin.”\\n\\n Users registered with the site anonymously; in other\\nwords, the site did not collect users’ identifying information,\\nincluding name, phone number, or mailing address. The\\nsite’s operator, Ultimate Software, believed that anonymity\\nwould promote users to share more personal and authentic\\nexperiences without inhibition. Experience Project’s\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 5\\n\\nfounder stated, “We don’t want to know [users’] real name,\\ntheir phone number, what town they’re from.” Id. “The\\nimpetus behind this policy [of anonymity] was to encourage\\nusers to share experiences with the least amount of inhibition\\npossible. The greater the anonymity, the more ‘honest’ the\\npost . . . .”\\n\\n Experience Project was live from 2007 until March\\n2016, during which its users shared 67 million experiences,\\nmade 15 million connections, and asked 5 million questions.\\nUsers could join groups and the site also recommended\\ngroups for users to join, based on the content of their posts\\nand other attributes, using machine-learning algorithms.\\nWhen a user posted content to a group, the site would send\\nan email notification to the other users active in that group.\\nThe site generated revenue through advertisements and the\\nsale of tokens that users used to post questions to other users\\nin their groups.\\n\\n Some of the site’s functions, including user anonymity\\nand grouping, facilitated illegal drug sales. Wesley Geer\\nwas involved in one such transaction, which turned fatal.\\nWesley suffered from drug addiction, which began when a\\ndoctor overprescribed him opioid pain killers after a serious\\nsports-related injury. After several unsuccessful\\nrehabilitation attempts, Wesley bought what he believed to\\nbe heroin from a fellow Experience Project user. Wesley\\nposted in a heroin-related group, “where can i [sic] score\\nheroin in jacksonville, fl.” The site sent him an email\\nnotification when another user, Hugo Margenat-Castro or\\n“Potheadjuice,” an Orlando-based drug dealer, posted in the\\nsame group. Wesley and Margenat-Castro connected off the\\nsite and Wesley bought heroin from Margenat-Castro on\\nAugust 18, 2015.\\n\\f6 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n Wesley died the next day from fentanyl toxicity. He did\\nnot know that the heroin Margenat-Castro sold him was\\nlaced with fentanyl. Margenat-Castro was ultimately\\narrested and prosecuted. He pleaded guilty in March 2017\\nadmitting that he sold heroin laced with fentanyl while active\\non Experience Project.\\n\\n In March 2016, Experience Project announced, in an\\nopen letter to its users, that it was shutting down. The letter\\nexpressed concern for the future of online privacy because\\nof government overreach. It stated that the site always\\nsupported proper law enforcement efforts but recognized\\nthat it did not have the resources to respond to increased\\ngovernment information requests. The site shut down on\\nApril 21, 2016.\\n\\n Plaintiff Kristanalea Dyroff, Wesley Greer’s mother,\\nfiled a complaint in San Francisco Superior Court. She\\nalleges that Ultimate Software: (1) allowed users to traffic\\nanonymously in illegal, deadly narcotics and to create\\ngroups dedicated to their sale and use; (2) steered users to\\nadditional groups dedicated to the sale and use of narcotics;\\n(3) sent users alerts to posts within groups that were\\ndedicated to the sale and use of narcotics; (4) permitted users\\nto remain active accountholders despite evidence that they\\nopenly engaged in drug trafficking and that law enforcement\\nhad undertaken related investigations; and (5) demonstrated\\nantipathy toward law enforcement efforts to stop illegal\\nactivity on Experience Project.\\n\\n Ultimate Software removed the action from state court\\nbased on diversity jurisdiction and filed a motion to dismiss\\nall claims under Federal Rule of Civil Procedure 12(b)(6).\\nThe district court granted the motion without prejudice.\\nDyroff filed a notice stating that she would not file an\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 7\\n\\namended complaint and asked the district court to enter\\njudgement. Dyroff timely appealed the judgment.\\n\\n STANDARD OF REVIEW\\n\\n We review de novo both a district court order dismissing\\na plaintiff’s claims pursuant to Federal Rule of Civil\\nProcedure 12(b)(6) and questions of statutory interpretation.\\nFields v. Twitter, Inc., 881 F.3d 739, 743 (9th Cir. 2018).\\nThe Court must “accept all factual allegations in the\\ncomplaint as true and construe the pleadings in the light most\\nfavorable to the nonmoving party.” Rowe v. Educ. Credit\\nMgmt. Corp., 559 F.3d 1028, 1029–30 (9th Cir. 2009). Only\\na complaint that states a plausible claim for relief may\\nsurvive a motion to dismiss. Ashcroft v. Iqbal, 556 U.S. 662,\\n678 (2009). Plausibility exists when a court may “draw the\\nreasonable inference that the defendant is liable for the\\nmisconduct alleged.” Id.\\n\\n DISCUSSION\\n\\n Plaintiff argues that in granting the motion to dismiss,\\nthe district court made three errors. First, she argues, the\\ndistrict court erred when it held that Communications\\nDecency Act (CDA) Section 230 immunizes Defendant\\nUltimate Software. Plaintiff reasons that Ultimate Software,\\nas the operator of the Experience Project website, was an\\ninformation content provider, as defined by the statute,\\nbecause its recommendation and notification functions were\\n“specifically designed to make subjective, editorial\\ndecisions about users based on their posts.” Second,\\naccording to Plaintiff, the district court erred when it found\\nthat her allegations of collusion between Ultimate Software\\nand drug dealers using Experience Project were not\\nplausible. Her third argument is that the district court erred\\nin finding that Ultimate Software owed no duty of care to her\\n\\f8 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\nson, Wesley Greer, an Experience Project user. We affirm\\nbecause the district court did not err in any of these respects.\\n\\nI. CDA Section 230 Immunizes Ultimate Software from\\n Plaintiff’s Claims\\n\\n The CDA provides that website operators are immune\\nfrom liability for third-party information (or content, like the\\nposts on Experience Project) unless the website operator “is\\nresponsible, in whole or in part, for the creation or\\ndevelopment of [the] information.” 47 U.S.C. §§ 230(c)(1)\\n& (f)(3). Ultimate Software did not create content on\\nExperience Project, in whole or in part. Accordingly,\\nUltimate Software, as the operator of Experience Project, is\\nimmune from liability under the CDA because its functions,\\nincluding recommendations and notifications, were content-\\nneutral tools used to facilitate communications. See Fair\\nHous. Council of San Fernando Valley v. Roommates.com,\\n521 F.3d 1157, 1167–69 (9th Cir 2008) (en banc).\\n\\n A. Scope of CDA Section 230 Immunity\\n\\n The CDA instructs us that “[n]o provider or user of an\\ninteractive computer service shall be treated as the publisher\\nor speaker of any information provided by another\\ninformation content provider.” 47 U.S.C. § 230(c)(1)\\n(emphasis added). The CDA defines an “interactive\\ncomputer service” as\\n\\n [A]ny information service, system, or access\\n software provider that provides or enables\\n computer access by multiple users to a\\n computer server, including specifically a\\n service or system that provides access to the\\n Internet and such systems operated or\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 9\\n\\n services offered by libraries or educational\\n institutions.\\n\\n47 U.S.C. § 230(f)(2).\\n\\n On the other hand, an “information content provider” is\\n\\n [A]ny person or entity that is responsible, in\\n whole or in part, for the creation or\\n development of information provided\\n through the Internet or any other interactive\\n computer service.\\n\\n47 U.S.C. § 230(f)(3).\\n\\n “The prototypical service qualifying for [CDA]\\nimmunity is an online messaging board (or bulletin board)\\non which Internet subscribers post comments and respond to\\ncomments posted by others.” Kimzey v. Yelp! Inc., 836 F.3d\\n1263, 1266 (9th Cir. 2016) (internal quotations omitted). In\\nother words, a website like Experience Project. Taking the\\nrelevant statutory definitions and case law in account, it\\nbecomes clear that, in general, Section 230(c)(1) “protects\\nwebsites from liability [under state or local law] for material\\nposted on the[ir] website[s] by someone else.” Doe v.\\nInternet Brands, Inc., 824 F.3d 846, 850 (9th Cir. 2016); see\\nalso 47 U.S.C. § 230(e)(3).\\n\\n Combining the above principles, in Barnes v. Yahoo!,\\nInc., we created three-prong test for Section 230 immunity.\\n570 F.3d 1096, 1100 (9th Cir. 2009). Immunity from liability\\nexists for “(1) a provider or user of an interactive computer\\nservice (2) whom a plaintiff seeks to treat, under a state law\\ncause of action, as a publisher or speaker (3) of information\\nprovided by another information content provider.” Id.\\nat 1100–01. When a plaintiff cannot allege enough facts to\\n\\f10 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\novercome Section 230 immunity, a plaintiff’s claims should\\nbe dismissed. See Kimzey, 836 F.3d at 1268–71. Ultimate\\nSoftware satisfies all three prongs of the test.\\n\\n B. Section 230 Immunity—The Barnes test\\n\\n 1. Defendant is an Interactive Computer Service\\n\\n We interpret the term “interactive computer service”\\nexpansively. Kimzey, 836 F.3d at 1268. Ultimate Software\\nwas an interactive computer service because it did not create\\nor publish its own content under the plain language of the\\nstatute. Rather, Ultimate Software published Experience\\nProject users’ posts and did not materially contribute to its\\nusers’ posts.\\n\\n Millions of users, including Plaintiff’s son, Wesley\\nGreer, set up accounts on Experience Project, a website, to\\ncommunicate with each other. Websites are the most\\ncommon interactive computer services. Kimzey, 836 F.3d at\\n1268; see also Roommates.com, 521 F.3d at 1162 n.6\\n(“[t]oday, the most common interactive computer services\\nare websites”).\\n\\n No binding legal authority supports Plaintiff’s\\ncontention that Ultimate Software became an information\\ncontent provider, losing its Section 230 immunity, by\\nfacilitating communication on Experience Project through\\ncontent-neutral website functions like group\\nrecommendations and post notifications. Ultimate Software,\\ntherefore, satisfies the first prong.\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 11\\n\\n 2. Plaintiff Treats Ultimate Software as a\\n Publisher or Speaker of Other’s\\n Information/Content\\n\\n An interactive computer service, like Ultimate Software,\\ncan also be an information content provider, but that is only\\nrelevant, for the purposes of Section 230 immunity, if the\\nwebsite it operates creates or develops the specific content\\nat issue. Carafano v. Metrosplash.com, Inc., 339 F.3d 1119,\\n1124 (9th Cir. 2003). Here, Ultimate Software was not an\\ninformation content provider because it did not create or\\ndevelop information (or content). 47 U.S.C. § 230(f)(3).\\nRather, it published information created or developed by\\nthird parties. Specifically, Experience Project did not create\\nor develop the posts that led to Greer’s death. Rather, it was\\nGreer, himself, who posted “where can i [sic] score heroin\\nin jacksonville, fl” on Experience Project. And it was the\\ndrug dealer, Margenat-Castro, who posted in response to\\nGreer’s post.\\n\\n It is true that Ultimate Software used features and\\nfunctions, including algorithms, to analyze user posts on\\nExperience Project and recommended other user groups.\\nThis includes the heroin-related discussion group to which\\nGreer posted and (through its emails and push notifications)\\nto the drug dealer who sold him the fentanyl-laced heroin.\\nPlaintiff, however, cannot plead around Section 230\\nimmunity by framing these website features as content. We\\nhave held that what matters is whether the claims “inherently\\nrequire[] the court to treat the defendant as the ‘publisher or\\nspeaker’ of content provided by another.” Barnes, 570 F.3d\\nat 1102. If they do, then Section 230(c)(1) provides\\nimmunity from liability. Id.\\n\\n By recommending user groups and sending email\\nnotifications, Ultimate Software, through its Experience\\n\\f12 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\nProject website, was acting as a publisher of others’ content.\\nThese functions—recommendations and notifications—are\\ntools meant to facilitate the communication and content of\\nothers. They are not content in and of themselves.\\n\\n Our recent decision, HomeAway.com, Inc. v. City of\\nSanta Monica, 918 F.3d 676 (9th Cir. 2019) is of no help to\\nPlaintiff. There, the City of Santa Monica required short-\\nterm vacation rentals to be licensed and imposed liability on\\nvacation rental hosting platforms—HomeAway.com and\\nAirbnb—that facilitated unlicensed short-term vacation\\nrentals. Id. at 680. The platforms sued, alleging, among other\\nthings, that Section 230 immunized them from liability. Id.\\nWe found that HomeAway.com and Airbnb did not meet the\\nsecond prong of the Barnes test because the Santa Monica\\nordinance did not “proscribe, mandate, or even discuss the\\ncontent of the [website] listings” and required only that the\\nwebsite’s transactions involve licensed properties. Id. at 683.\\nIn other words, the vacation rental platforms did not face\\nliability for the content of their listings; rather liability arose\\nfrom facilitating unlicensed booking transactions.\\n\\n Ultimate Software, therefore, satisfies the second prong\\nof the Barnes test.\\n\\n 3. Ultimate Software Published\\n Information/Content Provided by Another\\n Information Content Provider\\n\\n The third prong is also met because, as stated previously\\nand as detailed in Plaintiff’s complaint, the content at issue\\nwas created and developed by Greer and his drug dealer.\\nPlaintiff’s content “manipulation” theory is without support\\nin the statute and case law. First, Plaintiff misreads\\nRoommates.com when she argues it holds that a website\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 13\\n\\ndevelops content if it manipulates the content in a unique\\nway through content-neutral tools.\\n\\n The question in Roommates.com was whether Section\\n230 immunized a website, which matched people renting\\nrooms with people looking for somewhere to live, from\\nclaims that it violated federal and state housing anti-\\ndiscrimination laws by requiring subscribers to disclose,\\nusing dropdown menus and checkboxes, their sex, sexual\\norientation, and family status. See Roommates.com,\\n521 F.3d at 1161–2, 1165.\\n\\n We answered “no” to this question. We rested our\\ndecision, however, on the fact that Roommates.com\\naffirmatively required users to disclose information related\\nto protected classes through discriminatory questions and\\nanswer choices. As a result, this information, especially\\ninformation related to a user’s protected class, served as the\\nfocus of the registration process and, ultimately, became the\\ncornerstone of each user’s online profile. Moreover, the\\nwebsite designed its search function to guide users through\\nthe required discriminatory criteria. Id. at 1164, 1167. Under\\nthese set of facts, the website in Roommates.com was clearly\\nthe developer of the discriminatory content at issue. Id.\\nat 1170.\\n\\n In Roommates.com, we also identified the type of\\nconduct that does not constitute the “development” of\\ncontent under Section 230. Id. at 1169. For example, a\\nhousing website that lets users create their own criteria for\\nidentifying and choosing potential roommates (including\\ncriteria based on protected classes like race or sex) in a blank\\ntext box, does not become a developer of content if it does\\nnot require the use of that discriminatory criteria. Id. In other\\nwords, a website does not become a developer of content\\nwhen it provides neutral tools that a user exploits to create a\\n\\f14 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\nprofile or perform a search using criteria that constitutes a\\nprotected class. Id. We, furthermore, concluded that\\n“[w]here it is very clear that the website directly participates\\nin developing the alleged illegality—as it is clear here with\\nrespect to [Roommates.com’s] questions, answers and the\\nresulting profile pages—immunity will be lost.” However,\\n“in cases of enhancement by… inference—such as with\\nrespect to the ‘Additional Comments’ [on\\nRommates.com]—[S]ection 230 must be interpreted to\\nprotect websites not merely from ultimate liability, but from\\nhaving to fight costly and protracted legal battles.” Id.\\nat 1174–75.\\n\\n Here, Ultimate Software’s functions on Experience\\nProject most resemble the “Additional Comments” features\\nin Roommates.com in that Experience Project users,\\nincluding Wesley Greer, were not required to disclose that\\nthey were looking for heroin or other illegal drugs. Rather,\\nusers were given something along the lines of blank text\\nboxes in which they could post and share experiences,\\nquestions, and answers. The recommendation and\\nnotification functions helped facilitate this user-to-user\\ncommunication, but it did not materially contribute, as\\nPlaintiff argues, to the alleged unlawfulness of the content.\\nRoommates.com, 521 F.3d at 1175; see also Kimzey,\\n836 F.3d at 1269 n.4 (the material contribution test makes a\\n“‘crucial distinction between, on the one hand, taking actions\\n(traditional to publishers) that are necessary to the display of\\nunwelcome and actionable content and, on the other hand,\\nresponsibility for what makes the displayed content illegal\\nor actionable.’”).\\n\\n In summary, Plaintiff is unable to allege that Ultimate\\nSoftware materially contributed to the content posted on\\nExperience Project that led to Greer’s death. Plaintiff cannot\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 15\\n\\nand does not plead that Ultimate Software required users to\\npost specific content, made suggestions regarding the\\ncontent of potential user posts, or contributed to making\\nunlawful or objectionable user posts. Ultimate Software is\\nentitled to immunity under the plain terms of Section 230\\nand our case law as a publisher of third-party content.\\n\\nII. Plaintiff Does Not Plead Sufficient Facts to Show that\\n Ultimate Software Colluded with Drug Dealers on\\n Experience Project\\n\\n The complaint’s allegations as it relates to Plaintiff’s\\n“collusion” with bad actors does not establish an\\nindependent theory of liability. Rather, Plaintiff tries, again,\\nto circumvent Section 230 immunity by alleging that\\nUltimate Software knew or should have known that users\\nsold drugs on Experience Project, and it supported and\\nprotected these drug dealers through its anonymity policies.\\nThe district court characterized this claim well, stating “The\\nidea is that Ultimate Software is less Match.com and more\\nSilk Road (a notorious online platform for criminal\\nactivities, including selling illegal drugs).”\\n\\n To advance this collusion and inducement theory,\\nPlaintiff relies on a Washington Supreme Court decision,\\nJ.S. v. Village Voice Media Holdings, LLC, 184 Wash. 2d 95\\n(2015) (en banc). In Village Voice Media, plaintiffs, minors\\nfeatured in advertisements for sexual services, sued the\\noperators of the website Backpage.com alleging, among\\nother things, violations of state laws prohibiting the sexual\\nexploitation of children. Id. at 98. The court held that\\nplaintiffs sufficiently alleged that the website operators\\nhelped develop the illegal content and therefore were not\\nimmune from liability under Section 230. Id. at 103.\\n\\f16 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n Specifically, the court pointed to allegations that\\nBackpage.com required users to disclose certain information\\nwithin its “escorts” section that encouraged the sexual\\nexploitation of children. Id. at 102. One such allegation is\\nthat Backpage.com’s “content requirements [were]\\nspecifically designed to control the nature and context of\\n[escort] advertisements so that pimps can continue to use\\nBackpage.com to traffic in sex, including the trafficking of\\nchildren.” Id. at 102–03. In other words, the court found that\\nthe plaintiffs alleged enough facts such that it was plausible\\nto infer that Backpage.com’s content requirements—within\\nthe website’s escort section—were designed to facilitate the\\nprostitution of children.\\n\\n Here, Ultimate Software’s anonymity features along\\nwith its public statements expressing concern for internet\\nprivacy and detailing the burden of law enforcement\\ninformation requests are not facts whose inferences, viewed\\nin the light most favorable to Plaintiff, plausibly allege\\ncollusion with drug dealers or other bad actors. Today,\\nonline privacy is a ubiquitous public concern for both users\\nand technology companies. These statements do not\\nestablish, on the part of Ultimate Software, antipathy to law\\nenforcement, especially given the corresponding statements\\nabout always supporting “proper law enforcement requests.”\\n\\n Unlike the plaintiffs in Village Voice Media, Plaintiff\\nhere did not allege that Experience Project had a section for\\ndrug-related experiences on its website with specific content\\nposting requirements that facilitated illegal drug\\ntransactions. Plaintiff’s allegation that user anonymity\\nequals promoting drug transactions is not plausible. Iqbal,\\n556 U.S. at 678. The district court was right to dismiss all\\nclaims related to this supposed theory of liability because\\n\\f DYROFF V. THE ULTIMATE SOFTWARE GROUP 17\\n\\nUltimate Software is, as reasoned above, immune under\\nSection 230.\\n\\nIII. Ultimate Software Did Not Owe a Duty to\\n Plaintiff’s Son\\n\\n Ultimate Software owed Greer no duty of care because\\nExperience Project’s features amounted to content-neutral\\nfunctions that did not create a risk of harm. Plaintiff rests her\\n“failure to warn claim” on a misguided premise that\\nmisfeasance by Ultimate Software created a duty to Greer.\\n\\n When analyzing a duty of care in the context of third-\\nparty acts, California courts distinguish between\\n“misfeasance” and “nonfeasance.” Melton v. Boustred,\\n183 Cal. App. 4th 521, 531 (2010). Misfeasance is when a\\ndefendant makes the plaintiff’s position worse while\\nnonfeasance is when a defendant does not help a plaintiff.\\nLugtu v. Cal. Highway Patrol, 26 Cal. 4th 703, 716 (2001).\\nMisfeasance, unlike nonfeasance, creates an ordinary duty\\nof care where none may have existed before. See id.\\n\\n Ultimate Software did not make Plaintiff’s son, Greer,\\nworse off because the functions Plaintiff references—\\nrecommendations and notifications—were used regardless\\nof the groups in which a user participated. No website could\\nfunction if a duty of care was created when a website\\nfacilitates communication, in a content-neutral fashion, of its\\nusers’ content. See e.g., Klayman v. Zuckerberg, 753 F.3d\\n1354, 1359–60 (D.C. Cir. 2014) (no special relationship\\nbetween Facebook and its users). We decline to create such\\na relationship. Accordingly, the district was correct to\\ndismiss Plaintiff’s duty to warn claim.\\n\\f18 DYROFF V. THE ULTIMATE SOFTWARE GROUP\\n\\n CONCLUSION\\n\\n For the preceding reasons, we AFFIRM the district\\ncourt’s order granting Defendant Ultimate Software’s\\nmotion to dismiss.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4085,"cells":{"id":{"kind":"number","value":4649770,"string":"4,649,770"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Jose Valtierra v. Medtronic Inc."},"decision_date":{"kind":"string","value":"2019-08-20"},"docket_number":{"kind":"string","value":"17-15282"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Mary M. Schroeder and Milan D. Smith, Jr., Circuit Judges, and Jed S. Rakoff,* District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"SCHROEDER, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n JOSE VALTIERRA, No. 17-15282\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:15-cv-00865-\\n SMM\\n MEDTRONIC INC., a Minnesota\\n Company,\\n Defendant-Appellee. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Stephen M. McNamee, District Judge, Presiding\\n\\n Argued and Submitted June 14, 2019\\n San Francisco, California\\n\\n Filed August 20, 2019\\n\\n Before: Mary M. Schroeder and Milan D. Smith, Jr.,\\n Circuit Judges, and Jed S. Rakoff,* District Judge.\\n\\n Opinion by Judge Schroeder\\n\\n\\n\\n\\n *\\n The Honorable Jed S. Rakoff, United States District Judge for the\\nSouthern District of New York, sitting by designation.\\n\\f2 VALTIERRA V. MEDTRONIC\\n\\n SUMMARY**\\n\\n\\n Employment Discrimination\\n\\n The panel affirmed the district court’s grant of summary\\njudgment in favor of the defendant in an employment\\ndiscrimination action under the Americans with Disabilities\\nAct.\\n\\n Plaintiff claimed he was terminated from his employment\\non account of his morbid obesity, which the district court held\\nwas not a physical impairment and could not constitute a\\ndisability unless it was caused by an underlying physiological\\ncondition. Therefore, plaintiff could not establish disability\\ndiscrimination.\\n\\n The panel affirmed on other grounds, holding that, even\\nif plaintiff’s obesity were an impairment under the ADA, or\\nhe suffered from a disabling knee condition that the district\\ncourt could have considered, he could not show a causal\\nrelationship between these impairments and his termination.\\n\\n\\n COUNSEL\\n\\nJessica Miller (argued) and Michael Zoldan, Zoldan Law\\nGroup PLLC, Scottsdale, Arizona, for Plaintiff-Appellant.\\n\\nRobert Shawn Oller (argued), Littler Mendelson P.C.,\\nPhoenix, Arizona, for Defendant-Appellee.\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f VALTIERRA V. MEDTRONIC 3\\n\\nBarbara L. Sloan (argued), Attorney; Sydney A.R. Foster,\\nAssistant General Counsel; Jennifer S. Goldstein, Associate\\nGeneral Counsel; James L. Lee, Deputy General Counsel;\\nOffice of the General Counsel, United States Equal\\nEmployment Opportunity Commision, Washington, D.C.; for\\nAmicus Curiae Equal Employment Opportunity Commission.\\n\\nBryan P. Neal (argued) and Stephen F. Fink, Thompson &\\nKnight LLP, Dallas, Texas, for Amicus Curiae BNSF\\nRailway Company.\\n\\n\\n OPINION\\n\\nSCHROEDER, Circuit Judge:\\n\\n Plaintiff Jose Valtierra appeals the district court’s\\njudgment in favor of his former employer, Medtronic Inc., in\\nhis action under the Americans with Disabilities Act\\n(“ADA”). Valtierra claimed that he was terminated on\\naccount of his morbid obesity, which the district court held\\nwas not a physical impairment under the relevant Equal\\nEmployment Opportunity Commission (“EEOC”) regulations\\nand interpretative guidance. He contends on appeal that the\\ndistrict court misinterpreted that guidance. Medtronic, in\\naddition to defending the district court’s interpretation of the\\nEEOC guidance, defends the grant of summary judgment on\\nthe additional ground that the reason for plaintiff’s\\ntermination was not his obesity, but his falsification of\\nMedtronic’s business records to reflect that he had completed\\nrequired tasks.\\n\\n The underlying facts are not disputed. Valtierra began\\nworking for Medtronic in about 2004 as a facility\\n\\f4 VALTIERRA V. MEDTRONIC\\n\\nmaintenance technician. The company makes specialized\\nmedical devices used in the treatment and diagnosis of\\nvarious medical conditions such as diabetes and\\ncardiovascular disease. Valtierra’s job involved the repair\\nand maintenance of the company’s manufacturing equipment.\\nThe company informed him of his job assignments through\\na computer program that kept track of the work needed to\\nkeep the equipment in good condition.\\n\\n Valtierra was seriously overweight from the time of his\\nhiring until his termination. When he was hired he weighed\\nover 300 pounds, and at the time of the events relevant to this\\ncase in 2014, his weight had increased to more than\\n370 pounds. He requested and was given time off in late\\n2013 on account of joint pain associated with his weight and\\nreturned to work in December 2013 without medical\\nrestrictions. There is no dispute, however, that his weight\\nremained so considerable that he suffered from a condition\\ncommonly known as morbid obesity.\\n\\n In May 2014, his supervisor noticed that he seemed to be\\nhaving difficulty walking and was using the elevator instead\\nof the stairs. Later, concerned about Valtierra’s ability to do\\nthe job, the supervisor checked the computer system to see if\\nValtierra’s assignments had been completed. Although\\nValtierra had left for vacation the day before, the computer\\nindicated he had already completed twelve assignments that\\nshould have taken a more significant amount of time to\\ncomplete.\\n\\n When confronted with these discrepancies, Valtierra\\nadmitted that he had not performed all the work, but\\nexplained that he had intended to complete the work when he\\nreturned from vacation. He now further contends he should\\n\\f VALTIERRA V. MEDTRONIC 5\\n\\nnot have been assigned so many tasks to complete in so little\\ntime, because the employer knew he required\\naccommodations for his weight. Medtronic terminated him\\nfor falsifying the records.\\n\\n Valtierra then filed this action in the District Court for the\\nDistrict of Arizona. The gravamen of his complaint, as\\nrelevant to the issues before us, was that he suffered from a\\ndisability within the meaning of the ADA and his termination\\nconstituted unlawful discrimination in violation of that Act.\\nThe district court granted summary judgment for Medtronic,\\nholding, in accord with the decisions of several federal\\ncircuits, that obesity, no matter how great, cannot constitute\\na disability under the applicable EEOC regulations unless the\\nobesity is caused by an underlying physiological condition.\\nSince Valtierra could not identify any such underlying\\ncondition—perhaps, as the EEOC suggests, like most persons\\nwho suffer from obesity—he could not claim discrimination\\non account of a disability. The district court therefore\\ndismissed the action.\\n\\n At the time of the district court’s decision, three circuits\\nhad reached a similar conclusion. See Morriss v. BNSF Ry.\\nCo., 817 F.3d 1104 (8th Cir. 2016); EEOC v. Watkins Motor\\nLines, 463 F.3d 436 (6th Cir. 2006); Francis v. City of\\nMeriden, 129 F.3d 281 (2d Cir. 1997). Since then, the\\nSeventh Circuit has joined. Richardson v. Chicago Transit\\nAuth., 926 F.3d 881 (7th Cir. 2019). By contrast, the EEOC,\\nas amicus, supports the plaintiff’s argument that all of these\\ncourts have incorrectly analyzed the EEOC’s regulations and\\nguidance.\\n\\n We therefore turn to the relevant statutory and regulatory\\nprovisions, and the earlier federal appellate decisions that\\n\\f6 VALTIERRA V. MEDTRONIC\\n\\nconcluded morbid obesity alone cannot be the basis for a\\nclaim of disability discrimination. The ADA defines\\n“disability” as a “physical or mental impairment that\\nsubstantially limits one or more major life activities.”\\n42 U.S.C. § 12102(1)(A). Congress went on to provide some\\nexamples. The “major life activities” that Congress has listed\\ninclude, among others, “walking, standing, lifting . . . and\\nworking.” Id. § 12102(2)(A). The parties appear to agree\\nthat if Valtierra has a physical “impairment” within the\\nmeaning of subsection (1)(A), it sufficiently limits his\\nactivities so as to render him disabled. The question then\\nbecomes whether he has such an “impairment.”\\n\\n The EEOC has defined physical impairment in the\\nrelevant regulation, promulgated in 1992, as follows:\\n\\n Any physiological disorder or condition,\\n cosmetic disfigurement, or anatomical loss\\n affecting one or more body systems, such as\\n neurological, musculoskeletal, special sense\\n organs, respiratory (including speech organs),\\n cardiovascular, reproductive, digestive,\\n genitourinary, immune, circulatory, hemic,\\n lymphatic, skin, and endocrine[.]\\n\\n29 C.F.R. § 1630.2(h)(1).\\n\\n The EEOC produced an appendix to the regulations,\\ntermed “interpretive guidance,” encompassing more than\\n600 pages. The relevant passage discussing Section\\n1630.2(h) states:\\n\\n It is important to distinguish between\\n conditions that are impairments and physical,\\n\\f VALTIERRA V. MEDTRONIC 7\\n\\n psychological, environmental, cultural and\\n economic characteristics that are not\\n impairments. The definition of the term\\n “impairment” does not include physical\\n characteristics such as eye color, hair color,\\n lefthandedness or height, weight or muscle\\n tone that are within “normal” range and are\\n not the result of a physiological disorder. The\\n definition, likewise, does not include\\n characteristic predisposition to illness or\\n disease. Other conditions, such as pregnancy,\\n that are not the result of a physiological\\n disorder are also not impairments. However,\\n a pregnancy-related impairment that\\n substantially limits a major life activity is a\\n disability under the first prong of the\\n definition.\\n\\n29 C.F.R. pt. 1630, app., § 1630.2(h) (emphasis added).\\n\\n The EEOC argues morbid obesity is plainly physiological\\nin its effects, and that numerous federal agencies have\\ncategorized it as a disease. The agency further contends that\\nValtierra has raised at least a genuine issue of material fact\\nregarding whether his morbid obesity is an impairment,\\nbecause his medical records show that several of his bodily\\nsystems have been adversely affected.\\n\\n In this case, however, we need not take a definitive stand\\non the question of whether morbid obesity itself is an\\n“impairment” under the ADA. That is because, even\\nassuming that it is such an impairment, or that Valtierra\\nsuffered from a disabling knee condition that the district court\\ncould have considered, he would have to show some causal\\n\\f8 VALTIERRA V. MEDTRONIC\\n\\nrelationship between these impairments and his termination.\\nSee Hutton v. Elf Atochem N. Am., Inc., 273 F.3d 884, 891\\n(9th Cir. 2001) (a plaintiff must show he suffered adverse\\nemployment action “because of” his disability to succeed on\\nan ADA discrimination claim). He is unable to do so.\\nValtierra admits he closed twelve maintenance assignments\\nas having been completed when he had not done the work. In\\naddition, he had worked for Medtronic for more than ten\\nyears and had always weighed in excess of 300 pounds.\\nThere is no basis for concluding that he was terminated for\\nany reason other than Medtronic’s stated ground that he\\nfalsified records to show he had completed work assignments.\\n\\n Valtierra nevertheless maintains there is a triable issue\\nconcerning the reason for his termination. He suggests that\\nsimilarly situated employees were treated differently and the\\nfalsified records may have been a pretext for discrimination\\non the basis of his weight. The district court record does not\\nbear this out. Valtierra stated in his deposition that other\\nemployees were late in completing assignments, but he did\\nnot know whether any other employees closed out work\\nassignments without completing them. While the record does\\nsuggest that two other employees admitted to such\\nmisconduct, management was not informed of their actions\\nand never discovered that others had prematurely closed\\ntasks. Thus, Medtronic could not have treated Valtierra\\ndifferently because there is no evidence Medtronic ever knew\\nof similar misconduct on the part of others.\\n\\n The judgment of the district court is AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4427023/\", \"author_raw\": \"SCHROEDER, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MARY M SCHROEDER"},"judge2":{"kind":"string","value":"MILAN D SMITH JR"},"judge3":{"kind":"string","value":"JED S RAKOFF"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649770/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Jose Valtierra v. Medtronic Inc.\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-20\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n JOSE VALTIERRA, No. 17-15282\n Plaintiff-Appellant,\n D.C. No.\n v. 2:15-cv-00865-\n SMM\n MEDTRONIC INC., a Minnesota\n Company,\n Defendant-Appellee. OPINION\n\n\n Appeal from the United States District Court\n for the District of Arizona\n Stephen M. McNamee, District Judge, Presiding\n\n Argued and Submitted June 14, 2019\n San Francisco, California\n\n Filed August 20, 2019\n\n Before: Mary M. Schroeder and Milan D. Smith, Jr.,\n Circuit Judges, and Jed S. Rakoff,* District Judge.\n\n Opinion by Judge Schroeder\n\n\n\n\n *\n The Honorable Jed S. Rakoff, United States District Judge for the\nSouthern District of New York, sitting by designation.\n\f2 VALTIERRA V. MEDTRONIC\n\n SUMMARY**\n\n\n Employment Discrimination\n\n The panel affirmed the district court’s grant of summary\njudgment in favor of the defendant in an employment\ndiscrimination action under the Americans with Disabilities\nAct.\n\n Plaintiff claimed he was terminated from his employment\non account of his morbid obesity, which the district court held\nwas not a physical impairment and could not constitute a\ndisability unless it was caused by an underlying physiological\ncondition. Therefore, plaintiff could not establish disability\ndiscrimination.\n\n The panel affirmed on other grounds, holding that, even\nif plaintiff’s obesity were an impairment under the ADA, or\nhe suffered from a disabling knee condition that the district\ncourt could have considered, he could not show a causal\nrelationship between these impairments and his termination.\n\n\n COUNSEL\n\nJessica Miller (argued) and Michael Zoldan, Zoldan Law\nGroup PLLC, Scottsdale, Arizona, for Plaintiff-Appellant.\n\nRobert Shawn Oller (argued), Littler Mendelson P.C.,\nPhoenix, Arizona, for Defendant-Appellee.\n\n **\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f VALTIERRA V. MEDTRONIC 3\n\nBarbara L. Sloan (argued), Attorney; Sydney A.R. Foster,\nAssistant General Counsel; Jennifer S. Goldstein, Associate\nGeneral Counsel; James L. Lee, Deputy General Counsel;\nOffice of the General Counsel, United States Equal\nEmployment Opportunity Commision, Washington, D.C.; for\nAmicus Curiae Equal Employment Opportunity Commission.\n\nBryan P. Neal (argued) and Stephen F. Fink, Thompson &\nKnight LLP, Dallas, Texas, for Amicus Curiae BNSF\nRailway Company.\n\n\n OPINION\n\nSCHROEDER, Circuit Judge:\n\n Plaintiff Jose Valtierra appeals the district court’s\njudgment in favor of his former employer, Medtronic Inc., in\nhis action under the Americans with Disabilities Act\n(“ADA”). Valtierra claimed that he was terminated on\naccount of his morbid obesity, which the district court held\nwas not a physical impairment under the relevant Equal\nEmployment Opportunity Commission (“EEOC”) regulations\nand interpretative guidance. He contends on appeal that the\ndistrict court misinterpreted that guidance. Medtronic, in\naddition to defending the district court’s interpretation of the\nEEOC guidance, defends the grant of summary judgment on\nthe additional ground that the reason for plaintiff’s\ntermination was not his obesity, but his falsification of\nMedtronic’s business records to reflect that he had completed\nrequired tasks.\n\n The underlying facts are not disputed. Valtierra began\nworking for Medtronic in about 2004 as a facility\n\f4 VALTIERRA V. MEDTRONIC\n\nmaintenance technician. The company makes specialized\nmedical devices used in the treatment and diagnosis of\nvarious medical conditions such as diabetes and\ncardiovascular disease. Valtierra’s job involved the repair\nand maintenance of the company’s manufacturing equipment.\nThe company informed him of his job assignments through\na computer program that kept track of the work needed to\nkeep the equipment in good condition.\n\n Valtierra was seriously overweight from the time of his\nhiring until his termination. When he was hired he weighed\nover 300 pounds, and at the time of the events relevant to this\ncase in 2014, his weight had increased to more than\n370 pounds. He requested and was given time off in late\n2013 on account of joint pain associated with his weight and\nreturned to work in December 2013 without medical\nrestrictions. There is no dispute, however, that his weight\nremained so considerable that he suffered from a condition\ncommonly known as morbid obesity.\n\n In May 2014, his supervisor noticed that he seemed to be\nhaving difficulty walking and was using the elevator instead\nof the stairs. Later, concerned about Valtierra’s ability to do\nthe job, the supervisor checked the computer system to see if\nValtierra’s assignments had been completed. Although\nValtierra had left for vacation the day before, the computer\nindicated he had already completed twelve assignments that\nshould have taken a more significant amount of time to\ncomplete.\n\n When confronted with these discrepancies, Valtierra\nadmitted that he had not performed all the work, but\nexplained that he had intended to complete the work when he\nreturned from vacation. He now further contends he should\n\f VALTIERRA V. MEDTRONIC 5\n\nnot have been assigned so many tasks to complete in so little\ntime, because the employer knew he required\naccommodations for his weight. Medtronic terminated him\nfor falsifying the records.\n\n Valtierra then filed this action in the District Court for the\nDistrict of Arizona. The gravamen of his complaint, as\nrelevant to the issues before us, was that he suffered from a\ndisability within the meaning of the ADA and his termination\nconstituted unlawful discrimination in violation of that Act.\nThe district court granted summary judgment for Medtronic,\nholding, in accord with the decisions of several federal\ncircuits, that obesity, no matter how great, cannot constitute\na disability under the applicable EEOC regulations unless the\nobesity is caused by an underlying physiological condition.\nSince Valtierra could not identify any such underlying\ncondition—perhaps, as the EEOC suggests, like most persons\nwho suffer from obesity—he could not claim discrimination\non account of a disability. The district court therefore\ndismissed the action.\n\n At the time of the district court’s decision, three circuits\nhad reached a similar conclusion. See Morriss v. BNSF Ry.\nCo., 817 F.3d 1104 (8th Cir. 2016); EEOC v. Watkins Motor\nLines, 463 F.3d 436 (6th Cir. 2006); Francis v. City of\nMeriden, 129 F.3d 281 (2d Cir. 1997). Since then, the\nSeventh Circuit has joined. Richardson v. Chicago Transit\nAuth., 926 F.3d 881 (7th Cir. 2019). By contrast, the EEOC,\nas amicus, supports the plaintiff’s argument that all of these\ncourts have incorrectly analyzed the EEOC’s regulations and\nguidance.\n\n We therefore turn to the relevant statutory and regulatory\nprovisions, and the earlier federal appellate decisions that\n\f6 VALTIERRA V. MEDTRONIC\n\nconcluded morbid obesity alone cannot be the basis for a\nclaim of disability discrimination. The ADA defines\n“disability” as a “physical or mental impairment that\nsubstantially limits one or more major life activities.”\n42 U.S.C. § 12102(1)(A). Congress went on to provide some\nexamples. The “major life activities” that Congress has listed\ninclude, among others, “walking, standing, lifting . . . and\nworking.” Id. § 12102(2)(A). The parties appear to agree\nthat if Valtierra has a physical “impairment” within the\nmeaning of subsection (1)(A), it sufficiently limits his\nactivities so as to render him disabled. The question then\nbecomes whether he has such an “impairment.”\n\n The EEOC has defined physical impairment in the\nrelevant regulation, promulgated in 1992, as follows:\n\n Any physiological disorder or condition,\n cosmetic disfigurement, or anatomical loss\n affecting one or more body systems, such as\n neurological, musculoskeletal, special sense\n organs, respiratory (including speech organs),\n cardiovascular, reproductive, digestive,\n genitourinary, immune, circulatory, hemic,\n lymphatic, skin, and endocrine[.]\n\n29 C.F.R. § 1630.2(h)(1).\n\n The EEOC produced an appendix to the regulations,\ntermed “interpretive guidance,” encompassing more than\n600 pages. The relevant passage discussing Section\n1630.2(h) states:\n\n It is important to distinguish between\n conditions that are impairments and physical,\n\f VALTIERRA V. MEDTRONIC 7\n\n psychological, environmental, cultural and\n economic characteristics that are not\n impairments. The definition of the term\n “impairment” does not include physical\n characteristics such as eye color, hair color,\n lefthandedness or height, weight or muscle\n tone that are within “normal” range and are\n not the result of a physiological disorder. The\n definition, likewise, does not include\n characteristic predisposition to illness or\n disease. Other conditions, such as pregnancy,\n that are not the result of a physiological\n disorder are also not impairments. However,\n a pregnancy-related impairment that\n substantially limits a major life activity is a\n disability under the first prong of the\n definition.\n\n29 C.F.R. pt. 1630, app., § 1630.2(h) (emphasis added).\n\n The EEOC argues morbid obesity is plainly physiological\nin its effects, and that numerous federal agencies have\ncategorized it as a disease. The agency further contends that\nValtierra has raised at least a genuine issue of material fact\nregarding whether his morbid obesity is an impairment,\nbecause his medical records show that several of his bodily\nsystems have been adversely affected.\n\n In this case, however, we need not take a definitive stand\non the question of whether morbid obesity itself is an\n“impairment” under the ADA. That is because, even\nassuming that it is such an impairment, or that Valtierra\nsuffered from a disabling knee condition that the district court\ncould have considered, he would have to show some causal\n\f8 VALTIERRA V. MEDTRONIC\n\nrelationship between these impairments and his termination.\nSee Hutton v. Elf Atochem N. Am., Inc., 273 F.3d 884, 891\n(9th Cir. 2001) (a plaintiff must show he suffered adverse\nemployment action “because of” his disability to succeed on\nan ADA discrimination claim). He is unable to do so.\nValtierra admits he closed twelve maintenance assignments\nas having been completed when he had not done the work. In\naddition, he had worked for Medtronic for more than ten\nyears and had always weighed in excess of 300 pounds.\nThere is no basis for concluding that he was terminated for\nany reason other than Medtronic’s stated ground that he\nfalsified records to show he had completed work assignments.\n\n Valtierra nevertheless maintains there is a triable issue\nconcerning the reason for his termination. He suggests that\nsimilarly situated employees were treated differently and the\nfalsified records may have been a pretext for discrimination\non the basis of his weight. The district court record does not\nbear this out. Valtierra stated in his deposition that other\nemployees were late in completing assignments, but he did\nnot know whether any other employees closed out work\nassignments without completing them. While the record does\nsuggest that two other employees admitted to such\nmisconduct, management was not informed of their actions\nand never discovered that others had prematurely closed\ntasks. Thus, Medtronic could not have treated Valtierra\ndifferently because there is no evidence Medtronic ever knew\nof similar misconduct on the part of others.\n\n The judgment of the district court is AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Jose Valtierra v. Medtronic Inc.\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-20\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n JOSE VALTIERRA, No. 17-15282\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:15-cv-00865-\\n SMM\\n MEDTRONIC INC., a Minnesota\\n Company,\\n Defendant-Appellee. OPINION\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Stephen M. McNamee, District Judge, Presiding\\n\\n Argued and Submitted June 14, 2019\\n San Francisco, California\\n\\n Filed August 20, 2019\\n\\n Before: Mary M. Schroeder and Milan D. Smith, Jr.,\\n Circuit Judges, and Jed S. Rakoff,* District Judge.\\n\\n Opinion by Judge Schroeder\\n\\n\\n\\n\\n *\\n The Honorable Jed S. Rakoff, United States District Judge for the\\nSouthern District of New York, sitting by designation.\\n\\f2 VALTIERRA V. MEDTRONIC\\n\\n SUMMARY**\\n\\n\\n Employment Discrimination\\n\\n The panel affirmed the district court’s grant of summary\\njudgment in favor of the defendant in an employment\\ndiscrimination action under the Americans with Disabilities\\nAct.\\n\\n Plaintiff claimed he was terminated from his employment\\non account of his morbid obesity, which the district court held\\nwas not a physical impairment and could not constitute a\\ndisability unless it was caused by an underlying physiological\\ncondition. Therefore, plaintiff could not establish disability\\ndiscrimination.\\n\\n The panel affirmed on other grounds, holding that, even\\nif plaintiff’s obesity were an impairment under the ADA, or\\nhe suffered from a disabling knee condition that the district\\ncourt could have considered, he could not show a causal\\nrelationship between these impairments and his termination.\\n\\n\\n COUNSEL\\n\\nJessica Miller (argued) and Michael Zoldan, Zoldan Law\\nGroup PLLC, Scottsdale, Arizona, for Plaintiff-Appellant.\\n\\nRobert Shawn Oller (argued), Littler Mendelson P.C.,\\nPhoenix, Arizona, for Defendant-Appellee.\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f VALTIERRA V. MEDTRONIC 3\\n\\nBarbara L. Sloan (argued), Attorney; Sydney A.R. Foster,\\nAssistant General Counsel; Jennifer S. Goldstein, Associate\\nGeneral Counsel; James L. Lee, Deputy General Counsel;\\nOffice of the General Counsel, United States Equal\\nEmployment Opportunity Commision, Washington, D.C.; for\\nAmicus Curiae Equal Employment Opportunity Commission.\\n\\nBryan P. Neal (argued) and Stephen F. Fink, Thompson &\\nKnight LLP, Dallas, Texas, for Amicus Curiae BNSF\\nRailway Company.\\n\\n\\n OPINION\\n\\nSCHROEDER, Circuit Judge:\\n\\n Plaintiff Jose Valtierra appeals the district court’s\\njudgment in favor of his former employer, Medtronic Inc., in\\nhis action under the Americans with Disabilities Act\\n(“ADA”). Valtierra claimed that he was terminated on\\naccount of his morbid obesity, which the district court held\\nwas not a physical impairment under the relevant Equal\\nEmployment Opportunity Commission (“EEOC”) regulations\\nand interpretative guidance. He contends on appeal that the\\ndistrict court misinterpreted that guidance. Medtronic, in\\naddition to defending the district court’s interpretation of the\\nEEOC guidance, defends the grant of summary judgment on\\nthe additional ground that the reason for plaintiff’s\\ntermination was not his obesity, but his falsification of\\nMedtronic’s business records to reflect that he had completed\\nrequired tasks.\\n\\n The underlying facts are not disputed. Valtierra began\\nworking for Medtronic in about 2004 as a facility\\n\\f4 VALTIERRA V. MEDTRONIC\\n\\nmaintenance technician. The company makes specialized\\nmedical devices used in the treatment and diagnosis of\\nvarious medical conditions such as diabetes and\\ncardiovascular disease. Valtierra’s job involved the repair\\nand maintenance of the company’s manufacturing equipment.\\nThe company informed him of his job assignments through\\na computer program that kept track of the work needed to\\nkeep the equipment in good condition.\\n\\n Valtierra was seriously overweight from the time of his\\nhiring until his termination. When he was hired he weighed\\nover 300 pounds, and at the time of the events relevant to this\\ncase in 2014, his weight had increased to more than\\n370 pounds. He requested and was given time off in late\\n2013 on account of joint pain associated with his weight and\\nreturned to work in December 2013 without medical\\nrestrictions. There is no dispute, however, that his weight\\nremained so considerable that he suffered from a condition\\ncommonly known as morbid obesity.\\n\\n In May 2014, his supervisor noticed that he seemed to be\\nhaving difficulty walking and was using the elevator instead\\nof the stairs. Later, concerned about Valtierra’s ability to do\\nthe job, the supervisor checked the computer system to see if\\nValtierra’s assignments had been completed. Although\\nValtierra had left for vacation the day before, the computer\\nindicated he had already completed twelve assignments that\\nshould have taken a more significant amount of time to\\ncomplete.\\n\\n When confronted with these discrepancies, Valtierra\\nadmitted that he had not performed all the work, but\\nexplained that he had intended to complete the work when he\\nreturned from vacation. He now further contends he should\\n\\f VALTIERRA V. MEDTRONIC 5\\n\\nnot have been assigned so many tasks to complete in so little\\ntime, because the employer knew he required\\naccommodations for his weight. Medtronic terminated him\\nfor falsifying the records.\\n\\n Valtierra then filed this action in the District Court for the\\nDistrict of Arizona. The gravamen of his complaint, as\\nrelevant to the issues before us, was that he suffered from a\\ndisability within the meaning of the ADA and his termination\\nconstituted unlawful discrimination in violation of that Act.\\nThe district court granted summary judgment for Medtronic,\\nholding, in accord with the decisions of several federal\\ncircuits, that obesity, no matter how great, cannot constitute\\na disability under the applicable EEOC regulations unless the\\nobesity is caused by an underlying physiological condition.\\nSince Valtierra could not identify any such underlying\\ncondition—perhaps, as the EEOC suggests, like most persons\\nwho suffer from obesity—he could not claim discrimination\\non account of a disability. The district court therefore\\ndismissed the action.\\n\\n At the time of the district court’s decision, three circuits\\nhad reached a similar conclusion. See Morriss v. BNSF Ry.\\nCo., 817 F.3d 1104 (8th Cir. 2016); EEOC v. Watkins Motor\\nLines, 463 F.3d 436 (6th Cir. 2006); Francis v. City of\\nMeriden, 129 F.3d 281 (2d Cir. 1997). Since then, the\\nSeventh Circuit has joined. Richardson v. Chicago Transit\\nAuth., 926 F.3d 881 (7th Cir. 2019). By contrast, the EEOC,\\nas amicus, supports the plaintiff’s argument that all of these\\ncourts have incorrectly analyzed the EEOC’s regulations and\\nguidance.\\n\\n We therefore turn to the relevant statutory and regulatory\\nprovisions, and the earlier federal appellate decisions that\\n\\f6 VALTIERRA V. MEDTRONIC\\n\\nconcluded morbid obesity alone cannot be the basis for a\\nclaim of disability discrimination. The ADA defines\\n“disability” as a “physical or mental impairment that\\nsubstantially limits one or more major life activities.”\\n42 U.S.C. § 12102(1)(A). Congress went on to provide some\\nexamples. The “major life activities” that Congress has listed\\ninclude, among others, “walking, standing, lifting . . . and\\nworking.” Id. § 12102(2)(A). The parties appear to agree\\nthat if Valtierra has a physical “impairment” within the\\nmeaning of subsection (1)(A), it sufficiently limits his\\nactivities so as to render him disabled. The question then\\nbecomes whether he has such an “impairment.”\\n\\n The EEOC has defined physical impairment in the\\nrelevant regulation, promulgated in 1992, as follows:\\n\\n Any physiological disorder or condition,\\n cosmetic disfigurement, or anatomical loss\\n affecting one or more body systems, such as\\n neurological, musculoskeletal, special sense\\n organs, respiratory (including speech organs),\\n cardiovascular, reproductive, digestive,\\n genitourinary, immune, circulatory, hemic,\\n lymphatic, skin, and endocrine[.]\\n\\n29 C.F.R. § 1630.2(h)(1).\\n\\n The EEOC produced an appendix to the regulations,\\ntermed “interpretive guidance,” encompassing more than\\n600 pages. The relevant passage discussing Section\\n1630.2(h) states:\\n\\n It is important to distinguish between\\n conditions that are impairments and physical,\\n\\f VALTIERRA V. MEDTRONIC 7\\n\\n psychological, environmental, cultural and\\n economic characteristics that are not\\n impairments. The definition of the term\\n “impairment” does not include physical\\n characteristics such as eye color, hair color,\\n lefthandedness or height, weight or muscle\\n tone that are within “normal” range and are\\n not the result of a physiological disorder. The\\n definition, likewise, does not include\\n characteristic predisposition to illness or\\n disease. Other conditions, such as pregnancy,\\n that are not the result of a physiological\\n disorder are also not impairments. However,\\n a pregnancy-related impairment that\\n substantially limits a major life activity is a\\n disability under the first prong of the\\n definition.\\n\\n29 C.F.R. pt. 1630, app., § 1630.2(h) (emphasis added).\\n\\n The EEOC argues morbid obesity is plainly physiological\\nin its effects, and that numerous federal agencies have\\ncategorized it as a disease. The agency further contends that\\nValtierra has raised at least a genuine issue of material fact\\nregarding whether his morbid obesity is an impairment,\\nbecause his medical records show that several of his bodily\\nsystems have been adversely affected.\\n\\n In this case, however, we need not take a definitive stand\\non the question of whether morbid obesity itself is an\\n“impairment” under the ADA. That is because, even\\nassuming that it is such an impairment, or that Valtierra\\nsuffered from a disabling knee condition that the district court\\ncould have considered, he would have to show some causal\\n\\f8 VALTIERRA V. MEDTRONIC\\n\\nrelationship between these impairments and his termination.\\nSee Hutton v. Elf Atochem N. Am., Inc., 273 F.3d 884, 891\\n(9th Cir. 2001) (a plaintiff must show he suffered adverse\\nemployment action “because of” his disability to succeed on\\nan ADA discrimination claim). He is unable to do so.\\nValtierra admits he closed twelve maintenance assignments\\nas having been completed when he had not done the work. In\\naddition, he had worked for Medtronic for more than ten\\nyears and had always weighed in excess of 300 pounds.\\nThere is no basis for concluding that he was terminated for\\nany reason other than Medtronic’s stated ground that he\\nfalsified records to show he had completed work assignments.\\n\\n Valtierra nevertheless maintains there is a triable issue\\nconcerning the reason for his termination. He suggests that\\nsimilarly situated employees were treated differently and the\\nfalsified records may have been a pretext for discrimination\\non the basis of his weight. The district court record does not\\nbear this out. Valtierra stated in his deposition that other\\nemployees were late in completing assignments, but he did\\nnot know whether any other employees closed out work\\nassignments without completing them. While the record does\\nsuggest that two other employees admitted to such\\nmisconduct, management was not informed of their actions\\nand never discovered that others had prematurely closed\\ntasks. Thus, Medtronic could not have treated Valtierra\\ndifferently because there is no evidence Medtronic ever knew\\nof similar misconduct on the part of others.\\n\\n The judgment of the district court is AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4086,"cells":{"id":{"kind":"number","value":4649771,"string":"4,649,771"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Baldemar Zuniga v. William Barr"},"decision_date":{"kind":"string","value":"2019-08-20"},"docket_number":{"kind":"string","value":"16-72982"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Andrew J. Kleinfeld and Michelle T. Friedland, Circuit Judges, and William H. Pauley III, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PER CURIAM\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n BALDEMAR ZUNIGA, No. 16-72982\\n Petitioner,\\n Agency No.\\n v. A089-247-110\\n\\n WILLIAM P. BARR, Attorney General,\\n Respondent. OPINION\\n\\n On Petition for Review of an\\n Immigration Judge’s Decision\\n\\n Argued and Submitted May 17, 2019\\n Seattle, Washington\\n\\n Filed August 20, 2019\\n\\n Before: Andrew J. Kleinfeld and Michelle T. Friedland,\\nCircuit Judges, and William H. Pauley III, * District Judge.\\n\\n Per Curiam Opinion\\n\\n\\n\\n\\n *\\n The Honorable William H. Pauley III, United States District Judge\\nfor the Southern District of New York, sitting by designation.\\n\\f2 ZUNIGA V. BARR\\n\\n SUMMARY **\\n\\n\\n Immigration\\n\\n The panel granted Baldemar Zuniga’s petition for review\\nof an immigration judge’s decision affirming an asylum\\nofficer’s negative reasonable fear determination in expedited\\nremoval proceedings, and remanded, holding that non-\\ncitizens subject to expedited removal under 8 U.S.C. § 1228\\nhave a statutory right to counsel in reasonable fear\\nproceedings before immigration judges, and that the\\nimmigration judge deprived Zuniga of his right to counsel\\nby failing to obtain a knowing and voluntary waiver of that\\nright.\\n\\n The panel rejected the government’s argument that there\\nis no statutory right to counsel in reasonable fear\\nproceedings. The panel explained that 8 U.S.C. § 1228,\\nwhich governs expedited removal proceedings for non-\\ncitizens convicted of committing aggravated felonies, and\\nthrough which non-citizens can request reasonable fear\\ninterviews, explicitly provides that non-citizens have the\\nprivilege of being represented, at no expense to the\\ngovernment, by counsel. The panel further explained that\\nother subsections of § 1228 reinforce that right by requiring\\nthat proceedings for the removal of criminal non-citizens be\\nconducted in conformity with § 1229a, which in turn\\nprovides a statutory right to counsel in ordinary removal\\nproceedings, and directing the government to take\\nreasonable efforts not to impair an individual’s access and\\nright to counsel when considering whether to detain non-\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f ZUNIGA V. BARR 3\\n\\ncitizens. The panel also stated that the broader legislative\\ncontext outside of the specific provisions dealing with\\nexpedited removal proceedings for criminal non-citizens\\nalso supports the conclusion that there is a right to counsel\\nin reasonable fear proceedings.\\n\\n The panel rejected the government’s contention, which\\nrelied on a 1999 memo from the Executive Office for\\nImmigration Review interpreting 8 C.F.R. § 208.31, that IJs\\nhave discretion to decide whether a non-citizen may be\\nrepresented by counsel. The panel noted that the\\ngovernment was correct that the regulations specify only that\\nnon-citizens may be represented by counsel in the initial\\nreasonable fear interview before an asylum officer, and that\\nthey are silent as to representation by counsel in the review\\nhearing before the IJ, but the panel concluded that deference\\nto EOIR’s interpretation of that silence was not warranted\\nbecause its interpretation conflicted with the plain text of\\n8 U.S.C. § 1228.\\n\\n The panel held that the IJ violated Zuniga’s Fifth\\nAmendment right to due process by failing to obtain a\\nknowing and voluntary waiver of his right to counsel. The\\npanel further held that Zuniga need not show prejudice\\nwhere he was denied his statutory right.\\n\\n The panel remanded for a new hearing in which Zuniga’s\\nright to counsel is honored.\\n\\f4 ZUNIGA V. BARR\\n\\n COUNSEL\\n\\nRobert Pauw (argued), Gibbs Houston Pauw, Seattle,\\nWashington, for Petitioner.\\n\\nNehal Kamani (argued) and Kathryn M. McKinney,\\nAttorneys; Stephen J. Flynn, Assistant Director; Office of\\nImmigration Litigation, Civil Division, United States\\nDepartment of Justice, Washington, D.C.; for Respondent.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n This case presents us with a simple question: do non-\\ncitizens subject to expedited removal under 8 U.S.C. § 1228\\nhave a statutory right to counsel in reasonable fear\\nproceedings before immigration judges? The answer, based\\non the plain language of § 1228, is yes.\\n\\n Petitioner Baldemar Zuniga contends that in his hearing\\nbefore an Immigration Judge (“IJ”) to review a negative\\nreasonable fear determination made by an asylum officer,\\nthe IJ denied him his right to counsel. Because we conclude\\nthat Zuniga had a statutory right to counsel, that the colloquy\\nat the beginning of the hearing before the IJ was inadequate\\nto waive that right, and that no showing of prejudice is\\nrequired, we reverse and remand for further proceedings in\\nwhich Zuniga is given the opportunity to proceed with\\ncounsel.\\n\\n I.\\n\\n Baldemar Zuniga is a Mexican national who illegally\\nentered the United States as a child. In 2012, he was\\n\\f ZUNIGA V. BARR 5\\n\\nconvicted of participating in a conspiracy to manufacture\\nand distribute drugs and launder money. Zuniga testified in\\nopen court against two of his co-conspirators, who were part\\nof Mexico’s notorious Knights Templar cartel.\\n\\n While he was in prison, immigration authorities served\\nZuniga with a Notice of Intent to Issue a Final\\nAdministrative Removal Order. Because he had been\\nconvicted of a drug trafficking aggravated felony, Zuniga\\nwas placed in expedited removal proceedings pursuant to\\n8 U.S.C. § 1228(b). During an interview with Immigration\\nand Customs Enforcement (“ICE”), he expressed a fear that,\\nif removed to Mexico, he would face torture or persecution\\nby the cartel as retribution for his testimony. He was then\\nreferred to an asylum officer for an interview to evaluate\\nwhether his fear was “reasonable.”\\n\\n At the beginning of the reasonable fear interview, the\\nasylum officer explained that Zuniga had “the right to have\\n[his] attorney present for the interview.” Zuniga stated that\\nhe had an attorney helping with his reasonable fear case but\\nthat he was willing to continue with the reasonable fear\\ninterview without his attorney. 1 The asylum officer then\\nproceeded with the interview questions and concluded that\\nZuniga had not established a reasonable fear of persecution\\nor torture.\\n\\n Zuniga requested review of the asylum officer’s negative\\nreasonable fear determination by an Immigration Judge.\\n\\n 1\\n The asylum officer’s notes from the interview spell the name of\\nthe attorney Zuniga mentioned as “Robert Pawl.” His current attorney,\\nRobert Pauw, confirmed at oral argument before our court that he was\\nalso representing Zuniga at the time of Zuniga’s reasonable fear\\ninterview, and that he was the same attorney Zuniga mentioned to the\\nasylum officer.\\n\\f6 ZUNIGA V. BARR\\n\\nZuniga’s case was referred to the Immigration Court in San\\nFrancisco, California. The notice of referral to the IJ\\nincluded the following language: “You may be represented\\nin this proceeding, at no expense to the government, by an\\nattorney or other individual authorized and qualified to\\nrepresent persons before an Immigration Court. If you wish\\nto be so represented, your attorney or representative should\\nappear with you at this hearing.”\\n\\n Zuniga appeared at the hearing by video conference from\\na detention center in Mesa Verde, California. 2 At the outset\\nof the hearing, when announcing the case number and who\\nwas participating, the IJ stated that Zuniga did “not have a\\nlawyer.” Shortly thereafter, the IJ engaged in the following\\ncolloquy with Zuniga:\\n\\n JUDGE: In these proceedings you have the\\n right to counsel of your own choosing, but the\\n government will not pay for your attorney.\\n You should have received a copy of the free\\n legal service list. Did you get that list?\\n\\n ZUNIGA: Yes.\\n\\n JUDGE: So, sir, do you have a lawyer?\\n\\n ZUNIGA: I do not.\\n\\n\\n\\n\\n 2\\n In a later declaration, Zuniga testified that the video setup “made\\nit extremely difficult to understand everything that was happening” and\\nthat he had been under the impression that the hearing “was going to be\\na quick hearing where [he] was going to . . . ask for another extension.”\\n\\f ZUNIGA V. BARR 7\\n\\n JUDGE: All right, sir, were you interviewed\\n by an asylum officer regarding your fear of\\n returning to Mexico?\\n\\nThe IJ then proceeded with the substance of the hearing. The\\nIJ ultimately agreed with the asylum officer that Zuniga\\nlacked “a reasonable fear of persecution on account of a\\nground protected by the law[s] of the United States.”\\n\\n Zuniga petitioned our court for review of that decision,\\narguing that his due process rights were violated by the use\\nof the video conference in his reasonable fear review hearing\\nbefore the IJ and by the IJ’s failure to obtain a proper waiver\\nof his right to an attorney in that proceeding. He also argued\\nthat the IJ erred in determining that he lacked a reasonable\\nfear of persecution on a protected ground and in evaluating\\nhis fear of torture.\\n\\n II.\\n\\n We review de novo due process challenges to reasonable\\nfear proceedings. Colmenar v. I.N.S., 210 F.3d 967, 971 (9th\\nCir. 2000).\\n\\n III.\\n\\n A.\\n\\n In 1999, the Department of Justice created the reasonable\\nfear interview to serve as a “screening process to evaluate\\ntorture claims for aliens subject to streamlined\\nadministrative removal processes for aggravated felons . . .\\nand for aliens subject to reinstatement of a previous removal\\norder.” Regulations Concerning the Convention Against\\nTorture, 64 Fed. Reg. 8478, 8485 (Feb. 19, 1999). These\\ntwo groups of non-citizens are ineligible for asylum but may\\n\\f8 ZUNIGA V. BARR\\n\\nbe entitled to withholding or deferral of removal under\\n§ 241(b)(3) of the Immigration and Nationality Act (“INA”),\\ncurrently codified at 8 U.S.C. § 1231(b)(3), or under the\\nConvention Against Torture. 3 Regulations Concerning the\\nConvention Against Torture, 64 Fed. Reg. at 8485.\\n\\n As a first step of the reasonable fear process, non-citizens\\nwho express a fear of returning to their country of removal\\nto ICE are interviewed in a “non-adversarial manner” by an\\nasylum officer to determine whether they have a reasonable\\nfear of persecution or torture. 8 C.F.R. §§ 208.31(a), (c),\\n1208.31(a), (c). 4 If the asylum officer determines that a non-\\n\\n 3\\n The reasonable fear screening process was modeled on the credible\\nfear screening process, which had previously been created to provide a\\nlimited avenue for relief for inadmissible arriving aliens who are subject\\nto expedited removal under 8 U.S.C. § 1225. Such non-citizens can be\\nremoved without further review unless they express either a credible fear\\nof persecution or torture in the country of removal or an intent to seek\\nasylum. 8 U.S.C. § 1225(b)(1)(A)–(B); 8 C.F.R. §§ 208.30, 1208.30.\\nCredible fear requires establishing a “significant possibility” that the\\nnon-citizen could be eligible for asylum if given an opportunity in a more\\nfulsome proceeding. 8 U.S.C. § 1225(b)(1)(B)(v). The reasonable fear\\nscreening process was designed to fulfill a similar function for non-\\ncitizens being deported under 8 U.S.C. § 1228, but it requires them to\\nmeet a more demanding standard—they must demonstrate a “reasonable\\npossibility” of persecution or torture in the country of removal.\\nRegulations Concerning the Convention Against Torture, 64 Fed. Reg.\\nat 8485 (“Because the standard for showing entitlement to the[] forms of\\nprotection [available to these non-citizen aggravated felons] (a\\nprobability of persecution or torture) is significantly higher than the\\nstandard for asylum (a well-founded fear of persecution), the screening\\nstandard adopted for initial consideration of withholding and deferral\\nrequests in these contexts is also higher.”); see also 8 C.F.R. §§ 208.31,\\n1208.31.\\n 4\\n There are two identical sets of regulations contained in Chapters I\\nand V of 8 C.F.R. that are applicable to the Department of Homeland\\nSecurity and the Executive Office for Immigration Review, respectively.\\n\\f ZUNIGA V. BARR 9\\n\\ncitizen has a reasonable fear, he will be referred to an IJ for\\na merits hearing to determine eligibility for withholding of\\nremoval. Id. § 208.31(e). However, if the asylum officer\\ndetermines that the non-citizen “does not have a reasonable\\nfear of persecution or torture,” he “will be afforded the\\nopportunity for an expeditious review of the negative\\nscreening determination by an immigration judge.”\\nRegulations Concerning the Convention Against Torture,\\n64 Fed. Reg. at 8485 (emphasis added); see also 8 C.F.R.\\n§ 208.31(f)–(g). If a non-citizen exercises his right to\\nchallenge a negative reasonable fear determination, the IJ\\nreviews the asylum officer’s determination de novo.\\n8 C.F.R. § 208.31(f)–(g); Bartolome v. Sessions, 904 F.3d\\n803, 809 (9th Cir. 2018).\\n\\n If the IJ upholds the negative screening determination, it\\nconstitutes a final order of removal. 8 C.F.R. § 208.31(g)(1).\\nBecause the IJ’s decision is the final agency action on the\\nreasonable fear question, such a final removal order is ripe\\nfor judicial review by a federal court of appeals. Ayala v.\\nSessions, 855 F.3d 1012, 1015–16 (9th Cir. 2017).\\n\\n B.\\n\\n Zuniga argues that he had a right to counsel in his\\nreasonable fear review hearing before the IJ, that he did not\\nwaive that right because his colloquy with the IJ was\\ninsufficient to demonstrate that he made a knowing and\\nvoluntary waiver, and that this error requires automatic\\n\\n\\nAliens and Nationality; Homeland Security; Reorganization of\\nRegulations, 68 Fed. Reg. 9824, 9825 (Feb. 28, 2003). Because the text\\nis the same in both sets of regulations, for the sake of simplicity we will\\nrefer only to one set throughout the remainder of this opinion.\\n\\f10 ZUNIGA V. BARR\\n\\nreversal. The Government does not contest that, if Zuniga\\nhad a right to counsel, there was no adequate waiver here. 5\\nRather, the Government rests its defense to this petition\\nprimarily on the argument that there is no statutory right to\\ncounsel in reasonable fear proceedings. This argument\\nclearly fails.\\n\\n Expedited removal proceedings for non-citizens\\nconvicted of committing aggravated felonies (through which\\nnon-citizens can request reasonable fear interviews) are a\\ncreation of INA § 238(b), currently codified at 8 U.S.C.\\n§ 1228. Section 1228 explicitly provides that non-citizens\\n“shall have the privilege of being represented (at no expense\\nto the government) by [] counsel.” 8 U.S.C.\\n§ 1228(b)(4)(B). Other subsections of § 1228 reinforce the\\nright to counsel. The statute provides, for example, that the\\n“proceedings [for removal of criminal non-citizens] shall be\\nconducted in conformity with section 1229a,” 8 U.S.C.\\n§ 1228(a)(1), which provides a statutory right to counsel in\\nordinary removal proceedings, 8 U.S.C. § 1229a(b)(4)(A). 6\\n\\n 5\\n As explained below, we have previously held that if a non-citizen\\nappears without counsel in a removal hearing, “there must be a knowing\\nand voluntary waiver of the right to counsel” before the hearing can\\nproceed. Tawadrus v. Ashcroft, 364 F.3d 1099, 1103 (9th Cir. 2004). A\\nvalid waiver requires that the IJ “(1) inquire specifically as to whether\\npetitioner wishes to continue without a lawyer; and (2) receive a knowing\\nand voluntary affirmative response.” Id. (citations omitted).\\n 6\\n There is some overlap between 8 U.S.C. § 1228(a), which\\ndescribes the general procedures for removing criminal non-citizens, and\\n§ 1228(b), the provision under which Zuniga was removed, which\\napplies specifically to the removal of criminal non-citizens who are not\\npermanent residents. Both emphasize the need to conform with the\\nprocedures described in 8 U.S.C. § 1229a, but § 1228(b) provides some\\nspecific procedures applicable only to non-citizens who are not\\npermanent residents.\\n\\f ZUNIGA V. BARR 11\\n\\nSection 1228 also states that in deciding where to detain non-\\ncitizens under this section, “the Attorney General shall make\\nreasonable efforts to ensure that the alien’s access to counsel\\nand right to counsel under [8 U.S.C.] section 1362,” which\\nlikewise provides for the right to counsel in removal\\nproceedings, “are not impaired.” 8 U.S.C. § 1228(a)(2).\\n\\n This language makes clear that the statute contemplates\\na right to counsel in removal proceedings initiated under\\n8 U.S.C. § 1228, which includes reasonable fear review\\nproceedings. Nothing in the language of § 1228 indicates\\nthat the right to counsel is conditional or limited only to\\ncertain types of proceedings authorized under that statute,\\nexpedited or otherwise. Cf. United States v. Reyes-Bonilla,\\n671 F.3d 1036, 1045 (9th Cir. 2012) (noting, with citation to\\n8 U.S.C. § 1228, that the “right to counsel in expedited\\nremoval proceedings is . . . secured by statute”).\\n\\n The broader legislative context—outside of the specific\\nprovisions dealing with expedited removal proceedings for\\ncriminal non-citizens—also supports the conclusion that\\nthere is a right to counsel in reasonable fear proceedings.\\nThe INA gives non-citizens the right to be represented by an\\nattorney in most immigration proceedings as long as the\\ngovernment does not have to bear the expense. In particular,\\n8 U.S.C. § 1362 provides that “[i]n any removal proceedings\\nbefore an immigration judge and in any appeal proceedings\\nbefore the Attorney General from any such removal\\nproceedings,” non-citizens “shall have the privilege of being\\nrepresented (at no expense to the Government)” by counsel\\nof their choosing. See also 8 U.S.C. § 1229a(b)(4)(A)\\n(“[T]he alien shall have the privilege of being represented,\\nat no expense to the Government, by counsel of the alien’s\\nchoosing who is authorized to practice in such\\nproceedings.”). Because the text of § 1228 is both “clear and\\n\\f12 ZUNIGA V. BARR\\n\\nconsistent with the statutory scheme at issue, the plain\\nlanguage of the statute is conclusive” in providing a statutory\\nright to counsel. Emmert Indus. Corp. v. Artisan Assocs.,\\nInc., 497 F.3d 982, 987 (9th Cir. 2007) (quoting Molski v.\\nM.J. Cable, Inc., 481 F.3d 724, 732 (9th Cir. 2007)).\\n\\n The Government resists this conclusion, arguing that the\\nIJ has discretion to decide whether a non-citizen may be\\nrepresented by counsel. The Government’s argument is\\npremised on a 1999 memo from the Executive Office for\\nImmigration Review (“EOIR”). That memo interpreted\\n8 C.F.R. § 208.31 as giving an IJ the right to decide whether\\na non-citizen may be represented in the negative reasonable\\nfear determination review hearing because the regulation did\\nnot explicitly mention the right to counsel in those\\nproceedings. See Exec. Office for Immigration Review,\\nOffice of the Chief Immigration Judge, Operating Policies\\nand Procedures Memorandum No. 99-5: Implementation of\\nArticle 3 of the UN Convention Against Torture 8 (1999)\\n(“With regard to representation in reasonable fear review\\nproceedings, the . . . regulations are . . . silent. . . . Since\\nthere is no specific regulatory guidance on this point, the\\nissue is left to the discretion of the Immigration Judge.”).\\n\\n The Government is correct that the regulations specify\\nonly that non-citizens may be represented by counsel in the\\ninitial reasonable fear interview before an asylum officer,\\nand that they are silent as to representation by counsel in the\\nreview hearing before the IJ. Compare 8 C.F.R. § 208.31(c),\\nwith 8 C.F.R. § 208.31(g). But its argument that we should\\ndefer to EOIR’s interpretation of that silence fails because\\nthat interpretation conflicts with the plain text of 8 U.S.C.\\n§ 1228. We “need not accept an agency’s interpretation of\\nits own regulations if that interpretation is inconsistent with\\nthe statute under which the regulations were promulgated.”\\n\\f ZUNIGA V. BARR 13\\n\\nTurtle Island Restoration Network v. U.S. Dep’t of\\nCommerce, 878 F.3d 725, 733 (9th Cir. 2017) (quotation\\nmarks omitted); see also Auer v. Robbins, 519 U.S. 452, 461\\n(1997) (holding that courts should generally defer to an\\nagency’s interpretation of its own regulation unless, among\\nother circumstances, that interpretation is “plainly\\nerroneous” (quoting Robertson v. Methow Valley Citizens\\nCouncil, 490 U.S. 332, 359 (1989)). 7 We therefore conclude\\nthat Zuniga had a right to representation by an attorney in his\\nreasonable fear review hearing before the IJ.\\n\\n We have previously held that “[a]lthough there is no\\nSixth Amendment right to counsel in an immigration\\nhearing, Congress has recognized [a right to counsel] among\\nthe rights stemming from the Fifth Amendment guarantee of\\ndue process” by codifying it. Tawadrus v. Ashcroft,\\n364 F.3d 1099, 1103 (9th Cir. 2004); see also Rios-Berrios\\nv. I.N.S., 776 F.2d 859, 862 (9th Cir. 1985) (“[D]ue process\\nmandates that [a non-citizen] is entitled to counsel of his own\\nchoice at his own expense under terms of the [INA].”). And,\\nfurther, before a petitioner continues without counsel in an\\nimmigration proceeding for which there is a statutory right\\nto counsel, due process mandates that “there must be a\\nknowing and voluntary waiver of the right to counsel,”\\n\\n 7\\n Nor is the agency’s interpretation persuasive enough to otherwise\\ncommand respect. See Skidmore v. Swift & Co., 323 U.S. 134, 140\\n(1944) (explaining that “[t]he weight of” an agency interpretation “will\\ndepend upon the thoroughness evident in its consideration, the validity\\nof its reasoning, its consistency with earlier and later pronouncements,\\nand all those factors which give it power to persuade, if lacking power to\\ncontrol”). In addition to being contrary to the statute, the EOIR memo’s\\ninterpretation could lead to absurd results. For example, a non-citizen\\nwho had been informed by his hearing notice that he could bring an\\nattorney to represent him might arrive at the hearing only to learn that\\ncounsel was barred from the hearing at the discretion of the IJ.\\n\\f14 ZUNIGA V. BARR\\n\\nwhich requires that the IJ “(1) inquire specifically as to\\nwhether petitioner wishes to continue without a lawyer; and\\n(2) receive a knowing and voluntary affirmative response.”\\nTawadrus, 364 F.3d at 1103 (citations omitted). As the\\nGovernment concedes, Zuniga’s colloquy with the IJ was\\ninadequate to effect a valid waiver of the right to counsel\\nunder Tawadrus.\\n\\n The Government contends that a remand is nevertheless\\nunavailable to Zuniga because he cannot show prejudice\\nfrom the lack of counsel. This contention is foreclosed by\\nMontes-Lopez v. Holder, 694 F.3d 1085 (9th Cir. 2012), in\\nwhich we held that “an alien who shows that he has been\\ndenied the statutory right to be represented by counsel in an\\nimmigration proceeding need not also show that he was\\nprejudiced by the absence of the attorney.” Id. at 1093–94. 8\\n\\n IV.\\n\\n For the reasons described above, we conclude that the IJ\\nviolated Zuniga’s right to counsel in his reasonable fear\\nreview proceeding by failing to obtain a valid waiver, and\\nthat Zuniga is entitled to a new hearing before an IJ in which\\n\\n 8\\n Gomez-Velazco v. Sessions, 879 F.3d 989 (9th Cir. 2018), is not to\\nthe contrary. There, we held that we could not presume prejudice where\\na non-citizen, Gomez-Velazco, was denied counsel during one “discrete\\nstage” of expedited removal proceedings—his initial interaction with\\nofficers from the Department of Homeland Security. Id. at 994–95. In\\nGomez-Velazco, we distinguished Montes-Lopez on the ground that,\\nafter the discrete phase in which he lacked counsel, Gomez-Velazco was\\nable to consult with counsel before his removal order was actually\\nexecuted. Id. at 993–94. The same was not true here. Zuniga was denied\\nhis right to counsel at the final stage of the reasonable fear proceeding—\\nand thus, unlike the situation in Gomez-Velazco, Zuniga’s attorney had\\nno later opportunity to “remedy any damage done by [his] client’s un-\\ncounseled admissions or waivers.” Id. at 994.\\n\\f ZUNIGA V. BARR 15\\n\\nhis right to counsel is honored. Given this conclusion, we\\nneed not reach Zuniga’s other challenges to the proceedings\\nbefore the IJ or to the IJ’s reasoning.\\n\\n The Petition for Review is GRANTED and we\\nREMAND for further proceedings consistent with this\\nopinion.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4427024/\", \"author_raw\": \"PER CURIAM\"}]}"},"judge1":{"kind":"string","value":"ANDREW J KLEINFELD"},"judge2":{"kind":"string","value":"MICHELLE T FRIEDLAND"},"judge3":{"kind":"string","value":"WILLIAM H PAULEY III"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4649771/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Baldemar Zuniga v. William Barr\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-20\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n BALDEMAR ZUNIGA, No. 16-72982\n Petitioner,\n Agency No.\n v. A089-247-110\n\n WILLIAM P. BARR, Attorney General,\n Respondent. OPINION\n\n On Petition for Review of an\n Immigration Judge’s Decision\n\n Argued and Submitted May 17, 2019\n Seattle, Washington\n\n Filed August 20, 2019\n\n Before: Andrew J. Kleinfeld and Michelle T. Friedland,\nCircuit Judges, and William H. Pauley III, * District Judge.\n\n Per Curiam Opinion\n\n\n\n\n *\n The Honorable William H. Pauley III, United States District Judge\nfor the Southern District of New York, sitting by designation.\n\f2 ZUNIGA V. BARR\n\n SUMMARY **\n\n\n Immigration\n\n The panel granted Baldemar Zuniga’s petition for review\nof an immigration judge’s decision affirming an asylum\nofficer’s negative reasonable fear determination in expedited\nremoval proceedings, and remanded, holding that non-\ncitizens subject to expedited removal under 8 U.S.C. § 1228\nhave a statutory right to counsel in reasonable fear\nproceedings before immigration judges, and that the\nimmigration judge deprived Zuniga of his right to counsel\nby failing to obtain a knowing and voluntary waiver of that\nright.\n\n The panel rejected the government’s argument that there\nis no statutory right to counsel in reasonable fear\nproceedings. The panel explained that 8 U.S.C. § 1228,\nwhich governs expedited removal proceedings for non-\ncitizens convicted of committing aggravated felonies, and\nthrough which non-citizens can request reasonable fear\ninterviews, explicitly provides that non-citizens have the\nprivilege of being represented, at no expense to the\ngovernment, by counsel. The panel further explained that\nother subsections of § 1228 reinforce that right by requiring\nthat proceedings for the removal of criminal non-citizens be\nconducted in conformity with § 1229a, which in turn\nprovides a statutory right to counsel in ordinary removal\nproceedings, and directing the government to take\nreasonable efforts not to impair an individual’s access and\nright to counsel when considering whether to detain non-\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f ZUNIGA V. BARR 3\n\ncitizens. The panel also stated that the broader legislative\ncontext outside of the specific provisions dealing with\nexpedited removal proceedings for criminal non-citizens\nalso supports the conclusion that there is a right to counsel\nin reasonable fear proceedings.\n\n The panel rejected the government’s contention, which\nrelied on a 1999 memo from the Executive Office for\nImmigration Review interpreting 8 C.F.R. § 208.31, that IJs\nhave discretion to decide whether a non-citizen may be\nrepresented by counsel. The panel noted that the\ngovernment was correct that the regulations specify only that\nnon-citizens may be represented by counsel in the initial\nreasonable fear interview before an asylum officer, and that\nthey are silent as to representation by counsel in the review\nhearing before the IJ, but the panel concluded that deference\nto EOIR’s interpretation of that silence was not warranted\nbecause its interpretation conflicted with the plain text of\n8 U.S.C. § 1228.\n\n The panel held that the IJ violated Zuniga’s Fifth\nAmendment right to due process by failing to obtain a\nknowing and voluntary waiver of his right to counsel. The\npanel further held that Zuniga need not show prejudice\nwhere he was denied his statutory right.\n\n The panel remanded for a new hearing in which Zuniga’s\nright to counsel is honored.\n\f4 ZUNIGA V. BARR\n\n COUNSEL\n\nRobert Pauw (argued), Gibbs Houston Pauw, Seattle,\nWashington, for Petitioner.\n\nNehal Kamani (argued) and Kathryn M. McKinney,\nAttorneys; Stephen J. Flynn, Assistant Director; Office of\nImmigration Litigation, Civil Division, United States\nDepartment of Justice, Washington, D.C.; for Respondent.\n\n\n OPINION\n\nPER CURIAM:\n\n This case presents us with a simple question: do non-\ncitizens subject to expedited removal under 8 U.S.C. § 1228\nhave a statutory right to counsel in reasonable fear\nproceedings before immigration judges? The answer, based\non the plain language of § 1228, is yes.\n\n Petitioner Baldemar Zuniga contends that in his hearing\nbefore an Immigration Judge (“IJ”) to review a negative\nreasonable fear determination made by an asylum officer,\nthe IJ denied him his right to counsel. Because we conclude\nthat Zuniga had a statutory right to counsel, that the colloquy\nat the beginning of the hearing before the IJ was inadequate\nto waive that right, and that no showing of prejudice is\nrequired, we reverse and remand for further proceedings in\nwhich Zuniga is given the opportunity to proceed with\ncounsel.\n\n I.\n\n Baldemar Zuniga is a Mexican national who illegally\nentered the United States as a child. In 2012, he was\n\f ZUNIGA V. BARR 5\n\nconvicted of participating in a conspiracy to manufacture\nand distribute drugs and launder money. Zuniga testified in\nopen court against two of his co-conspirators, who were part\nof Mexico’s notorious Knights Templar cartel.\n\n While he was in prison, immigration authorities served\nZuniga with a Notice of Intent to Issue a Final\nAdministrative Removal Order. Because he had been\nconvicted of a drug trafficking aggravated felony, Zuniga\nwas placed in expedited removal proceedings pursuant to\n8 U.S.C. § 1228(b). During an interview with Immigration\nand Customs Enforcement (“ICE”), he expressed a fear that,\nif removed to Mexico, he would face torture or persecution\nby the cartel as retribution for his testimony. He was then\nreferred to an asylum officer for an interview to evaluate\nwhether his fear was “reasonable.”\n\n At the beginning of the reasonable fear interview, the\nasylum officer explained that Zuniga had “the right to have\n[his] attorney present for the interview.” Zuniga stated that\nhe had an attorney helping with his reasonable fear case but\nthat he was willing to continue with the reasonable fear\ninterview without his attorney. 1 The asylum officer then\nproceeded with the interview questions and concluded that\nZuniga had not established a reasonable fear of persecution\nor torture.\n\n Zuniga requested review of the asylum officer’s negative\nreasonable fear determination by an Immigration Judge.\n\n 1\n The asylum officer’s notes from the interview spell the name of\nthe attorney Zuniga mentioned as “Robert Pawl.” His current attorney,\nRobert Pauw, confirmed at oral argument before our court that he was\nalso representing Zuniga at the time of Zuniga’s reasonable fear\ninterview, and that he was the same attorney Zuniga mentioned to the\nasylum officer.\n\f6 ZUNIGA V. BARR\n\nZuniga’s case was referred to the Immigration Court in San\nFrancisco, California. The notice of referral to the IJ\nincluded the following language: “You may be represented\nin this proceeding, at no expense to the government, by an\nattorney or other individual authorized and qualified to\nrepresent persons before an Immigration Court. If you wish\nto be so represented, your attorney or representative should\nappear with you at this hearing.”\n\n Zuniga appeared at the hearing by video conference from\na detention center in Mesa Verde, California. 2 At the outset\nof the hearing, when announcing the case number and who\nwas participating, the IJ stated that Zuniga did “not have a\nlawyer.” Shortly thereafter, the IJ engaged in the following\ncolloquy with Zuniga:\n\n JUDGE: In these proceedings you have the\n right to counsel of your own choosing, but the\n government will not pay for your attorney.\n You should have received a copy of the free\n legal service list. Did you get that list?\n\n ZUNIGA: Yes.\n\n JUDGE: So, sir, do you have a lawyer?\n\n ZUNIGA: I do not.\n\n\n\n\n 2\n In a later declaration, Zuniga testified that the video setup “made\nit extremely difficult to understand everything that was happening” and\nthat he had been under the impression that the hearing “was going to be\na quick hearing where [he] was going to . . . ask for another extension.”\n\f ZUNIGA V. BARR 7\n\n JUDGE: All right, sir, were you interviewed\n by an asylum officer regarding your fear of\n returning to Mexico?\n\nThe IJ then proceeded with the substance of the hearing. The\nIJ ultimately agreed with the asylum officer that Zuniga\nlacked “a reasonable fear of persecution on account of a\nground protected by the law[s] of the United States.”\n\n Zuniga petitioned our court for review of that decision,\narguing that his due process rights were violated by the use\nof the video conference in his reasonable fear review hearing\nbefore the IJ and by the IJ’s failure to obtain a proper waiver\nof his right to an attorney in that proceeding. He also argued\nthat the IJ erred in determining that he lacked a reasonable\nfear of persecution on a protected ground and in evaluating\nhis fear of torture.\n\n II.\n\n We review de novo due process challenges to reasonable\nfear proceedings. Colmenar v. I.N.S., 210 F.3d 967, 971 (9th\nCir. 2000).\n\n III.\n\n A.\n\n In 1999, the Department of Justice created the reasonable\nfear interview to serve as a “screening process to evaluate\ntorture claims for aliens subject to streamlined\nadministrative removal processes for aggravated felons . . .\nand for aliens subject to reinstatement of a previous removal\norder.” Regulations Concerning the Convention Against\nTorture, 64 Fed. Reg. 8478, 8485 (Feb. 19, 1999). These\ntwo groups of non-citizens are ineligible for asylum but may\n\f8 ZUNIGA V. BARR\n\nbe entitled to withholding or deferral of removal under\n§ 241(b)(3) of the Immigration and Nationality Act (“INA”),\ncurrently codified at 8 U.S.C. § 1231(b)(3), or under the\nConvention Against Torture. 3 Regulations Concerning the\nConvention Against Torture, 64 Fed. Reg. at 8485.\n\n As a first step of the reasonable fear process, non-citizens\nwho express a fear of returning to their country of removal\nto ICE are interviewed in a “non-adversarial manner” by an\nasylum officer to determine whether they have a reasonable\nfear of persecution or torture. 8 C.F.R. §§ 208.31(a), (c),\n1208.31(a), (c). 4 If the asylum officer determines that a non-\n\n 3\n The reasonable fear screening process was modeled on the credible\nfear screening process, which had previously been created to provide a\nlimited avenue for relief for inadmissible arriving aliens who are subject\nto expedited removal under 8 U.S.C. § 1225. Such non-citizens can be\nremoved without further review unless they express either a credible fear\nof persecution or torture in the country of removal or an intent to seek\nasylum. 8 U.S.C. § 1225(b)(1)(A)–(B); 8 C.F.R. §§ 208.30, 1208.30.\nCredible fear requires establishing a “significant possibility” that the\nnon-citizen could be eligible for asylum if given an opportunity in a more\nfulsome proceeding. 8 U.S.C. § 1225(b)(1)(B)(v). The reasonable fear\nscreening process was designed to fulfill a similar function for non-\ncitizens being deported under 8 U.S.C. § 1228, but it requires them to\nmeet a more demanding standard—they must demonstrate a “reasonable\npossibility” of persecution or torture in the country of removal.\nRegulations Concerning the Convention Against Torture, 64 Fed. Reg.\nat 8485 (“Because the standard for showing entitlement to the[] forms of\nprotection [available to these non-citizen aggravated felons] (a\nprobability of persecution or torture) is significantly higher than the\nstandard for asylum (a well-founded fear of persecution), the screening\nstandard adopted for initial consideration of withholding and deferral\nrequests in these contexts is also higher.”); see also 8 C.F.R. §§ 208.31,\n1208.31.\n 4\n There are two identical sets of regulations contained in Chapters I\nand V of 8 C.F.R. that are applicable to the Department of Homeland\nSecurity and the Executive Office for Immigration Review, respectively.\n\f ZUNIGA V. BARR 9\n\ncitizen has a reasonable fear, he will be referred to an IJ for\na merits hearing to determine eligibility for withholding of\nremoval. Id. § 208.31(e). However, if the asylum officer\ndetermines that the non-citizen “does not have a reasonable\nfear of persecution or torture,” he “will be afforded the\nopportunity for an expeditious review of the negative\nscreening determination by an immigration judge.”\nRegulations Concerning the Convention Against Torture,\n64 Fed. Reg. at 8485 (emphasis added); see also 8 C.F.R.\n§ 208.31(f)–(g). If a non-citizen exercises his right to\nchallenge a negative reasonable fear determination, the IJ\nreviews the asylum officer’s determination de novo.\n8 C.F.R. § 208.31(f)–(g); Bartolome v. Sessions, 904 F.3d\n803, 809 (9th Cir. 2018).\n\n If the IJ upholds the negative screening determination, it\nconstitutes a final order of removal. 8 C.F.R. § 208.31(g)(1).\nBecause the IJ’s decision is the final agency action on the\nreasonable fear question, such a final removal order is ripe\nfor judicial review by a federal court of appeals. Ayala v.\nSessions, 855 F.3d 1012, 1015–16 (9th Cir. 2017).\n\n B.\n\n Zuniga argues that he had a right to counsel in his\nreasonable fear review hearing before the IJ, that he did not\nwaive that right because his colloquy with the IJ was\ninsufficient to demonstrate that he made a knowing and\nvoluntary waiver, and that this error requires automatic\n\n\nAliens and Nationality; Homeland Security; Reorganization of\nRegulations, 68 Fed. Reg. 9824, 9825 (Feb. 28, 2003). Because the text\nis the same in both sets of regulations, for the sake of simplicity we will\nrefer only to one set throughout the remainder of this opinion.\n\f10 ZUNIGA V. BARR\n\nreversal. The Government does not contest that, if Zuniga\nhad a right to counsel, there was no adequate waiver here. 5\nRather, the Government rests its defense to this petition\nprimarily on the argument that there is no statutory right to\ncounsel in reasonable fear proceedings. This argument\nclearly fails.\n\n Expedited removal proceedings for non-citizens\nconvicted of committing aggravated felonies (through which\nnon-citizens can request reasonable fear interviews) are a\ncreation of INA § 238(b), currently codified at 8 U.S.C.\n§ 1228. Section 1228 explicitly provides that non-citizens\n“shall have the privilege of being represented (at no expense\nto the government) by [] counsel.” 8 U.S.C.\n§ 1228(b)(4)(B). Other subsections of § 1228 reinforce the\nright to counsel. The statute provides, for example, that the\n“proceedings [for removal of criminal non-citizens] shall be\nconducted in conformity with section 1229a,” 8 U.S.C.\n§ 1228(a)(1), which provides a statutory right to counsel in\nordinary removal proceedings, 8 U.S.C. § 1229a(b)(4)(A). 6\n\n 5\n As explained below, we have previously held that if a non-citizen\nappears without counsel in a removal hearing, “there must be a knowing\nand voluntary waiver of the right to counsel” before the hearing can\nproceed. Tawadrus v. Ashcroft, 364 F.3d 1099, 1103 (9th Cir. 2004). A\nvalid waiver requires that the IJ “(1) inquire specifically as to whether\npetitioner wishes to continue without a lawyer; and (2) receive a knowing\nand voluntary affirmative response.” Id. (citations omitted).\n 6\n There is some overlap between 8 U.S.C. § 1228(a), which\ndescribes the general procedures for removing criminal non-citizens, and\n§ 1228(b), the provision under which Zuniga was removed, which\napplies specifically to the removal of criminal non-citizens who are not\npermanent residents. Both emphasize the need to conform with the\nprocedures described in 8 U.S.C. § 1229a, but § 1228(b) provides some\nspecific procedures applicable only to non-citizens who are not\npermanent residents.\n\f ZUNIGA V. BARR 11\n\nSection 1228 also states that in deciding where to detain non-\ncitizens under this section, “the Attorney General shall make\nreasonable efforts to ensure that the alien’s access to counsel\nand right to counsel under [8 U.S.C.] section 1362,” which\nlikewise provides for the right to counsel in removal\nproceedings, “are not impaired.” 8 U.S.C. § 1228(a)(2).\n\n This language makes clear that the statute contemplates\na right to counsel in removal proceedings initiated under\n8 U.S.C. § 1228, which includes reasonable fear review\nproceedings. Nothing in the language of § 1228 indicates\nthat the right to counsel is conditional or limited only to\ncertain types of proceedings authorized under that statute,\nexpedited or otherwise. Cf. United States v. Reyes-Bonilla,\n671 F.3d 1036, 1045 (9th Cir. 2012) (noting, with citation to\n8 U.S.C. § 1228, that the “right to counsel in expedited\nremoval proceedings is . . . secured by statute”).\n\n The broader legislative context—outside of the specific\nprovisions dealing with expedited removal proceedings for\ncriminal non-citizens—also supports the conclusion that\nthere is a right to counsel in reasonable fear proceedings.\nThe INA gives non-citizens the right to be represented by an\nattorney in most immigration proceedings as long as the\ngovernment does not have to bear the expense. In particular,\n8 U.S.C. § 1362 provides that “[i]n any removal proceedings\nbefore an immigration judge and in any appeal proceedings\nbefore the Attorney General from any such removal\nproceedings,” non-citizens “shall have the privilege of being\nrepresented (at no expense to the Government)” by counsel\nof their choosing. See also 8 U.S.C. § 1229a(b)(4)(A)\n(“[T]he alien shall have the privilege of being represented,\nat no expense to the Government, by counsel of the alien’s\nchoosing who is authorized to practice in such\nproceedings.”). Because the text of § 1228 is both “clear and\n\f12 ZUNIGA V. BARR\n\nconsistent with the statutory scheme at issue, the plain\nlanguage of the statute is conclusive” in providing a statutory\nright to counsel. Emmert Indus. Corp. v. Artisan Assocs.,\nInc., 497 F.3d 982, 987 (9th Cir. 2007) (quoting Molski v.\nM.J. Cable, Inc., 481 F.3d 724, 732 (9th Cir. 2007)).\n\n The Government resists this conclusion, arguing that the\nIJ has discretion to decide whether a non-citizen may be\nrepresented by counsel. The Government’s argument is\npremised on a 1999 memo from the Executive Office for\nImmigration Review (“EOIR”). That memo interpreted\n8 C.F.R. § 208.31 as giving an IJ the right to decide whether\na non-citizen may be represented in the negative reasonable\nfear determination review hearing because the regulation did\nnot explicitly mention the right to counsel in those\nproceedings. See Exec. Office for Immigration Review,\nOffice of the Chief Immigration Judge, Operating Policies\nand Procedures Memorandum No. 99-5: Implementation of\nArticle 3 of the UN Convention Against Torture 8 (1999)\n(“With regard to representation in reasonable fear review\nproceedings, the . . . regulations are . . . silent. . . . Since\nthere is no specific regulatory guidance on this point, the\nissue is left to the discretion of the Immigration Judge.”).\n\n The Government is correct that the regulations specify\nonly that non-citizens may be represented by counsel in the\ninitial reasonable fear interview before an asylum officer,\nand that they are silent as to representation by counsel in the\nreview hearing before the IJ. Compare 8 C.F.R. § 208.31(c),\nwith 8 C.F.R. § 208.31(g). But its argument that we should\ndefer to EOIR’s interpretation of that silence fails because\nthat interpretation conflicts with the plain text of 8 U.S.C.\n§ 1228. We “need not accept an agency’s interpretation of\nits own regulations if that interpretation is inconsistent with\nthe statute under which the regulations were promulgated.”\n\f ZUNIGA V. BARR 13\n\nTurtle Island Restoration Network v. U.S. Dep’t of\nCommerce, 878 F.3d 725, 733 (9th Cir. 2017) (quotation\nmarks omitted); see also Auer v. Robbins, 519 U.S. 452, 461\n(1997) (holding that courts should generally defer to an\nagency’s interpretation of its own regulation unless, among\nother circumstances, that interpretation is “plainly\nerroneous” (quoting Robertson v. Methow Valley Citizens\nCouncil, 490 U.S. 332, 359 (1989)). 7 We therefore conclude\nthat Zuniga had a right to representation by an attorney in his\nreasonable fear review hearing before the IJ.\n\n We have previously held that “[a]lthough there is no\nSixth Amendment right to counsel in an immigration\nhearing, Congress has recognized [a right to counsel] among\nthe rights stemming from the Fifth Amendment guarantee of\ndue process” by codifying it. Tawadrus v. Ashcroft,\n364 F.3d 1099, 1103 (9th Cir. 2004); see also Rios-Berrios\nv. I.N.S., 776 F.2d 859, 862 (9th Cir. 1985) (“[D]ue process\nmandates that [a non-citizen] is entitled to counsel of his own\nchoice at his own expense under terms of the [INA].”). And,\nfurther, before a petitioner continues without counsel in an\nimmigration proceeding for which there is a statutory right\nto counsel, due process mandates that “there must be a\nknowing and voluntary waiver of the right to counsel,”\n\n 7\n Nor is the agency’s interpretation persuasive enough to otherwise\ncommand respect. See Skidmore v. Swift & Co., 323 U.S. 134, 140\n(1944) (explaining that “[t]he weight of” an agency interpretation “will\ndepend upon the thoroughness evident in its consideration, the validity\nof its reasoning, its consistency with earlier and later pronouncements,\nand all those factors which give it power to persuade, if lacking power to\ncontrol”). In addition to being contrary to the statute, the EOIR memo’s\ninterpretation could lead to absurd results. For example, a non-citizen\nwho had been informed by his hearing notice that he could bring an\nattorney to represent him might arrive at the hearing only to learn that\ncounsel was barred from the hearing at the discretion of the IJ.\n\f14 ZUNIGA V. BARR\n\nwhich requires that the IJ “(1) inquire specifically as to\nwhether petitioner wishes to continue without a lawyer; and\n(2) receive a knowing and voluntary affirmative response.”\nTawadrus, 364 F.3d at 1103 (citations omitted). As the\nGovernment concedes, Zuniga’s colloquy with the IJ was\ninadequate to effect a valid waiver of the right to counsel\nunder Tawadrus.\n\n The Government contends that a remand is nevertheless\nunavailable to Zuniga because he cannot show prejudice\nfrom the lack of counsel. This contention is foreclosed by\nMontes-Lopez v. Holder, 694 F.3d 1085 (9th Cir. 2012), in\nwhich we held that “an alien who shows that he has been\ndenied the statutory right to be represented by counsel in an\nimmigration proceeding need not also show that he was\nprejudiced by the absence of the attorney.” Id. at 1093–94. 8\n\n IV.\n\n For the reasons described above, we conclude that the IJ\nviolated Zuniga’s right to counsel in his reasonable fear\nreview proceeding by failing to obtain a valid waiver, and\nthat Zuniga is entitled to a new hearing before an IJ in which\n\n 8\n Gomez-Velazco v. Sessions, 879 F.3d 989 (9th Cir. 2018), is not to\nthe contrary. There, we held that we could not presume prejudice where\na non-citizen, Gomez-Velazco, was denied counsel during one “discrete\nstage” of expedited removal proceedings—his initial interaction with\nofficers from the Department of Homeland Security. Id. at 994–95. In\nGomez-Velazco, we distinguished Montes-Lopez on the ground that,\nafter the discrete phase in which he lacked counsel, Gomez-Velazco was\nable to consult with counsel before his removal order was actually\nexecuted. Id. at 993–94. The same was not true here. Zuniga was denied\nhis right to counsel at the final stage of the reasonable fear proceeding—\nand thus, unlike the situation in Gomez-Velazco, Zuniga’s attorney had\nno later opportunity to “remedy any damage done by [his] client’s un-\ncounseled admissions or waivers.” Id. at 994.\n\f ZUNIGA V. BARR 15\n\nhis right to counsel is honored. Given this conclusion, we\nneed not reach Zuniga’s other challenges to the proceedings\nbefore the IJ or to the IJ’s reasoning.\n\n The Petition for Review is GRANTED and we\nREMAND for further proceedings consistent with this\nopinion.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Baldemar Zuniga v. William Barr\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-20\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n BALDEMAR ZUNIGA, No. 16-72982\\n Petitioner,\\n Agency No.\\n v. A089-247-110\\n\\n WILLIAM P. BARR, Attorney General,\\n Respondent. OPINION\\n\\n On Petition for Review of an\\n Immigration Judge’s Decision\\n\\n Argued and Submitted May 17, 2019\\n Seattle, Washington\\n\\n Filed August 20, 2019\\n\\n Before: Andrew J. Kleinfeld and Michelle T. Friedland,\\nCircuit Judges, and William H. Pauley III, * District Judge.\\n\\n Per Curiam Opinion\\n\\n\\n\\n\\n *\\n The Honorable William H. Pauley III, United States District Judge\\nfor the Southern District of New York, sitting by designation.\\n\\f2 ZUNIGA V. BARR\\n\\n SUMMARY **\\n\\n\\n Immigration\\n\\n The panel granted Baldemar Zuniga’s petition for review\\nof an immigration judge’s decision affirming an asylum\\nofficer’s negative reasonable fear determination in expedited\\nremoval proceedings, and remanded, holding that non-\\ncitizens subject to expedited removal under 8 U.S.C. § 1228\\nhave a statutory right to counsel in reasonable fear\\nproceedings before immigration judges, and that the\\nimmigration judge deprived Zuniga of his right to counsel\\nby failing to obtain a knowing and voluntary waiver of that\\nright.\\n\\n The panel rejected the government’s argument that there\\nis no statutory right to counsel in reasonable fear\\nproceedings. The panel explained that 8 U.S.C. § 1228,\\nwhich governs expedited removal proceedings for non-\\ncitizens convicted of committing aggravated felonies, and\\nthrough which non-citizens can request reasonable fear\\ninterviews, explicitly provides that non-citizens have the\\nprivilege of being represented, at no expense to the\\ngovernment, by counsel. The panel further explained that\\nother subsections of § 1228 reinforce that right by requiring\\nthat proceedings for the removal of criminal non-citizens be\\nconducted in conformity with § 1229a, which in turn\\nprovides a statutory right to counsel in ordinary removal\\nproceedings, and directing the government to take\\nreasonable efforts not to impair an individual’s access and\\nright to counsel when considering whether to detain non-\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f ZUNIGA V. BARR 3\\n\\ncitizens. The panel also stated that the broader legislative\\ncontext outside of the specific provisions dealing with\\nexpedited removal proceedings for criminal non-citizens\\nalso supports the conclusion that there is a right to counsel\\nin reasonable fear proceedings.\\n\\n The panel rejected the government’s contention, which\\nrelied on a 1999 memo from the Executive Office for\\nImmigration Review interpreting 8 C.F.R. § 208.31, that IJs\\nhave discretion to decide whether a non-citizen may be\\nrepresented by counsel. The panel noted that the\\ngovernment was correct that the regulations specify only that\\nnon-citizens may be represented by counsel in the initial\\nreasonable fear interview before an asylum officer, and that\\nthey are silent as to representation by counsel in the review\\nhearing before the IJ, but the panel concluded that deference\\nto EOIR’s interpretation of that silence was not warranted\\nbecause its interpretation conflicted with the plain text of\\n8 U.S.C. § 1228.\\n\\n The panel held that the IJ violated Zuniga’s Fifth\\nAmendment right to due process by failing to obtain a\\nknowing and voluntary waiver of his right to counsel. The\\npanel further held that Zuniga need not show prejudice\\nwhere he was denied his statutory right.\\n\\n The panel remanded for a new hearing in which Zuniga’s\\nright to counsel is honored.\\n\\f4 ZUNIGA V. BARR\\n\\n COUNSEL\\n\\nRobert Pauw (argued), Gibbs Houston Pauw, Seattle,\\nWashington, for Petitioner.\\n\\nNehal Kamani (argued) and Kathryn M. McKinney,\\nAttorneys; Stephen J. Flynn, Assistant Director; Office of\\nImmigration Litigation, Civil Division, United States\\nDepartment of Justice, Washington, D.C.; for Respondent.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n This case presents us with a simple question: do non-\\ncitizens subject to expedited removal under 8 U.S.C. § 1228\\nhave a statutory right to counsel in reasonable fear\\nproceedings before immigration judges? The answer, based\\non the plain language of § 1228, is yes.\\n\\n Petitioner Baldemar Zuniga contends that in his hearing\\nbefore an Immigration Judge (“IJ”) to review a negative\\nreasonable fear determination made by an asylum officer,\\nthe IJ denied him his right to counsel. Because we conclude\\nthat Zuniga had a statutory right to counsel, that the colloquy\\nat the beginning of the hearing before the IJ was inadequate\\nto waive that right, and that no showing of prejudice is\\nrequired, we reverse and remand for further proceedings in\\nwhich Zuniga is given the opportunity to proceed with\\ncounsel.\\n\\n I.\\n\\n Baldemar Zuniga is a Mexican national who illegally\\nentered the United States as a child. In 2012, he was\\n\\f ZUNIGA V. BARR 5\\n\\nconvicted of participating in a conspiracy to manufacture\\nand distribute drugs and launder money. Zuniga testified in\\nopen court against two of his co-conspirators, who were part\\nof Mexico’s notorious Knights Templar cartel.\\n\\n While he was in prison, immigration authorities served\\nZuniga with a Notice of Intent to Issue a Final\\nAdministrative Removal Order. Because he had been\\nconvicted of a drug trafficking aggravated felony, Zuniga\\nwas placed in expedited removal proceedings pursuant to\\n8 U.S.C. § 1228(b). During an interview with Immigration\\nand Customs Enforcement (“ICE”), he expressed a fear that,\\nif removed to Mexico, he would face torture or persecution\\nby the cartel as retribution for his testimony. He was then\\nreferred to an asylum officer for an interview to evaluate\\nwhether his fear was “reasonable.”\\n\\n At the beginning of the reasonable fear interview, the\\nasylum officer explained that Zuniga had “the right to have\\n[his] attorney present for the interview.” Zuniga stated that\\nhe had an attorney helping with his reasonable fear case but\\nthat he was willing to continue with the reasonable fear\\ninterview without his attorney. 1 The asylum officer then\\nproceeded with the interview questions and concluded that\\nZuniga had not established a reasonable fear of persecution\\nor torture.\\n\\n Zuniga requested review of the asylum officer’s negative\\nreasonable fear determination by an Immigration Judge.\\n\\n 1\\n The asylum officer’s notes from the interview spell the name of\\nthe attorney Zuniga mentioned as “Robert Pawl.” His current attorney,\\nRobert Pauw, confirmed at oral argument before our court that he was\\nalso representing Zuniga at the time of Zuniga’s reasonable fear\\ninterview, and that he was the same attorney Zuniga mentioned to the\\nasylum officer.\\n\\f6 ZUNIGA V. BARR\\n\\nZuniga’s case was referred to the Immigration Court in San\\nFrancisco, California. The notice of referral to the IJ\\nincluded the following language: “You may be represented\\nin this proceeding, at no expense to the government, by an\\nattorney or other individual authorized and qualified to\\nrepresent persons before an Immigration Court. If you wish\\nto be so represented, your attorney or representative should\\nappear with you at this hearing.”\\n\\n Zuniga appeared at the hearing by video conference from\\na detention center in Mesa Verde, California. 2 At the outset\\nof the hearing, when announcing the case number and who\\nwas participating, the IJ stated that Zuniga did “not have a\\nlawyer.” Shortly thereafter, the IJ engaged in the following\\ncolloquy with Zuniga:\\n\\n JUDGE: In these proceedings you have the\\n right to counsel of your own choosing, but the\\n government will not pay for your attorney.\\n You should have received a copy of the free\\n legal service list. Did you get that list?\\n\\n ZUNIGA: Yes.\\n\\n JUDGE: So, sir, do you have a lawyer?\\n\\n ZUNIGA: I do not.\\n\\n\\n\\n\\n 2\\n In a later declaration, Zuniga testified that the video setup “made\\nit extremely difficult to understand everything that was happening” and\\nthat he had been under the impression that the hearing “was going to be\\na quick hearing where [he] was going to . . . ask for another extension.”\\n\\f ZUNIGA V. BARR 7\\n\\n JUDGE: All right, sir, were you interviewed\\n by an asylum officer regarding your fear of\\n returning to Mexico?\\n\\nThe IJ then proceeded with the substance of the hearing. The\\nIJ ultimately agreed with the asylum officer that Zuniga\\nlacked “a reasonable fear of persecution on account of a\\nground protected by the law[s] of the United States.”\\n\\n Zuniga petitioned our court for review of that decision,\\narguing that his due process rights were violated by the use\\nof the video conference in his reasonable fear review hearing\\nbefore the IJ and by the IJ’s failure to obtain a proper waiver\\nof his right to an attorney in that proceeding. He also argued\\nthat the IJ erred in determining that he lacked a reasonable\\nfear of persecution on a protected ground and in evaluating\\nhis fear of torture.\\n\\n II.\\n\\n We review de novo due process challenges to reasonable\\nfear proceedings. Colmenar v. I.N.S., 210 F.3d 967, 971 (9th\\nCir. 2000).\\n\\n III.\\n\\n A.\\n\\n In 1999, the Department of Justice created the reasonable\\nfear interview to serve as a “screening process to evaluate\\ntorture claims for aliens subject to streamlined\\nadministrative removal processes for aggravated felons . . .\\nand for aliens subject to reinstatement of a previous removal\\norder.” Regulations Concerning the Convention Against\\nTorture, 64 Fed. Reg. 8478, 8485 (Feb. 19, 1999). These\\ntwo groups of non-citizens are ineligible for asylum but may\\n\\f8 ZUNIGA V. BARR\\n\\nbe entitled to withholding or deferral of removal under\\n§ 241(b)(3) of the Immigration and Nationality Act (“INA”),\\ncurrently codified at 8 U.S.C. § 1231(b)(3), or under the\\nConvention Against Torture. 3 Regulations Concerning the\\nConvention Against Torture, 64 Fed. Reg. at 8485.\\n\\n As a first step of the reasonable fear process, non-citizens\\nwho express a fear of returning to their country of removal\\nto ICE are interviewed in a “non-adversarial manner” by an\\nasylum officer to determine whether they have a reasonable\\nfear of persecution or torture. 8 C.F.R. §§ 208.31(a), (c),\\n1208.31(a), (c). 4 If the asylum officer determines that a non-\\n\\n 3\\n The reasonable fear screening process was modeled on the credible\\nfear screening process, which had previously been created to provide a\\nlimited avenue for relief for inadmissible arriving aliens who are subject\\nto expedited removal under 8 U.S.C. § 1225. Such non-citizens can be\\nremoved without further review unless they express either a credible fear\\nof persecution or torture in the country of removal or an intent to seek\\nasylum. 8 U.S.C. § 1225(b)(1)(A)–(B); 8 C.F.R. §§ 208.30, 1208.30.\\nCredible fear requires establishing a “significant possibility” that the\\nnon-citizen could be eligible for asylum if given an opportunity in a more\\nfulsome proceeding. 8 U.S.C. § 1225(b)(1)(B)(v). The reasonable fear\\nscreening process was designed to fulfill a similar function for non-\\ncitizens being deported under 8 U.S.C. § 1228, but it requires them to\\nmeet a more demanding standard—they must demonstrate a “reasonable\\npossibility” of persecution or torture in the country of removal.\\nRegulations Concerning the Convention Against Torture, 64 Fed. Reg.\\nat 8485 (“Because the standard for showing entitlement to the[] forms of\\nprotection [available to these non-citizen aggravated felons] (a\\nprobability of persecution or torture) is significantly higher than the\\nstandard for asylum (a well-founded fear of persecution), the screening\\nstandard adopted for initial consideration of withholding and deferral\\nrequests in these contexts is also higher.”); see also 8 C.F.R. §§ 208.31,\\n1208.31.\\n 4\\n There are two identical sets of regulations contained in Chapters I\\nand V of 8 C.F.R. that are applicable to the Department of Homeland\\nSecurity and the Executive Office for Immigration Review, respectively.\\n\\f ZUNIGA V. BARR 9\\n\\ncitizen has a reasonable fear, he will be referred to an IJ for\\na merits hearing to determine eligibility for withholding of\\nremoval. Id. § 208.31(e). However, if the asylum officer\\ndetermines that the non-citizen “does not have a reasonable\\nfear of persecution or torture,” he “will be afforded the\\nopportunity for an expeditious review of the negative\\nscreening determination by an immigration judge.”\\nRegulations Concerning the Convention Against Torture,\\n64 Fed. Reg. at 8485 (emphasis added); see also 8 C.F.R.\\n§ 208.31(f)–(g). If a non-citizen exercises his right to\\nchallenge a negative reasonable fear determination, the IJ\\nreviews the asylum officer’s determination de novo.\\n8 C.F.R. § 208.31(f)–(g); Bartolome v. Sessions, 904 F.3d\\n803, 809 (9th Cir. 2018).\\n\\n If the IJ upholds the negative screening determination, it\\nconstitutes a final order of removal. 8 C.F.R. § 208.31(g)(1).\\nBecause the IJ’s decision is the final agency action on the\\nreasonable fear question, such a final removal order is ripe\\nfor judicial review by a federal court of appeals. Ayala v.\\nSessions, 855 F.3d 1012, 1015–16 (9th Cir. 2017).\\n\\n B.\\n\\n Zuniga argues that he had a right to counsel in his\\nreasonable fear review hearing before the IJ, that he did not\\nwaive that right because his colloquy with the IJ was\\ninsufficient to demonstrate that he made a knowing and\\nvoluntary waiver, and that this error requires automatic\\n\\n\\nAliens and Nationality; Homeland Security; Reorganization of\\nRegulations, 68 Fed. Reg. 9824, 9825 (Feb. 28, 2003). Because the text\\nis the same in both sets of regulations, for the sake of simplicity we will\\nrefer only to one set throughout the remainder of this opinion.\\n\\f10 ZUNIGA V. BARR\\n\\nreversal. The Government does not contest that, if Zuniga\\nhad a right to counsel, there was no adequate waiver here. 5\\nRather, the Government rests its defense to this petition\\nprimarily on the argument that there is no statutory right to\\ncounsel in reasonable fear proceedings. This argument\\nclearly fails.\\n\\n Expedited removal proceedings for non-citizens\\nconvicted of committing aggravated felonies (through which\\nnon-citizens can request reasonable fear interviews) are a\\ncreation of INA § 238(b), currently codified at 8 U.S.C.\\n§ 1228. Section 1228 explicitly provides that non-citizens\\n“shall have the privilege of being represented (at no expense\\nto the government) by [] counsel.” 8 U.S.C.\\n§ 1228(b)(4)(B). Other subsections of § 1228 reinforce the\\nright to counsel. The statute provides, for example, that the\\n“proceedings [for removal of criminal non-citizens] shall be\\nconducted in conformity with section 1229a,” 8 U.S.C.\\n§ 1228(a)(1), which provides a statutory right to counsel in\\nordinary removal proceedings, 8 U.S.C. § 1229a(b)(4)(A). 6\\n\\n 5\\n As explained below, we have previously held that if a non-citizen\\nappears without counsel in a removal hearing, “there must be a knowing\\nand voluntary waiver of the right to counsel” before the hearing can\\nproceed. Tawadrus v. Ashcroft, 364 F.3d 1099, 1103 (9th Cir. 2004). A\\nvalid waiver requires that the IJ “(1) inquire specifically as to whether\\npetitioner wishes to continue without a lawyer; and (2) receive a knowing\\nand voluntary affirmative response.” Id. (citations omitted).\\n 6\\n There is some overlap between 8 U.S.C. § 1228(a), which\\ndescribes the general procedures for removing criminal non-citizens, and\\n§ 1228(b), the provision under which Zuniga was removed, which\\napplies specifically to the removal of criminal non-citizens who are not\\npermanent residents. Both emphasize the need to conform with the\\nprocedures described in 8 U.S.C. § 1229a, but § 1228(b) provides some\\nspecific procedures applicable only to non-citizens who are not\\npermanent residents.\\n\\f ZUNIGA V. BARR 11\\n\\nSection 1228 also states that in deciding where to detain non-\\ncitizens under this section, “the Attorney General shall make\\nreasonable efforts to ensure that the alien’s access to counsel\\nand right to counsel under [8 U.S.C.] section 1362,” which\\nlikewise provides for the right to counsel in removal\\nproceedings, “are not impaired.” 8 U.S.C. § 1228(a)(2).\\n\\n This language makes clear that the statute contemplates\\na right to counsel in removal proceedings initiated under\\n8 U.S.C. § 1228, which includes reasonable fear review\\nproceedings. Nothing in the language of § 1228 indicates\\nthat the right to counsel is conditional or limited only to\\ncertain types of proceedings authorized under that statute,\\nexpedited or otherwise. Cf. United States v. Reyes-Bonilla,\\n671 F.3d 1036, 1045 (9th Cir. 2012) (noting, with citation to\\n8 U.S.C. § 1228, that the “right to counsel in expedited\\nremoval proceedings is . . . secured by statute”).\\n\\n The broader legislative context—outside of the specific\\nprovisions dealing with expedited removal proceedings for\\ncriminal non-citizens—also supports the conclusion that\\nthere is a right to counsel in reasonable fear proceedings.\\nThe INA gives non-citizens the right to be represented by an\\nattorney in most immigration proceedings as long as the\\ngovernment does not have to bear the expense. In particular,\\n8 U.S.C. § 1362 provides that “[i]n any removal proceedings\\nbefore an immigration judge and in any appeal proceedings\\nbefore the Attorney General from any such removal\\nproceedings,” non-citizens “shall have the privilege of being\\nrepresented (at no expense to the Government)” by counsel\\nof their choosing. See also 8 U.S.C. § 1229a(b)(4)(A)\\n(“[T]he alien shall have the privilege of being represented,\\nat no expense to the Government, by counsel of the alien’s\\nchoosing who is authorized to practice in such\\nproceedings.”). Because the text of § 1228 is both “clear and\\n\\f12 ZUNIGA V. BARR\\n\\nconsistent with the statutory scheme at issue, the plain\\nlanguage of the statute is conclusive” in providing a statutory\\nright to counsel. Emmert Indus. Corp. v. Artisan Assocs.,\\nInc., 497 F.3d 982, 987 (9th Cir. 2007) (quoting Molski v.\\nM.J. Cable, Inc., 481 F.3d 724, 732 (9th Cir. 2007)).\\n\\n The Government resists this conclusion, arguing that the\\nIJ has discretion to decide whether a non-citizen may be\\nrepresented by counsel. The Government’s argument is\\npremised on a 1999 memo from the Executive Office for\\nImmigration Review (“EOIR”). That memo interpreted\\n8 C.F.R. § 208.31 as giving an IJ the right to decide whether\\na non-citizen may be represented in the negative reasonable\\nfear determination review hearing because the regulation did\\nnot explicitly mention the right to counsel in those\\nproceedings. See Exec. Office for Immigration Review,\\nOffice of the Chief Immigration Judge, Operating Policies\\nand Procedures Memorandum No. 99-5: Implementation of\\nArticle 3 of the UN Convention Against Torture 8 (1999)\\n(“With regard to representation in reasonable fear review\\nproceedings, the . . . regulations are . . . silent. . . . Since\\nthere is no specific regulatory guidance on this point, the\\nissue is left to the discretion of the Immigration Judge.”).\\n\\n The Government is correct that the regulations specify\\nonly that non-citizens may be represented by counsel in the\\ninitial reasonable fear interview before an asylum officer,\\nand that they are silent as to representation by counsel in the\\nreview hearing before the IJ. Compare 8 C.F.R. § 208.31(c),\\nwith 8 C.F.R. § 208.31(g). But its argument that we should\\ndefer to EOIR’s interpretation of that silence fails because\\nthat interpretation conflicts with the plain text of 8 U.S.C.\\n§ 1228. We “need not accept an agency’s interpretation of\\nits own regulations if that interpretation is inconsistent with\\nthe statute under which the regulations were promulgated.”\\n\\f ZUNIGA V. BARR 13\\n\\nTurtle Island Restoration Network v. U.S. Dep’t of\\nCommerce, 878 F.3d 725, 733 (9th Cir. 2017) (quotation\\nmarks omitted); see also Auer v. Robbins, 519 U.S. 452, 461\\n(1997) (holding that courts should generally defer to an\\nagency’s interpretation of its own regulation unless, among\\nother circumstances, that interpretation is “plainly\\nerroneous” (quoting Robertson v. Methow Valley Citizens\\nCouncil, 490 U.S. 332, 359 (1989)). 7 We therefore conclude\\nthat Zuniga had a right to representation by an attorney in his\\nreasonable fear review hearing before the IJ.\\n\\n We have previously held that “[a]lthough there is no\\nSixth Amendment right to counsel in an immigration\\nhearing, Congress has recognized [a right to counsel] among\\nthe rights stemming from the Fifth Amendment guarantee of\\ndue process” by codifying it. Tawadrus v. Ashcroft,\\n364 F.3d 1099, 1103 (9th Cir. 2004); see also Rios-Berrios\\nv. I.N.S., 776 F.2d 859, 862 (9th Cir. 1985) (“[D]ue process\\nmandates that [a non-citizen] is entitled to counsel of his own\\nchoice at his own expense under terms of the [INA].”). And,\\nfurther, before a petitioner continues without counsel in an\\nimmigration proceeding for which there is a statutory right\\nto counsel, due process mandates that “there must be a\\nknowing and voluntary waiver of the right to counsel,”\\n\\n 7\\n Nor is the agency’s interpretation persuasive enough to otherwise\\ncommand respect. See Skidmore v. Swift & Co., 323 U.S. 134, 140\\n(1944) (explaining that “[t]he weight of” an agency interpretation “will\\ndepend upon the thoroughness evident in its consideration, the validity\\nof its reasoning, its consistency with earlier and later pronouncements,\\nand all those factors which give it power to persuade, if lacking power to\\ncontrol”). In addition to being contrary to the statute, the EOIR memo’s\\ninterpretation could lead to absurd results. For example, a non-citizen\\nwho had been informed by his hearing notice that he could bring an\\nattorney to represent him might arrive at the hearing only to learn that\\ncounsel was barred from the hearing at the discretion of the IJ.\\n\\f14 ZUNIGA V. BARR\\n\\nwhich requires that the IJ “(1) inquire specifically as to\\nwhether petitioner wishes to continue without a lawyer; and\\n(2) receive a knowing and voluntary affirmative response.”\\nTawadrus, 364 F.3d at 1103 (citations omitted). As the\\nGovernment concedes, Zuniga’s colloquy with the IJ was\\ninadequate to effect a valid waiver of the right to counsel\\nunder Tawadrus.\\n\\n The Government contends that a remand is nevertheless\\nunavailable to Zuniga because he cannot show prejudice\\nfrom the lack of counsel. This contention is foreclosed by\\nMontes-Lopez v. Holder, 694 F.3d 1085 (9th Cir. 2012), in\\nwhich we held that “an alien who shows that he has been\\ndenied the statutory right to be represented by counsel in an\\nimmigration proceeding need not also show that he was\\nprejudiced by the absence of the attorney.” Id. at 1093–94. 8\\n\\n IV.\\n\\n For the reasons described above, we conclude that the IJ\\nviolated Zuniga’s right to counsel in his reasonable fear\\nreview proceeding by failing to obtain a valid waiver, and\\nthat Zuniga is entitled to a new hearing before an IJ in which\\n\\n 8\\n Gomez-Velazco v. Sessions, 879 F.3d 989 (9th Cir. 2018), is not to\\nthe contrary. There, we held that we could not presume prejudice where\\na non-citizen, Gomez-Velazco, was denied counsel during one “discrete\\nstage” of expedited removal proceedings—his initial interaction with\\nofficers from the Department of Homeland Security. Id. at 994–95. In\\nGomez-Velazco, we distinguished Montes-Lopez on the ground that,\\nafter the discrete phase in which he lacked counsel, Gomez-Velazco was\\nable to consult with counsel before his removal order was actually\\nexecuted. Id. at 993–94. The same was not true here. Zuniga was denied\\nhis right to counsel at the final stage of the reasonable fear proceeding—\\nand thus, unlike the situation in Gomez-Velazco, Zuniga’s attorney had\\nno later opportunity to “remedy any damage done by [his] client’s un-\\ncounseled admissions or waivers.” Id. at 994.\\n\\f ZUNIGA V. BARR 15\\n\\nhis right to counsel is honored. Given this conclusion, we\\nneed not reach Zuniga’s other challenges to the proceedings\\nbefore the IJ or to the IJ’s reasoning.\\n\\n The Petition for Review is GRANTED and we\\nREMAND for further proceedings consistent with this\\nopinion.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4087,"cells":{"id":{"kind":"number","value":4654476,"string":"4,654,476"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Jeffrey Green"},"decision_date":{"kind":"string","value":"2019-08-21"},"docket_number":{"kind":"string","value":"17-30227"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: A. Wallace Tashima, William A. Fletcher, and Marsha S. Berzon, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"BERZON, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-30227\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:16-cr-00063-SLG-1\\n\\nJEFFREY R. GREEN, OPINION\\n Defendant-Appellant.\\n\\n\\n Appeal from the United States District Court\\n for the District of Alaska\\n Sharon L. Gleason, District Judge, Presiding\\n\\n Argued and Submitted June 12, 2019\\n Anchorage, Alaska\\n\\n Filed August 21, 2019\\n\\n Before: A. Wallace Tashima, William A. Fletcher,\\n and Marsha S. Berzon, Circuit Judges.\\n\\n Opinion by Judge Berzon\\n\\f2 UNITED STATES V. GREEN\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n Vacating a sentence and remanding for resentencing, the\\npanel held that the district court erred by concluding that it\\ncould not listen to the defendant’s allocution before\\ndetermining whether a reduction of acceptance of\\nresponsibility was warranted under the Sentencing\\nGuidelines, and that this misapprehension was plain error\\nthat affected the defendant’s substantial rights and seriously\\naffected the fairness of the proceedings.\\n\\n\\n COUNSEL\\n\\nKrista Hart (argued), Sacramento, California, for Defendant-\\nAppellant.\\n\\nJonas M. Walker (argued), Assistant United States Attorney;\\nBrian Schroder, United States Attorney; United States\\nAttorney’s Office, Anchorage, Alaska; for Plaintiff-\\nAppellee.\\n\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. GREEN 3\\n\\n OPINION\\n\\nBERZON, Circuit Judge:\\n\\n Must a district court decide on a defendant’s eligibility\\nfor an acceptance-of-responsibility reduction in his\\nGuidelines level before listening to the defendant’s\\nallocution? Our answer is “No.”\\n\\n I\\n\\n On June 3, 2016, a group of police officers went to\\nJeffrey Green’s apartment in Anchorage, Alaska, and\\narrested Green on an outstanding warrant. While patting him\\ndown, an officer found a loaded revolver in Green’s pocket.\\nDuring a later search, the officers found two pistols stored\\ninside a safe in a storage closet accessible from the\\napartment. Both pistols had been reported stolen.\\n\\n The government charged Green, who had a long history\\nof felony convictions, with a single count of possession of a\\nfirearm as a felon in violation of 18 U.S.C. § 922(g)(1). Six\\nmonths later, Green pleaded guilty. During his plea\\ncolloquy, Green admitted that he possessed the single\\nrevolver found in his pocket during the arrest and that he was\\na felon. But Green did not admit to all the conduct alleged\\nin the single-count indictment. He made no admissions—or\\nstatements of any kind—regarding either of the pistols found\\nin the safe. 1 The district court found that Green’s admission\\nregarding the revolver, coupled with his admission regarding\\n\\n\\n\\n 1\\n The government offered Green a plea deal contingent on Green\\nadmitting possession of all three firearms. Green rejected this offer and\\ninstead pleaded guilty without an agreement with the government.\\n\\f4 UNITED STATES V. GREEN\\n\\nhis criminal history, provided a sufficient factual basis for\\nthe plea under Federal Rule of Criminal Procedure 11(b)(3).\\n\\n After Green pleaded guilty, the district court directed the\\nprobation department to prepare a presentence report. That\\nreport concluded that Green should be assessed an offense\\nlevel under the Sentencing Guidelines premised on\\npossession of a total of three guns, two of which were stolen.\\nSee U.S. Sentencing Guidelines Manual § 2K2.1(b)(1),\\n(b)(4)(A) (U.S. Sentencing Comm’n 2016). 2 The\\npresentence report also concluded that Green was not\\nentitled to any reduction for accepting responsibility because\\nhe had not admitted possession of the two pistols found in\\nthe safe. See id. § 3E1.1(a). Green objected to each of these\\nconclusions. He primarily argued that the government\\n“ha[d] not proven that the two additional firearms found in\\nthe storage closet were in Green’s possession.”\\n\\n Because Green so objected, the court held an evidentiary\\nhearing to determine whether Green possessed those\\nfirearms. See id. § 6A1.3 cmt. At this hearing, the\\ngovernment introduced a recording of Green speaking to a\\nwoman by phone after he was arrested. Green asked the\\nwoman, “Did they get my safe?” The woman replied, “I\\ndon’t know. There was a locksmith. They had a locksmith\\ncome there. Yes. Shane said they got into it.” Green then\\nresponded: “Oh, my God.” Based on this audio and other\\nevidence introduced by the government the district court\\nconcluded that the government had shown by a\\n\\n 2\\n These Guidelines sections provide that “[i]f the offense involved\\nthree [to seven] firearms, increase” the offense level by two levels,\\nGuidelines Manual § 2K2.1(b)(1), and that “[i]f any firearm [involved in\\nthe offense] was stolen, increase [the offense level] by 2 levels,” id.\\n§ 2K2.1(b)(4)(A).\\n\\f UNITED STATES V. GREEN 5\\n\\npreponderance of the evidence that Green possessed the two\\npistols, and that he should thus be assessed the offense level\\nfor possession of stolen guns and for possession of three or\\nmore guns. The court left open whether Green should be\\nawarded a reduction for accepting responsibility pursuant to\\nSentencing Guidelines section 3E1.1.\\n\\n Two weeks later, on November 3, 2017, the district court\\nheld a second sentencing hearing. At the outset of this\\nsecond hearing, the court entertained argument as to whether\\nit should find that Green accepted responsibility under\\nsection 3E1.1. During argument, defense counsel told the\\ncourt that Green “intends to allocute to this Court.” Counsel\\nfurther stated that he thought “the only way [Green will] be\\nable to express [the] contrition [required by section 3E1.1]\\n. . . is in that allocution.”\\n\\n After hearing counsel’s argument but before hearing\\nGreen’s allocution the court announced its conclusion\\nregarding the acceptance-of-responsibility reduction—that\\nthe reduction was not appropriate. The sentencing court\\nexplained that it reached this conclusion largely because it\\nviewed this case as analogous to United States v. Ginn,\\n87 F.3d 367 (9th Cir. 1996), which held that an acceptance-\\nof-responsibility reduction was appropriate only where a\\ndefendant charged with multiple counts had accepted\\nresponsibility for all of the “counts of which he is\\nconvicted.” Id. at 370; cf. United States v. Garrido, 596 F.3d\\n613, 619 (9th Cir. 2010) (holding that to remain eligible for\\nthe acceptance-of-responsibility reduction, a defendant need\\nnot accept responsibility for counts excluded from grouping\\nunder the Guidelines).\\n\\n The district court recognized that “evaluating the\\nacceptance[-of-responsibility reduction] . . . after an\\nallocution might be helpful” in some circumstances. But it\\n\\f6 UNITED STATES V. GREEN\\n\\nbelieved that under our case law, it could not hear from\\nGreen before determining the applicability of the reduction.\\nIn the district court’s view, “the Ninth Circuit very clearly\\ninstructs district judges to determine the guidelines at the\\noutset of sentencing proceeding[s],” and so before hearing\\nfrom the defendant.\\n\\n After explaining its decision regarding the acceptance-\\nof-responsibility reduction, the court heard further argument\\nfrom counsel on the appropriate sentence under the statutory\\nsentencing factors. See 18 U.S.C. § 3553(a). It then\\nprovided Green his opportunity to speak. During his\\nallocution, Green explained at length that he was “extremely\\nsorry” for his actions.\\n\\n Immediately following the allocution, the court\\nsentenced Green to a 108-month term of imprisonment. The\\nimposed sentence was within the range of 100 to 120 months\\nrecommended by the Guidelines calculation adopted by the\\ndistrict court. Had Green received the acceptance-of-\\nresponsibility reduction, the Guidelines recommended term\\nwould have been between 77 and 96 months. Green appeals\\nhis sentence.\\n\\n II\\n\\n At issue in this appeal is the district court’s decision not\\nto reduce Green’s offense level under section 3E1.1 of the\\nSentencing Guidelines based on acceptance of\\nresponsibility. We review any factual finding embedded in\\nthat determination for clear error. United States v.\\nRodriguez, 851 F.3d 931, 949 (9th Cir. 2017). But we\\n“review de novo whether the district court misapprehended\\nthe law with respect to the acceptance of responsibility\\nreduction.” United States v. Cortes, 299 F.3d 1030, 1037\\n\\f UNITED STATES V. GREEN 7\\n\\n(9th Cir. 2002); see also United States v. Doe, 778 F.3d 814,\\n821 (9th Cir. 2015).\\n\\n A misapprehension of law occurred here. The district\\ncourt believed that it had to determine whether Green had\\naccepted responsibility before allowing Green his\\nopportunity to allocute. That is not so.\\n\\n Green did not raise this error in the district court or in his\\ninitial briefing, so we review for plain error. See United\\nStates v. Sum of $185,336.07 U.S. Currency, 731 F.3d 189,\\n195 (2d Cir.2013); United States v. McKinney, 707 F.2d 381,\\n383 (9th Cir. 1983). We conclude that this misapprehension\\nof law constituted plain error, vacate Green’s sentence, and\\nremand for resentencing.\\n\\n A\\n\\n Before addressing the merits of this appeal, a brief\\nreview of the acceptance-of-responsibility reduction is\\nhelpful. Sentencing Guidelines section 3E1.1(a) calls for\\nreducing a defendant’s offense level by two points when the\\ndistrict court finds that the defendant has “clearly\\ndemonstrate[d] acceptance of responsibility for his offense.”\\n“The primary goal of the reduction is to reward defendants\\nwho are genuinely contrite.” United States v. McKinney,\\n15 F.3d 849, 853 (9th Cir. 1994). When a defendant’s\\n“statements and conduct ma[k]e it clear that his contrition\\n[is] sincere, he [is] entitled to the reduction,” Cortes,\\n299 F.3d at 1038, even if that contrition is expressed at the\\n“eleventh[]hour,” United States v. Hill, 953 F.2d 452, 461\\n(9th Cir. 1991).\\n\\n The Sentencing Guidelines commentary provides courts\\nwith guidance on how to evaluate whether a defendant has\\ndemonstrated the requisite contrition. The Guidelines\\n\\f8 UNITED STATES V. GREEN\\n\\ncommentary instructs that courts may consider whether the\\ndefendant has, for example, “truthfully admitt[ed] the\\nconduct comprising the offense(s) of conviction,”\\n“voluntar[ily] terminat[ed] or withdraw[n] from criminal\\nconduct or associations,” or “voluntar[illy] pa[id] restitution\\nprior to adjudication of guilt.” See Guidelines Manual\\n§ 3E1.1 cmt. n.1.\\n\\n The Guidelines and our case law also provide specific\\nguidance on how sentencing courts should evaluate the\\nacceptance-of-responsibility reduction where, as here, the\\ndefendant has entered a plea of guilty. The Guidelines\\ncommentary explains: “Entry of a plea of guilty prior to the\\ncommencement of trial combined with truthfully admitting\\nthe conduct comprising the offense of conviction, and\\ntruthfully admitting or not falsely denying any additional\\nrelevant conduct for which he is accountable . . . will\\nconstitute significant evidence of acceptance of\\nresponsibility.” Id. § 3E1.1 cmt. n.3. When a district court\\ndetermines that a defendant’s plea was adequately supported\\nby the defendant’s admissions, as the district court did here,\\nthat determination “compels the inference that [the\\ndefendant] had ‘truthfully admitt[ed] the conduct\\ncomprising the offense of conviction.’” United States v.\\nVance, 62 F.3d 1152, 1158 (9th Cir. 1995) (second alteration\\nin original). Although the Guidelines thus suggest that a\\nguilty plea supported by truthful admissions by the\\ndefendant creates a presumption that the defendant will\\nreceive the acceptance-of-responsibility reduction, this\\npresumption can be negated by other evidence that suggests\\nthe defendant has not in fact accepted responsibility for his\\nactions. Guidelines Manual § 3E1.1 cmt. n.3. Such\\ninconsistent conduct can include, for example, “falsely\\nden[ying], or frivolously contest[ing], relevant conduct that\\nthe court determines to be true.” Id. § 3E1.1 cmt. n.1(A).\\n\\f UNITED STATES V. GREEN 9\\n\\n B\\n\\n The district court in this case did not cite any specific\\nauthority to support its premise that a sentencing court must\\nreach its conclusion regarding acceptance of responsibility\\nbefore hearing from the defendant. The government argues\\non appeal that Kimbrough v. United States, 552 U.S. 85, 108\\n(2007), Gall v. United States, 552 U.S. 38, 49 (2007), and\\nUnited States v. Carty, 520 F.3d 984, 991 (9th Cir. 2008) (en\\nbanc), preclude a sentencing court from hearing a\\ndefendant’s allocution before determining whether an\\nacceptance-of-responsibility reduction is appropriate.\\n\\n None of these cases so limits a sentencing court.\\nKimbrough, Gall, and Carty each make clear that sentencing\\ncourts “must treat the Guidelines as the ‘starting point and\\nthe initial benchmark’” during sentencing. Kimbrough,\\n552 U.S. at 108 (quoting Gall, 552 U.S. at 49); see also Rita\\nv. United States, 551 U.S. 338, 347–38 (2007). Carty\\nexplained that “[a]ll sentencing proceedings are to begin by\\ndetermining the applicable Guidelines range.” 520 F.3d at\\n991. Likewise, Gall states that “a district court should begin\\nall sentencing proceedings by correctly calculating the\\napplicable Guidelines range.” 552 U.S. at 49.\\n\\n The language in these cases, taken out of context, can\\nperhaps be read to comport with the district court’s\\nunderstanding. In context, however, it is apparent that these\\ncases were not addressing the question before us—whether\\na district court must begin a sentencing hearing with a\\nGuidelines calculation rather than announcing it once (1) all\\nnecessary factual information has been presented, (2) all\\nrelevant factual issues have been resolved, and (3) all\\npertinent argument has been heard.\\n\\f10 UNITED STATES V. GREEN\\n\\n The cases explain that when sentencing a defendant, a\\ndistrict court must determine the recommended sentencing\\nrange under the Sentencing Guidelines before making a\\nholistic and individualized determination as to the\\nappropriate sentence under those Guidelines and the\\n18 U.S.C. § 3553(a) factors. Carty, 520 F.3d at 991. It is in\\nthat sense that the ultimate “sentencing proceedings”—that\\nis, the determination and announcement of the sentence to\\nbe imposed—“are to begin by determining the applicable\\nGuidelines range.” Id. None of these cases requires a district\\ncourt to make a final calculation as to the appropriate\\nGuidelines range at the outset of a sentencing hearing, even\\nif the judge lacks the necessary information and legal\\nargumentation to make an accurate, well-considered\\ndetermination.\\n\\n Indeed, in this case the district court did conduct an\\nevidentiary hearing to determine whether it was appropriate\\nto impose additional offense levels for the possession of\\nstolen firearms, and the possession of three or more firearms.\\nIf Kimbrough, Gall, and Carty set forth the rigid rule the\\ngovernment now suggests, that rule would also make\\nholding such an evidentiary hearing improper. Thankfully,\\nKimbrough, Gall, and Carty do not set forth such a\\ncounterproductive requirement. Neither the Guidelines nor\\nthe pertinent case law prefers ignorance over appropriate\\ninformation collection and considered judicial reflection\\nbefore calculating the applicable Guidelines range.\\n\\n Three additional considerations support our conclusion\\nthat a district court can hear from a defendant before\\ndetermining whether an acceptance-of-responsibility\\nreduction is indicated under the Guidelines.\\n\\n First, as a matter of practice, sentencing courts\\nfrequently consider the defendant’s allocution when\\n\\f UNITED STATES V. GREEN 11\\n\\nconsidering whether an acceptance-of-responsibility\\nreduction is warranted under the Guidelines. See, e.g.,\\nUnited States v. Goodson, 920 F.3d 1209, 1211 (8th Cir.\\n2019); United States v. Leasure, 122 F.3d 837, 841 (9th Cir.\\n1997); Hill, 953 F.2d at 461, United States v. Watt, 910 F.2d\\n587, 589, (9th Cir. 1990); see also United States v. Harvey,\\n597 F. App’x 455, 456 (9th Cir. 2015); United States v.\\nKamsomphou, 111 F. App’x 937, 940 (9th Cir. 2004);\\nUnited States v. Thompson, 49 F. App’x 749, 750 (9th Cir.\\n2002).\\n\\n Second, the Sentencing Guidelines recognize that, in\\nsome cases, a “factor important to the sentencing\\ndetermination [will be] reasonably in dispute.” Guidelines\\nManual § 6A1.3(a). The Guidelines state that, in these\\ncircumstances, “the parties shall be given an adequate\\nopportunity to present information to the court regarding that\\nfactor.” Id. (emphasis added). Thus, the Guidelines\\nrecognize that in some circumstances, “determining the\\napplicable Guidelines range,” Carty, 520 F.3d at 991, may\\nrequire a sentencing judge to evaluate affidavits, hear\\ntestimony or hold argument, depending on the nature of the\\nfactor in dispute. See Guidelines Manual § 6A1.3(a).\\n\\n Third, and relatedly, when a court must resolve “disputed\\nsentencing factors” as part of its sentencing determination,\\nthe Guidelines state that the sentencing court “shall resolve\\n[those] disputed . . . factors . . . in accordance with [Federal]\\nRule [of Criminal Procedure] 32(i).” Guidelines Manual\\n§ 6A1.3(b) (emphasis added). Federal Rule of Criminal\\nProcedure 32(i)(4)(A)(ii), in turn, requires the sentencing\\ncourt to “address the defendant personally in order to permit\\nthe defendant to speak or present any information to mitigate\\nthe sentence.” This rule, like its longstanding common law\\npredecessor, affords the defendant the right “to present any\\n\\f12 UNITED STATES V. GREEN\\n\\ninformation in mitigation of punishment.” Green v. United\\nStates, 365 U.S. 301, 304 (1961). There is particular need to\\ngive the defendant an opportunity to speak when a\\nsentencing court must evaluate the defendant’s character and\\ncredibility. In such a circumstance, even the “most\\npersuasive counsel may not be able to speak for a defendant\\nas the defendant might, with halting eloquence, speak for\\nhimself.” Id.\\n\\n A sentencing court is faced with precisely such a\\ncharacter-evaluation situation when assessing whether to\\napply the acceptance-of-responsibility reduction. The\\ncourt’s decision on the reduction turns on whether the\\ndefendant has expressed contrition and whether such\\nexpression is “sincere.” Cortes, 299 F.3d at 1038. A\\ndefendant can hardly demonstrate sincere contrition to the\\ncourt through his allocution if he cannot speak until after the\\nsentencing court has already made up its mind as to whether\\nhe has done so.\\n\\n C\\n\\n On appeal, Green’s counsel did not initially argue that\\nthe district court misconstrued the law by concluding that it\\ncould not first hear from Green before determining whether\\nhe had accepted responsibility.3 Thus, this error is subject\\nto plain error review. 4 See United States v. Depue, 912 F.3d\\n1227, 1233 (9th Cir. 2019) (en banc). Under plain error\\n\\n 3\\n Green raised this basis for vacating the sentence only after\\nreceiving a request from this court for additional briefing on the issue.\\n\\n 4\\n The government argues that Green has waived any challenge to the\\ntiming of the district court’s consideration of the acceptance-of-\\nresponsibility reduction. But it cites no evidence that Green\\naffirmatively waived the claim. Depue, 912 F.3d at 1233.\\n\\f UNITED STATES V. GREEN 13\\n\\nreview, a reviewing court may grant relief where the district\\ncourt erred so long as the error is contrary to the law at the\\ntime of the appeal, affects the defendant’s substantial rights,\\nand seriously affects the fairness, integrity, or public\\nreputation of judicial proceedings. Id. at 1232. We conclude\\nthat each of these conditions is met.\\n\\n First, the district court’s conclusion was undoubtably\\ncontrary to the law. As already discussed, no authority\\nsupports the proposition that a sentencing court cannot first\\nhear from a defendant before determining whether a\\nreduction for acceptance of responsibility is appropriate.\\nAnd, all other relevant considerations—the practice of\\nsentencing courts, the Sentencing Guidelines, and Rule 32(i)\\nof the Federal Rules of Criminal Procedure—indicate that\\ndistrict courts regularly do and regularly should consider a\\ndefendant’s allocution before determining whether to apply\\nthe acceptance-of-responsibility reduction.\\n\\n The district court’s error also affected Green’s\\nsubstantial rights and seriously affected the fairness of the\\njudicial proceedings. In the sentencing context, these two\\nconsiderations will normally merge. An error “affects\\nsubstantial rights if the defendant can ‘demonstrate a\\nreasonable probability that [he] would have received a\\ndifferent sentence if the district court had not erred.’” Id.\\nat 1234 (quoting United States v. Joseph, 716 F.3d 1273,\\n1280 (9th Cir. 2013)) (second alternation in original).\\nSimilarly, an error seriously affects the fairness of a\\nsentencing proceeding if that error “may have increased the\\nlength of a defendant’s sentence.” United States v. Tapia,\\n665 F.3d 1059, 1063 (9th Cir. 2011). In other words, we\\nmust conclude that there is a reasonable probability that\\nGreen could receive a lower sentence on remand once the\\n\\f14 UNITED STATES V. GREEN\\n\\ndistrict court considers his allocution. There is such\\nprobability.\\n\\n To address the reasonable probability of a lower\\nsentence, we first address Green’s contention that the district\\ncourt erred by comparing the circumstances in this case to\\nthe circumstances in United States v. Ginn. When stating its\\nconclusion that Green was not eligible for the acceptance-of-\\nresponsibility reduction, the district court reasoned that Ginn\\ncounsels against granting an acceptance-of-responsibility\\nreduction where a defendant has pleaded guilty to a single\\ncount of possession of firearms, but does not admit to\\npossession of all the firearms alleged in the indictment. If\\nthe district court properly analogized to Ginn, then there\\nwould be little probability that the district court would\\nresolve the issue of Green’s acceptance of responsibility in\\nGreen’s favor on remand.\\n\\n We conclude it was error to rely on Ginn to determine\\nwhether the acceptance-of-responsibility reduction was\\nappropriate. Ginn and Garrido, read together, make a\\ndefendant ineligible for an acceptance-of-responsibility\\nreduction where he has not accepted responsibility for all\\ncounts of which he was convicted, excluding counts that\\ncannot be grouped under the Guidelines. Ginn, 87 F.3d at\\n370; Garrido, 596 F.3d at 619. That rule has little\\napplicability to this case. Here, the government charged\\nGreen with only one count. Three guns were (superfluously)\\ncharged in that single count, and Green pleaded guilty to that\\ncount. Green was not convicted of possessing three guns,\\nand he had no obligation to admit to possessing all three to\\naccept responsibility for the only crime of which he was\\nconvicted. See Vance, 62 F.3d at 1158; see also Guidelines\\nManual § 3E1.1 cmt. n.1(A).\\n\\f UNITED STATES V. GREEN 15\\n\\n As the rule set forth in Ginn and Garrido regarding\\nacceptance-of-responsibility reductions for defendants\\nconvicted of multiple counts is neither binding nor relevant\\nin this case, the district court will on remand have to\\nreconsider whether the acceptance-of-responsibility\\nreduction is appropriate. In making that determination,\\nGreen’s posture concerning the two guns found in the safe\\nmay have some bearing. The two guns in the safe were\\nfound by a preponderance of the evidence to be in Green’s\\npossession and so to be related conduct. See Guidelines\\nManual §§ 1B1.3, 3E1.1 cmt. n.1(A). “A defendant is not\\nrequired to volunteer, or affirmatively admit, relevant\\nconduct beyond the offense of conviction in order to obtain\\na[n acceptance-of-responsibility] reduction.” Id. § 3E1.1\\ncmt. n.1(A). But “a defendant who falsely denies, or\\nfrivolously contests, relevant conduct that the court\\ndetermines to be true has acted in a manner inconsistent with\\nacceptance of responsibility.” Id.; see also id. § 1B1.3.\\n\\n The district court specifically found that Green did not\\nfalsely deny possession of these two guns. But it did not\\ndetermine whether Green “frivolously contest[ed]”\\npossession of these guns. Thus, on remand the district court\\ncan consider whether Green has “frivolously contest[ed]”\\npossession of the two other guns, see United States v.\\nRamos-Medina, 706 F.3d 932, 942 (9th Cir. 2013), or\\nwhether he has otherwise “acted in a manner inconsistent\\nwith acceptance of responsibility.” Guidelines Manual\\n§ 3E1.1 cmt. n.1(A); see also id. § 3E1.1 cmt. n.3. If he has\\nengaged in such inconsistent conduct, the Guidelines direct\\nthat the court should consider whether such inconsistent\\nconduct outweighs the affirmative evidence indicating\\nGreen has accepted responsibility, including his decision to\\nenter “a plea of guilty prior to the commencement of trial”\\n\\f16 UNITED STATES V. GREEN\\n\\nand “admit[] the conduct comprising the offense of\\nconviction.” Id. § 3E1.1 cmt. n.3.\\n\\n Under these standards, the district court could conclude,\\ntaking into account the allocution, that Green meets the\\nacceptance-of-responsibility requirements with regard to\\nrelated conduct. There is at least a reasonable probability\\nthat after allocution, the district court could determine an\\nacceptance-of-responsibility reduction appropriate. And\\nbecause a decision to grant an acceptance-of-responsibility\\nreduction would likely lead to a less severe sentence for\\nGreen, the failure to consider Green’s allocution—coupled\\nwith the district court’s faulty reliance on Ginn—affected\\nGreen’s substantial rights and seriously affected the fairness,\\nintegrity, or public reputation of judicial proceedings.\\n\\n * * *\\n\\n We hold that the sentencing court erred by concluding\\nthat it could not first hear from the defendant before\\ndetermining whether a reduction for acceptance of\\nresponsibility was warranted under the Sentencing\\nGuidelines. We also conclude that this misapprehension was\\nplain error and so vacate the sentence and remand for\\nresentencing.\\n\\n Sentence VACATED and REMANDED for\\nresentencing.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4431729/\", \"author_raw\": \"BERZON, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"A WALLACE TASHIMA"},"judge2":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge3":{"kind":"string","value":"MARSHA S BERZON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4654476/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Jeffrey Green\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-21\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-30227\n Plaintiff-Appellee,\n D.C. No.\n v. 3:16-cr-00063-SLG-1\n\nJEFFREY R. GREEN, OPINION\n Defendant-Appellant.\n\n\n Appeal from the United States District Court\n for the District of Alaska\n Sharon L. Gleason, District Judge, Presiding\n\n Argued and Submitted June 12, 2019\n Anchorage, Alaska\n\n Filed August 21, 2019\n\n Before: A. Wallace Tashima, William A. Fletcher,\n and Marsha S. Berzon, Circuit Judges.\n\n Opinion by Judge Berzon\n\f2 UNITED STATES V. GREEN\n\n SUMMARY *\n\n\n Criminal Law\n\n Vacating a sentence and remanding for resentencing, the\npanel held that the district court erred by concluding that it\ncould not listen to the defendant’s allocution before\ndetermining whether a reduction of acceptance of\nresponsibility was warranted under the Sentencing\nGuidelines, and that this misapprehension was plain error\nthat affected the defendant’s substantial rights and seriously\naffected the fairness of the proceedings.\n\n\n COUNSEL\n\nKrista Hart (argued), Sacramento, California, for Defendant-\nAppellant.\n\nJonas M. Walker (argued), Assistant United States Attorney;\nBrian Schroder, United States Attorney; United States\nAttorney’s Office, Anchorage, Alaska; for Plaintiff-\nAppellee.\n\n\n\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f UNITED STATES V. GREEN 3\n\n OPINION\n\nBERZON, Circuit Judge:\n\n Must a district court decide on a defendant’s eligibility\nfor an acceptance-of-responsibility reduction in his\nGuidelines level before listening to the defendant’s\nallocution? Our answer is “No.”\n\n I\n\n On June 3, 2016, a group of police officers went to\nJeffrey Green’s apartment in Anchorage, Alaska, and\narrested Green on an outstanding warrant. While patting him\ndown, an officer found a loaded revolver in Green’s pocket.\nDuring a later search, the officers found two pistols stored\ninside a safe in a storage closet accessible from the\napartment. Both pistols had been reported stolen.\n\n The government charged Green, who had a long history\nof felony convictions, with a single count of possession of a\nfirearm as a felon in violation of 18 U.S.C. § 922(g)(1). Six\nmonths later, Green pleaded guilty. During his plea\ncolloquy, Green admitted that he possessed the single\nrevolver found in his pocket during the arrest and that he was\na felon. But Green did not admit to all the conduct alleged\nin the single-count indictment. He made no admissions—or\nstatements of any kind—regarding either of the pistols found\nin the safe. 1 The district court found that Green’s admission\nregarding the revolver, coupled with his admission regarding\n\n\n\n 1\n The government offered Green a plea deal contingent on Green\nadmitting possession of all three firearms. Green rejected this offer and\ninstead pleaded guilty without an agreement with the government.\n\f4 UNITED STATES V. GREEN\n\nhis criminal history, provided a sufficient factual basis for\nthe plea under Federal Rule of Criminal Procedure 11(b)(3).\n\n After Green pleaded guilty, the district court directed the\nprobation department to prepare a presentence report. That\nreport concluded that Green should be assessed an offense\nlevel under the Sentencing Guidelines premised on\npossession of a total of three guns, two of which were stolen.\nSee U.S. Sentencing Guidelines Manual § 2K2.1(b)(1),\n(b)(4)(A) (U.S. Sentencing Comm’n 2016). 2 The\npresentence report also concluded that Green was not\nentitled to any reduction for accepting responsibility because\nhe had not admitted possession of the two pistols found in\nthe safe. See id. § 3E1.1(a). Green objected to each of these\nconclusions. He primarily argued that the government\n“ha[d] not proven that the two additional firearms found in\nthe storage closet were in Green’s possession.”\n\n Because Green so objected, the court held an evidentiary\nhearing to determine whether Green possessed those\nfirearms. See id. § 6A1.3 cmt. At this hearing, the\ngovernment introduced a recording of Green speaking to a\nwoman by phone after he was arrested. Green asked the\nwoman, “Did they get my safe?” The woman replied, “I\ndon’t know. There was a locksmith. They had a locksmith\ncome there. Yes. Shane said they got into it.” Green then\nresponded: “Oh, my God.” Based on this audio and other\nevidence introduced by the government the district court\nconcluded that the government had shown by a\n\n 2\n These Guidelines sections provide that “[i]f the offense involved\nthree [to seven] firearms, increase” the offense level by two levels,\nGuidelines Manual § 2K2.1(b)(1), and that “[i]f any firearm [involved in\nthe offense] was stolen, increase [the offense level] by 2 levels,” id.\n§ 2K2.1(b)(4)(A).\n\f UNITED STATES V. GREEN 5\n\npreponderance of the evidence that Green possessed the two\npistols, and that he should thus be assessed the offense level\nfor possession of stolen guns and for possession of three or\nmore guns. The court left open whether Green should be\nawarded a reduction for accepting responsibility pursuant to\nSentencing Guidelines section 3E1.1.\n\n Two weeks later, on November 3, 2017, the district court\nheld a second sentencing hearing. At the outset of this\nsecond hearing, the court entertained argument as to whether\nit should find that Green accepted responsibility under\nsection 3E1.1. During argument, defense counsel told the\ncourt that Green “intends to allocute to this Court.” Counsel\nfurther stated that he thought “the only way [Green will] be\nable to express [the] contrition [required by section 3E1.1]\n. . . is in that allocution.”\n\n After hearing counsel’s argument but before hearing\nGreen’s allocution the court announced its conclusion\nregarding the acceptance-of-responsibility reduction—that\nthe reduction was not appropriate. The sentencing court\nexplained that it reached this conclusion largely because it\nviewed this case as analogous to United States v. Ginn,\n87 F.3d 367 (9th Cir. 1996), which held that an acceptance-\nof-responsibility reduction was appropriate only where a\ndefendant charged with multiple counts had accepted\nresponsibility for all of the “counts of which he is\nconvicted.” Id. at 370; cf. United States v. Garrido, 596 F.3d\n613, 619 (9th Cir. 2010) (holding that to remain eligible for\nthe acceptance-of-responsibility reduction, a defendant need\nnot accept responsibility for counts excluded from grouping\nunder the Guidelines).\n\n The district court recognized that “evaluating the\nacceptance[-of-responsibility reduction] . . . after an\nallocution might be helpful” in some circumstances. But it\n\f6 UNITED STATES V. GREEN\n\nbelieved that under our case law, it could not hear from\nGreen before determining the applicability of the reduction.\nIn the district court’s view, “the Ninth Circuit very clearly\ninstructs district judges to determine the guidelines at the\noutset of sentencing proceeding[s],” and so before hearing\nfrom the defendant.\n\n After explaining its decision regarding the acceptance-\nof-responsibility reduction, the court heard further argument\nfrom counsel on the appropriate sentence under the statutory\nsentencing factors. See 18 U.S.C. § 3553(a). It then\nprovided Green his opportunity to speak. During his\nallocution, Green explained at length that he was “extremely\nsorry” for his actions.\n\n Immediately following the allocution, the court\nsentenced Green to a 108-month term of imprisonment. The\nimposed sentence was within the range of 100 to 120 months\nrecommended by the Guidelines calculation adopted by the\ndistrict court. Had Green received the acceptance-of-\nresponsibility reduction, the Guidelines recommended term\nwould have been between 77 and 96 months. Green appeals\nhis sentence.\n\n II\n\n At issue in this appeal is the district court’s decision not\nto reduce Green’s offense level under section 3E1.1 of the\nSentencing Guidelines based on acceptance of\nresponsibility. We review any factual finding embedded in\nthat determination for clear error. United States v.\nRodriguez, 851 F.3d 931, 949 (9th Cir. 2017). But we\n“review de novo whether the district court misapprehended\nthe law with respect to the acceptance of responsibility\nreduction.” United States v. Cortes, 299 F.3d 1030, 1037\n\f UNITED STATES V. GREEN 7\n\n(9th Cir. 2002); see also United States v. Doe, 778 F.3d 814,\n821 (9th Cir. 2015).\n\n A misapprehension of law occurred here. The district\ncourt believed that it had to determine whether Green had\naccepted responsibility before allowing Green his\nopportunity to allocute. That is not so.\n\n Green did not raise this error in the district court or in his\ninitial briefing, so we review for plain error. See United\nStates v. Sum of $185,336.07 U.S. Currency, 731 F.3d 189,\n195 (2d Cir.2013); United States v. McKinney, 707 F.2d 381,\n383 (9th Cir. 1983). We conclude that this misapprehension\nof law constituted plain error, vacate Green’s sentence, and\nremand for resentencing.\n\n A\n\n Before addressing the merits of this appeal, a brief\nreview of the acceptance-of-responsibility reduction is\nhelpful. Sentencing Guidelines section 3E1.1(a) calls for\nreducing a defendant’s offense level by two points when the\ndistrict court finds that the defendant has “clearly\ndemonstrate[d] acceptance of responsibility for his offense.”\n“The primary goal of the reduction is to reward defendants\nwho are genuinely contrite.” United States v. McKinney,\n15 F.3d 849, 853 (9th Cir. 1994). When a defendant’s\n“statements and conduct ma[k]e it clear that his contrition\n[is] sincere, he [is] entitled to the reduction,” Cortes,\n299 F.3d at 1038, even if that contrition is expressed at the\n“eleventh[]hour,” United States v. Hill, 953 F.2d 452, 461\n(9th Cir. 1991).\n\n The Sentencing Guidelines commentary provides courts\nwith guidance on how to evaluate whether a defendant has\ndemonstrated the requisite contrition. The Guidelines\n\f8 UNITED STATES V. GREEN\n\ncommentary instructs that courts may consider whether the\ndefendant has, for example, “truthfully admitt[ed] the\nconduct comprising the offense(s) of conviction,”\n“voluntar[ily] terminat[ed] or withdraw[n] from criminal\nconduct or associations,” or “voluntar[illy] pa[id] restitution\nprior to adjudication of guilt.” See Guidelines Manual\n§ 3E1.1 cmt. n.1.\n\n The Guidelines and our case law also provide specific\nguidance on how sentencing courts should evaluate the\nacceptance-of-responsibility reduction where, as here, the\ndefendant has entered a plea of guilty. The Guidelines\ncommentary explains: “Entry of a plea of guilty prior to the\ncommencement of trial combined with truthfully admitting\nthe conduct comprising the offense of conviction, and\ntruthfully admitting or not falsely denying any additional\nrelevant conduct for which he is accountable . . . will\nconstitute significant evidence of acceptance of\nresponsibility.” Id. § 3E1.1 cmt. n.3. When a district court\ndetermines that a defendant’s plea was adequately supported\nby the defendant’s admissions, as the district court did here,\nthat determination “compels the inference that [the\ndefendant] had ‘truthfully admitt[ed] the conduct\ncomprising the offense of conviction.’” United States v.\nVance, 62 F.3d 1152, 1158 (9th Cir. 1995) (second alteration\nin original). Although the Guidelines thus suggest that a\nguilty plea supported by truthful admissions by the\ndefendant creates a presumption that the defendant will\nreceive the acceptance-of-responsibility reduction, this\npresumption can be negated by other evidence that suggests\nthe defendant has not in fact accepted responsibility for his\nactions. Guidelines Manual § 3E1.1 cmt. n.3. Such\ninconsistent conduct can include, for example, “falsely\nden[ying], or frivolously contest[ing], relevant conduct that\nthe court determines to be true.” Id. § 3E1.1 cmt. n.1(A).\n\f UNITED STATES V. GREEN 9\n\n B\n\n The district court in this case did not cite any specific\nauthority to support its premise that a sentencing court must\nreach its conclusion regarding acceptance of responsibility\nbefore hearing from the defendant. The government argues\non appeal that Kimbrough v. United States, 552 U.S. 85, 108\n(2007), Gall v. United States, 552 U.S. 38, 49 (2007), and\nUnited States v. Carty, 520 F.3d 984, 991 (9th Cir. 2008) (en\nbanc), preclude a sentencing court from hearing a\ndefendant’s allocution before determining whether an\nacceptance-of-responsibility reduction is appropriate.\n\n None of these cases so limits a sentencing court.\nKimbrough, Gall, and Carty each make clear that sentencing\ncourts “must treat the Guidelines as the ‘starting point and\nthe initial benchmark’” during sentencing. Kimbrough,\n552 U.S. at 108 (quoting Gall, 552 U.S. at 49); see also Rita\nv. United States, 551 U.S. 338, 347–38 (2007). Carty\nexplained that “[a]ll sentencing proceedings are to begin by\ndetermining the applicable Guidelines range.” 520 F.3d at\n991. Likewise, Gall states that “a district court should begin\nall sentencing proceedings by correctly calculating the\napplicable Guidelines range.” 552 U.S. at 49.\n\n The language in these cases, taken out of context, can\nperhaps be read to comport with the district court’s\nunderstanding. In context, however, it is apparent that these\ncases were not addressing the question before us—whether\na district court must begin a sentencing hearing with a\nGuidelines calculation rather than announcing it once (1) all\nnecessary factual information has been presented, (2) all\nrelevant factual issues have been resolved, and (3) all\npertinent argument has been heard.\n\f10 UNITED STATES V. GREEN\n\n The cases explain that when sentencing a defendant, a\ndistrict court must determine the recommended sentencing\nrange under the Sentencing Guidelines before making a\nholistic and individualized determination as to the\nappropriate sentence under those Guidelines and the\n18 U.S.C. § 3553(a) factors. Carty, 520 F.3d at 991. It is in\nthat sense that the ultimate “sentencing proceedings”—that\nis, the determination and announcement of the sentence to\nbe imposed—“are to begin by determining the applicable\nGuidelines range.” Id. None of these cases requires a district\ncourt to make a final calculation as to the appropriate\nGuidelines range at the outset of a sentencing hearing, even\nif the judge lacks the necessary information and legal\nargumentation to make an accurate, well-considered\ndetermination.\n\n Indeed, in this case the district court did conduct an\nevidentiary hearing to determine whether it was appropriate\nto impose additional offense levels for the possession of\nstolen firearms, and the possession of three or more firearms.\nIf Kimbrough, Gall, and Carty set forth the rigid rule the\ngovernment now suggests, that rule would also make\nholding such an evidentiary hearing improper. Thankfully,\nKimbrough, Gall, and Carty do not set forth such a\ncounterproductive requirement. Neither the Guidelines nor\nthe pertinent case law prefers ignorance over appropriate\ninformation collection and considered judicial reflection\nbefore calculating the applicable Guidelines range.\n\n Three additional considerations support our conclusion\nthat a district court can hear from a defendant before\ndetermining whether an acceptance-of-responsibility\nreduction is indicated under the Guidelines.\n\n First, as a matter of practice, sentencing courts\nfrequently consider the defendant’s allocution when\n\f UNITED STATES V. GREEN 11\n\nconsidering whether an acceptance-of-responsibility\nreduction is warranted under the Guidelines. See, e.g.,\nUnited States v. Goodson, 920 F.3d 1209, 1211 (8th Cir.\n2019); United States v. Leasure, 122 F.3d 837, 841 (9th Cir.\n1997); Hill, 953 F.2d at 461, United States v. Watt, 910 F.2d\n587, 589, (9th Cir. 1990); see also United States v. Harvey,\n597 F. App’x 455, 456 (9th Cir. 2015); United States v.\nKamsomphou, 111 F. App’x 937, 940 (9th Cir. 2004);\nUnited States v. Thompson, 49 F. App’x 749, 750 (9th Cir.\n2002).\n\n Second, the Sentencing Guidelines recognize that, in\nsome cases, a “factor important to the sentencing\ndetermination [will be] reasonably in dispute.” Guidelines\nManual § 6A1.3(a). The Guidelines state that, in these\ncircumstances, “the parties shall be given an adequate\nopportunity to present information to the court regarding that\nfactor.” Id. (emphasis added). Thus, the Guidelines\nrecognize that in some circumstances, “determining the\napplicable Guidelines range,” Carty, 520 F.3d at 991, may\nrequire a sentencing judge to evaluate affidavits, hear\ntestimony or hold argument, depending on the nature of the\nfactor in dispute. See Guidelines Manual § 6A1.3(a).\n\n Third, and relatedly, when a court must resolve “disputed\nsentencing factors” as part of its sentencing determination,\nthe Guidelines state that the sentencing court “shall resolve\n[those] disputed . . . factors . . . in accordance with [Federal]\nRule [of Criminal Procedure] 32(i).” Guidelines Manual\n§ 6A1.3(b) (emphasis added). Federal Rule of Criminal\nProcedure 32(i)(4)(A)(ii), in turn, requires the sentencing\ncourt to “address the defendant personally in order to permit\nthe defendant to speak or present any information to mitigate\nthe sentence.” This rule, like its longstanding common law\npredecessor, affords the defendant the right “to present any\n\f12 UNITED STATES V. GREEN\n\ninformation in mitigation of punishment.” Green v. United\nStates, 365 U.S. 301, 304 (1961). There is particular need to\ngive the defendant an opportunity to speak when a\nsentencing court must evaluate the defendant’s character and\ncredibility. In such a circumstance, even the “most\npersuasive counsel may not be able to speak for a defendant\nas the defendant might, with halting eloquence, speak for\nhimself.” Id.\n\n A sentencing court is faced with precisely such a\ncharacter-evaluation situation when assessing whether to\napply the acceptance-of-responsibility reduction. The\ncourt’s decision on the reduction turns on whether the\ndefendant has expressed contrition and whether such\nexpression is “sincere.” Cortes, 299 F.3d at 1038. A\ndefendant can hardly demonstrate sincere contrition to the\ncourt through his allocution if he cannot speak until after the\nsentencing court has already made up its mind as to whether\nhe has done so.\n\n C\n\n On appeal, Green’s counsel did not initially argue that\nthe district court misconstrued the law by concluding that it\ncould not first hear from Green before determining whether\nhe had accepted responsibility.3 Thus, this error is subject\nto plain error review. 4 See United States v. Depue, 912 F.3d\n1227, 1233 (9th Cir. 2019) (en banc). Under plain error\n\n 3\n Green raised this basis for vacating the sentence only after\nreceiving a request from this court for additional briefing on the issue.\n\n 4\n The government argues that Green has waived any challenge to the\ntiming of the district court’s consideration of the acceptance-of-\nresponsibility reduction. But it cites no evidence that Green\naffirmatively waived the claim. Depue, 912 F.3d at 1233.\n\f UNITED STATES V. GREEN 13\n\nreview, a reviewing court may grant relief where the district\ncourt erred so long as the error is contrary to the law at the\ntime of the appeal, affects the defendant’s substantial rights,\nand seriously affects the fairness, integrity, or public\nreputation of judicial proceedings. Id. at 1232. We conclude\nthat each of these conditions is met.\n\n First, the district court’s conclusion was undoubtably\ncontrary to the law. As already discussed, no authority\nsupports the proposition that a sentencing court cannot first\nhear from a defendant before determining whether a\nreduction for acceptance of responsibility is appropriate.\nAnd, all other relevant considerations—the practice of\nsentencing courts, the Sentencing Guidelines, and Rule 32(i)\nof the Federal Rules of Criminal Procedure—indicate that\ndistrict courts regularly do and regularly should consider a\ndefendant’s allocution before determining whether to apply\nthe acceptance-of-responsibility reduction.\n\n The district court’s error also affected Green’s\nsubstantial rights and seriously affected the fairness of the\njudicial proceedings. In the sentencing context, these two\nconsiderations will normally merge. An error “affects\nsubstantial rights if the defendant can ‘demonstrate a\nreasonable probability that [he] would have received a\ndifferent sentence if the district court had not erred.’” Id.\nat 1234 (quoting United States v. Joseph, 716 F.3d 1273,\n1280 (9th Cir. 2013)) (second alternation in original).\nSimilarly, an error seriously affects the fairness of a\nsentencing proceeding if that error “may have increased the\nlength of a defendant’s sentence.” United States v. Tapia,\n665 F.3d 1059, 1063 (9th Cir. 2011). In other words, we\nmust conclude that there is a reasonable probability that\nGreen could receive a lower sentence on remand once the\n\f14 UNITED STATES V. GREEN\n\ndistrict court considers his allocution. There is such\nprobability.\n\n To address the reasonable probability of a lower\nsentence, we first address Green’s contention that the district\ncourt erred by comparing the circumstances in this case to\nthe circumstances in United States v. Ginn. When stating its\nconclusion that Green was not eligible for the acceptance-of-\nresponsibility reduction, the district court reasoned that Ginn\ncounsels against granting an acceptance-of-responsibility\nreduction where a defendant has pleaded guilty to a single\ncount of possession of firearms, but does not admit to\npossession of all the firearms alleged in the indictment. If\nthe district court properly analogized to Ginn, then there\nwould be little probability that the district court would\nresolve the issue of Green’s acceptance of responsibility in\nGreen’s favor on remand.\n\n We conclude it was error to rely on Ginn to determine\nwhether the acceptance-of-responsibility reduction was\nappropriate. Ginn and Garrido, read together, make a\ndefendant ineligible for an acceptance-of-responsibility\nreduction where he has not accepted responsibility for all\ncounts of which he was convicted, excluding counts that\ncannot be grouped under the Guidelines. Ginn, 87 F.3d at\n370; Garrido, 596 F.3d at 619. That rule has little\napplicability to this case. Here, the government charged\nGreen with only one count. Three guns were (superfluously)\ncharged in that single count, and Green pleaded guilty to that\ncount. Green was not convicted of possessing three guns,\nand he had no obligation to admit to possessing all three to\naccept responsibility for the only crime of which he was\nconvicted. See Vance, 62 F.3d at 1158; see also Guidelines\nManual § 3E1.1 cmt. n.1(A).\n\f UNITED STATES V. GREEN 15\n\n As the rule set forth in Ginn and Garrido regarding\nacceptance-of-responsibility reductions for defendants\nconvicted of multiple counts is neither binding nor relevant\nin this case, the district court will on remand have to\nreconsider whether the acceptance-of-responsibility\nreduction is appropriate. In making that determination,\nGreen’s posture concerning the two guns found in the safe\nmay have some bearing. The two guns in the safe were\nfound by a preponderance of the evidence to be in Green’s\npossession and so to be related conduct. See Guidelines\nManual §§ 1B1.3, 3E1.1 cmt. n.1(A). “A defendant is not\nrequired to volunteer, or affirmatively admit, relevant\nconduct beyond the offense of conviction in order to obtain\na[n acceptance-of-responsibility] reduction.” Id. § 3E1.1\ncmt. n.1(A). But “a defendant who falsely denies, or\nfrivolously contests, relevant conduct that the court\ndetermines to be true has acted in a manner inconsistent with\nacceptance of responsibility.” Id.; see also id. § 1B1.3.\n\n The district court specifically found that Green did not\nfalsely deny possession of these two guns. But it did not\ndetermine whether Green “frivolously contest[ed]”\npossession of these guns. Thus, on remand the district court\ncan consider whether Green has “frivolously contest[ed]”\npossession of the two other guns, see United States v.\nRamos-Medina, 706 F.3d 932, 942 (9th Cir. 2013), or\nwhether he has otherwise “acted in a manner inconsistent\nwith acceptance of responsibility.” Guidelines Manual\n§ 3E1.1 cmt. n.1(A); see also id. § 3E1.1 cmt. n.3. If he has\nengaged in such inconsistent conduct, the Guidelines direct\nthat the court should consider whether such inconsistent\nconduct outweighs the affirmative evidence indicating\nGreen has accepted responsibility, including his decision to\nenter “a plea of guilty prior to the commencement of trial”\n\f16 UNITED STATES V. GREEN\n\nand “admit[] the conduct comprising the offense of\nconviction.” Id. § 3E1.1 cmt. n.3.\n\n Under these standards, the district court could conclude,\ntaking into account the allocution, that Green meets the\nacceptance-of-responsibility requirements with regard to\nrelated conduct. There is at least a reasonable probability\nthat after allocution, the district court could determine an\nacceptance-of-responsibility reduction appropriate. And\nbecause a decision to grant an acceptance-of-responsibility\nreduction would likely lead to a less severe sentence for\nGreen, the failure to consider Green’s allocution—coupled\nwith the district court’s faulty reliance on Ginn—affected\nGreen’s substantial rights and seriously affected the fairness,\nintegrity, or public reputation of judicial proceedings.\n\n * * *\n\n We hold that the sentencing court erred by concluding\nthat it could not first hear from the defendant before\ndetermining whether a reduction for acceptance of\nresponsibility was warranted under the Sentencing\nGuidelines. We also conclude that this misapprehension was\nplain error and so vacate the sentence and remand for\nresentencing.\n\n Sentence VACATED and REMANDED for\nresentencing.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Jeffrey Green\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-21\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 17-30227\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:16-cr-00063-SLG-1\\n\\nJEFFREY R. GREEN, OPINION\\n Defendant-Appellant.\\n\\n\\n Appeal from the United States District Court\\n for the District of Alaska\\n Sharon L. Gleason, District Judge, Presiding\\n\\n Argued and Submitted June 12, 2019\\n Anchorage, Alaska\\n\\n Filed August 21, 2019\\n\\n Before: A. Wallace Tashima, William A. Fletcher,\\n and Marsha S. Berzon, Circuit Judges.\\n\\n Opinion by Judge Berzon\\n\\f2 UNITED STATES V. GREEN\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n Vacating a sentence and remanding for resentencing, the\\npanel held that the district court erred by concluding that it\\ncould not listen to the defendant’s allocution before\\ndetermining whether a reduction of acceptance of\\nresponsibility was warranted under the Sentencing\\nGuidelines, and that this misapprehension was plain error\\nthat affected the defendant’s substantial rights and seriously\\naffected the fairness of the proceedings.\\n\\n\\n COUNSEL\\n\\nKrista Hart (argued), Sacramento, California, for Defendant-\\nAppellant.\\n\\nJonas M. Walker (argued), Assistant United States Attorney;\\nBrian Schroder, United States Attorney; United States\\nAttorney’s Office, Anchorage, Alaska; for Plaintiff-\\nAppellee.\\n\\n\\n\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f UNITED STATES V. GREEN 3\\n\\n OPINION\\n\\nBERZON, Circuit Judge:\\n\\n Must a district court decide on a defendant’s eligibility\\nfor an acceptance-of-responsibility reduction in his\\nGuidelines level before listening to the defendant’s\\nallocution? Our answer is “No.”\\n\\n I\\n\\n On June 3, 2016, a group of police officers went to\\nJeffrey Green’s apartment in Anchorage, Alaska, and\\narrested Green on an outstanding warrant. While patting him\\ndown, an officer found a loaded revolver in Green’s pocket.\\nDuring a later search, the officers found two pistols stored\\ninside a safe in a storage closet accessible from the\\napartment. Both pistols had been reported stolen.\\n\\n The government charged Green, who had a long history\\nof felony convictions, with a single count of possession of a\\nfirearm as a felon in violation of 18 U.S.C. § 922(g)(1). Six\\nmonths later, Green pleaded guilty. During his plea\\ncolloquy, Green admitted that he possessed the single\\nrevolver found in his pocket during the arrest and that he was\\na felon. But Green did not admit to all the conduct alleged\\nin the single-count indictment. He made no admissions—or\\nstatements of any kind—regarding either of the pistols found\\nin the safe. 1 The district court found that Green’s admission\\nregarding the revolver, coupled with his admission regarding\\n\\n\\n\\n 1\\n The government offered Green a plea deal contingent on Green\\nadmitting possession of all three firearms. Green rejected this offer and\\ninstead pleaded guilty without an agreement with the government.\\n\\f4 UNITED STATES V. GREEN\\n\\nhis criminal history, provided a sufficient factual basis for\\nthe plea under Federal Rule of Criminal Procedure 11(b)(3).\\n\\n After Green pleaded guilty, the district court directed the\\nprobation department to prepare a presentence report. That\\nreport concluded that Green should be assessed an offense\\nlevel under the Sentencing Guidelines premised on\\npossession of a total of three guns, two of which were stolen.\\nSee U.S. Sentencing Guidelines Manual § 2K2.1(b)(1),\\n(b)(4)(A) (U.S. Sentencing Comm’n 2016). 2 The\\npresentence report also concluded that Green was not\\nentitled to any reduction for accepting responsibility because\\nhe had not admitted possession of the two pistols found in\\nthe safe. See id. § 3E1.1(a). Green objected to each of these\\nconclusions. He primarily argued that the government\\n“ha[d] not proven that the two additional firearms found in\\nthe storage closet were in Green’s possession.”\\n\\n Because Green so objected, the court held an evidentiary\\nhearing to determine whether Green possessed those\\nfirearms. See id. § 6A1.3 cmt. At this hearing, the\\ngovernment introduced a recording of Green speaking to a\\nwoman by phone after he was arrested. Green asked the\\nwoman, “Did they get my safe?” The woman replied, “I\\ndon’t know. There was a locksmith. They had a locksmith\\ncome there. Yes. Shane said they got into it.” Green then\\nresponded: “Oh, my God.” Based on this audio and other\\nevidence introduced by the government the district court\\nconcluded that the government had shown by a\\n\\n 2\\n These Guidelines sections provide that “[i]f the offense involved\\nthree [to seven] firearms, increase” the offense level by two levels,\\nGuidelines Manual § 2K2.1(b)(1), and that “[i]f any firearm [involved in\\nthe offense] was stolen, increase [the offense level] by 2 levels,” id.\\n§ 2K2.1(b)(4)(A).\\n\\f UNITED STATES V. GREEN 5\\n\\npreponderance of the evidence that Green possessed the two\\npistols, and that he should thus be assessed the offense level\\nfor possession of stolen guns and for possession of three or\\nmore guns. The court left open whether Green should be\\nawarded a reduction for accepting responsibility pursuant to\\nSentencing Guidelines section 3E1.1.\\n\\n Two weeks later, on November 3, 2017, the district court\\nheld a second sentencing hearing. At the outset of this\\nsecond hearing, the court entertained argument as to whether\\nit should find that Green accepted responsibility under\\nsection 3E1.1. During argument, defense counsel told the\\ncourt that Green “intends to allocute to this Court.” Counsel\\nfurther stated that he thought “the only way [Green will] be\\nable to express [the] contrition [required by section 3E1.1]\\n. . . is in that allocution.”\\n\\n After hearing counsel’s argument but before hearing\\nGreen’s allocution the court announced its conclusion\\nregarding the acceptance-of-responsibility reduction—that\\nthe reduction was not appropriate. The sentencing court\\nexplained that it reached this conclusion largely because it\\nviewed this case as analogous to United States v. Ginn,\\n87 F.3d 367 (9th Cir. 1996), which held that an acceptance-\\nof-responsibility reduction was appropriate only where a\\ndefendant charged with multiple counts had accepted\\nresponsibility for all of the “counts of which he is\\nconvicted.” Id. at 370; cf. United States v. Garrido, 596 F.3d\\n613, 619 (9th Cir. 2010) (holding that to remain eligible for\\nthe acceptance-of-responsibility reduction, a defendant need\\nnot accept responsibility for counts excluded from grouping\\nunder the Guidelines).\\n\\n The district court recognized that “evaluating the\\nacceptance[-of-responsibility reduction] . . . after an\\nallocution might be helpful” in some circumstances. But it\\n\\f6 UNITED STATES V. GREEN\\n\\nbelieved that under our case law, it could not hear from\\nGreen before determining the applicability of the reduction.\\nIn the district court’s view, “the Ninth Circuit very clearly\\ninstructs district judges to determine the guidelines at the\\noutset of sentencing proceeding[s],” and so before hearing\\nfrom the defendant.\\n\\n After explaining its decision regarding the acceptance-\\nof-responsibility reduction, the court heard further argument\\nfrom counsel on the appropriate sentence under the statutory\\nsentencing factors. See 18 U.S.C. § 3553(a). It then\\nprovided Green his opportunity to speak. During his\\nallocution, Green explained at length that he was “extremely\\nsorry” for his actions.\\n\\n Immediately following the allocution, the court\\nsentenced Green to a 108-month term of imprisonment. The\\nimposed sentence was within the range of 100 to 120 months\\nrecommended by the Guidelines calculation adopted by the\\ndistrict court. Had Green received the acceptance-of-\\nresponsibility reduction, the Guidelines recommended term\\nwould have been between 77 and 96 months. Green appeals\\nhis sentence.\\n\\n II\\n\\n At issue in this appeal is the district court’s decision not\\nto reduce Green’s offense level under section 3E1.1 of the\\nSentencing Guidelines based on acceptance of\\nresponsibility. We review any factual finding embedded in\\nthat determination for clear error. United States v.\\nRodriguez, 851 F.3d 931, 949 (9th Cir. 2017). But we\\n“review de novo whether the district court misapprehended\\nthe law with respect to the acceptance of responsibility\\nreduction.” United States v. Cortes, 299 F.3d 1030, 1037\\n\\f UNITED STATES V. GREEN 7\\n\\n(9th Cir. 2002); see also United States v. Doe, 778 F.3d 814,\\n821 (9th Cir. 2015).\\n\\n A misapprehension of law occurred here. The district\\ncourt believed that it had to determine whether Green had\\naccepted responsibility before allowing Green his\\nopportunity to allocute. That is not so.\\n\\n Green did not raise this error in the district court or in his\\ninitial briefing, so we review for plain error. See United\\nStates v. Sum of $185,336.07 U.S. Currency, 731 F.3d 189,\\n195 (2d Cir.2013); United States v. McKinney, 707 F.2d 381,\\n383 (9th Cir. 1983). We conclude that this misapprehension\\nof law constituted plain error, vacate Green’s sentence, and\\nremand for resentencing.\\n\\n A\\n\\n Before addressing the merits of this appeal, a brief\\nreview of the acceptance-of-responsibility reduction is\\nhelpful. Sentencing Guidelines section 3E1.1(a) calls for\\nreducing a defendant’s offense level by two points when the\\ndistrict court finds that the defendant has “clearly\\ndemonstrate[d] acceptance of responsibility for his offense.”\\n“The primary goal of the reduction is to reward defendants\\nwho are genuinely contrite.” United States v. McKinney,\\n15 F.3d 849, 853 (9th Cir. 1994). When a defendant’s\\n“statements and conduct ma[k]e it clear that his contrition\\n[is] sincere, he [is] entitled to the reduction,” Cortes,\\n299 F.3d at 1038, even if that contrition is expressed at the\\n“eleventh[]hour,” United States v. Hill, 953 F.2d 452, 461\\n(9th Cir. 1991).\\n\\n The Sentencing Guidelines commentary provides courts\\nwith guidance on how to evaluate whether a defendant has\\ndemonstrated the requisite contrition. The Guidelines\\n\\f8 UNITED STATES V. GREEN\\n\\ncommentary instructs that courts may consider whether the\\ndefendant has, for example, “truthfully admitt[ed] the\\nconduct comprising the offense(s) of conviction,”\\n“voluntar[ily] terminat[ed] or withdraw[n] from criminal\\nconduct or associations,” or “voluntar[illy] pa[id] restitution\\nprior to adjudication of guilt.” See Guidelines Manual\\n§ 3E1.1 cmt. n.1.\\n\\n The Guidelines and our case law also provide specific\\nguidance on how sentencing courts should evaluate the\\nacceptance-of-responsibility reduction where, as here, the\\ndefendant has entered a plea of guilty. The Guidelines\\ncommentary explains: “Entry of a plea of guilty prior to the\\ncommencement of trial combined with truthfully admitting\\nthe conduct comprising the offense of conviction, and\\ntruthfully admitting or not falsely denying any additional\\nrelevant conduct for which he is accountable . . . will\\nconstitute significant evidence of acceptance of\\nresponsibility.” Id. § 3E1.1 cmt. n.3. When a district court\\ndetermines that a defendant’s plea was adequately supported\\nby the defendant’s admissions, as the district court did here,\\nthat determination “compels the inference that [the\\ndefendant] had ‘truthfully admitt[ed] the conduct\\ncomprising the offense of conviction.’” United States v.\\nVance, 62 F.3d 1152, 1158 (9th Cir. 1995) (second alteration\\nin original). Although the Guidelines thus suggest that a\\nguilty plea supported by truthful admissions by the\\ndefendant creates a presumption that the defendant will\\nreceive the acceptance-of-responsibility reduction, this\\npresumption can be negated by other evidence that suggests\\nthe defendant has not in fact accepted responsibility for his\\nactions. Guidelines Manual § 3E1.1 cmt. n.3. Such\\ninconsistent conduct can include, for example, “falsely\\nden[ying], or frivolously contest[ing], relevant conduct that\\nthe court determines to be true.” Id. § 3E1.1 cmt. n.1(A).\\n\\f UNITED STATES V. GREEN 9\\n\\n B\\n\\n The district court in this case did not cite any specific\\nauthority to support its premise that a sentencing court must\\nreach its conclusion regarding acceptance of responsibility\\nbefore hearing from the defendant. The government argues\\non appeal that Kimbrough v. United States, 552 U.S. 85, 108\\n(2007), Gall v. United States, 552 U.S. 38, 49 (2007), and\\nUnited States v. Carty, 520 F.3d 984, 991 (9th Cir. 2008) (en\\nbanc), preclude a sentencing court from hearing a\\ndefendant’s allocution before determining whether an\\nacceptance-of-responsibility reduction is appropriate.\\n\\n None of these cases so limits a sentencing court.\\nKimbrough, Gall, and Carty each make clear that sentencing\\ncourts “must treat the Guidelines as the ‘starting point and\\nthe initial benchmark’” during sentencing. Kimbrough,\\n552 U.S. at 108 (quoting Gall, 552 U.S. at 49); see also Rita\\nv. United States, 551 U.S. 338, 347–38 (2007). Carty\\nexplained that “[a]ll sentencing proceedings are to begin by\\ndetermining the applicable Guidelines range.” 520 F.3d at\\n991. Likewise, Gall states that “a district court should begin\\nall sentencing proceedings by correctly calculating the\\napplicable Guidelines range.” 552 U.S. at 49.\\n\\n The language in these cases, taken out of context, can\\nperhaps be read to comport with the district court’s\\nunderstanding. In context, however, it is apparent that these\\ncases were not addressing the question before us—whether\\na district court must begin a sentencing hearing with a\\nGuidelines calculation rather than announcing it once (1) all\\nnecessary factual information has been presented, (2) all\\nrelevant factual issues have been resolved, and (3) all\\npertinent argument has been heard.\\n\\f10 UNITED STATES V. GREEN\\n\\n The cases explain that when sentencing a defendant, a\\ndistrict court must determine the recommended sentencing\\nrange under the Sentencing Guidelines before making a\\nholistic and individualized determination as to the\\nappropriate sentence under those Guidelines and the\\n18 U.S.C. § 3553(a) factors. Carty, 520 F.3d at 991. It is in\\nthat sense that the ultimate “sentencing proceedings”—that\\nis, the determination and announcement of the sentence to\\nbe imposed—“are to begin by determining the applicable\\nGuidelines range.” Id. None of these cases requires a district\\ncourt to make a final calculation as to the appropriate\\nGuidelines range at the outset of a sentencing hearing, even\\nif the judge lacks the necessary information and legal\\nargumentation to make an accurate, well-considered\\ndetermination.\\n\\n Indeed, in this case the district court did conduct an\\nevidentiary hearing to determine whether it was appropriate\\nto impose additional offense levels for the possession of\\nstolen firearms, and the possession of three or more firearms.\\nIf Kimbrough, Gall, and Carty set forth the rigid rule the\\ngovernment now suggests, that rule would also make\\nholding such an evidentiary hearing improper. Thankfully,\\nKimbrough, Gall, and Carty do not set forth such a\\ncounterproductive requirement. Neither the Guidelines nor\\nthe pertinent case law prefers ignorance over appropriate\\ninformation collection and considered judicial reflection\\nbefore calculating the applicable Guidelines range.\\n\\n Three additional considerations support our conclusion\\nthat a district court can hear from a defendant before\\ndetermining whether an acceptance-of-responsibility\\nreduction is indicated under the Guidelines.\\n\\n First, as a matter of practice, sentencing courts\\nfrequently consider the defendant’s allocution when\\n\\f UNITED STATES V. GREEN 11\\n\\nconsidering whether an acceptance-of-responsibility\\nreduction is warranted under the Guidelines. See, e.g.,\\nUnited States v. Goodson, 920 F.3d 1209, 1211 (8th Cir.\\n2019); United States v. Leasure, 122 F.3d 837, 841 (9th Cir.\\n1997); Hill, 953 F.2d at 461, United States v. Watt, 910 F.2d\\n587, 589, (9th Cir. 1990); see also United States v. Harvey,\\n597 F. App’x 455, 456 (9th Cir. 2015); United States v.\\nKamsomphou, 111 F. App’x 937, 940 (9th Cir. 2004);\\nUnited States v. Thompson, 49 F. App’x 749, 750 (9th Cir.\\n2002).\\n\\n Second, the Sentencing Guidelines recognize that, in\\nsome cases, a “factor important to the sentencing\\ndetermination [will be] reasonably in dispute.” Guidelines\\nManual § 6A1.3(a). The Guidelines state that, in these\\ncircumstances, “the parties shall be given an adequate\\nopportunity to present information to the court regarding that\\nfactor.” Id. (emphasis added). Thus, the Guidelines\\nrecognize that in some circumstances, “determining the\\napplicable Guidelines range,” Carty, 520 F.3d at 991, may\\nrequire a sentencing judge to evaluate affidavits, hear\\ntestimony or hold argument, depending on the nature of the\\nfactor in dispute. See Guidelines Manual § 6A1.3(a).\\n\\n Third, and relatedly, when a court must resolve “disputed\\nsentencing factors” as part of its sentencing determination,\\nthe Guidelines state that the sentencing court “shall resolve\\n[those] disputed . . . factors . . . in accordance with [Federal]\\nRule [of Criminal Procedure] 32(i).” Guidelines Manual\\n§ 6A1.3(b) (emphasis added). Federal Rule of Criminal\\nProcedure 32(i)(4)(A)(ii), in turn, requires the sentencing\\ncourt to “address the defendant personally in order to permit\\nthe defendant to speak or present any information to mitigate\\nthe sentence.” This rule, like its longstanding common law\\npredecessor, affords the defendant the right “to present any\\n\\f12 UNITED STATES V. GREEN\\n\\ninformation in mitigation of punishment.” Green v. United\\nStates, 365 U.S. 301, 304 (1961). There is particular need to\\ngive the defendant an opportunity to speak when a\\nsentencing court must evaluate the defendant’s character and\\ncredibility. In such a circumstance, even the “most\\npersuasive counsel may not be able to speak for a defendant\\nas the defendant might, with halting eloquence, speak for\\nhimself.” Id.\\n\\n A sentencing court is faced with precisely such a\\ncharacter-evaluation situation when assessing whether to\\napply the acceptance-of-responsibility reduction. The\\ncourt’s decision on the reduction turns on whether the\\ndefendant has expressed contrition and whether such\\nexpression is “sincere.” Cortes, 299 F.3d at 1038. A\\ndefendant can hardly demonstrate sincere contrition to the\\ncourt through his allocution if he cannot speak until after the\\nsentencing court has already made up its mind as to whether\\nhe has done so.\\n\\n C\\n\\n On appeal, Green’s counsel did not initially argue that\\nthe district court misconstrued the law by concluding that it\\ncould not first hear from Green before determining whether\\nhe had accepted responsibility.3 Thus, this error is subject\\nto plain error review. 4 See United States v. Depue, 912 F.3d\\n1227, 1233 (9th Cir. 2019) (en banc). Under plain error\\n\\n 3\\n Green raised this basis for vacating the sentence only after\\nreceiving a request from this court for additional briefing on the issue.\\n\\n 4\\n The government argues that Green has waived any challenge to the\\ntiming of the district court’s consideration of the acceptance-of-\\nresponsibility reduction. But it cites no evidence that Green\\naffirmatively waived the claim. Depue, 912 F.3d at 1233.\\n\\f UNITED STATES V. GREEN 13\\n\\nreview, a reviewing court may grant relief where the district\\ncourt erred so long as the error is contrary to the law at the\\ntime of the appeal, affects the defendant’s substantial rights,\\nand seriously affects the fairness, integrity, or public\\nreputation of judicial proceedings. Id. at 1232. We conclude\\nthat each of these conditions is met.\\n\\n First, the district court’s conclusion was undoubtably\\ncontrary to the law. As already discussed, no authority\\nsupports the proposition that a sentencing court cannot first\\nhear from a defendant before determining whether a\\nreduction for acceptance of responsibility is appropriate.\\nAnd, all other relevant considerations—the practice of\\nsentencing courts, the Sentencing Guidelines, and Rule 32(i)\\nof the Federal Rules of Criminal Procedure—indicate that\\ndistrict courts regularly do and regularly should consider a\\ndefendant’s allocution before determining whether to apply\\nthe acceptance-of-responsibility reduction.\\n\\n The district court’s error also affected Green’s\\nsubstantial rights and seriously affected the fairness of the\\njudicial proceedings. In the sentencing context, these two\\nconsiderations will normally merge. An error “affects\\nsubstantial rights if the defendant can ‘demonstrate a\\nreasonable probability that [he] would have received a\\ndifferent sentence if the district court had not erred.’” Id.\\nat 1234 (quoting United States v. Joseph, 716 F.3d 1273,\\n1280 (9th Cir. 2013)) (second alternation in original).\\nSimilarly, an error seriously affects the fairness of a\\nsentencing proceeding if that error “may have increased the\\nlength of a defendant’s sentence.” United States v. Tapia,\\n665 F.3d 1059, 1063 (9th Cir. 2011). In other words, we\\nmust conclude that there is a reasonable probability that\\nGreen could receive a lower sentence on remand once the\\n\\f14 UNITED STATES V. GREEN\\n\\ndistrict court considers his allocution. There is such\\nprobability.\\n\\n To address the reasonable probability of a lower\\nsentence, we first address Green’s contention that the district\\ncourt erred by comparing the circumstances in this case to\\nthe circumstances in United States v. Ginn. When stating its\\nconclusion that Green was not eligible for the acceptance-of-\\nresponsibility reduction, the district court reasoned that Ginn\\ncounsels against granting an acceptance-of-responsibility\\nreduction where a defendant has pleaded guilty to a single\\ncount of possession of firearms, but does not admit to\\npossession of all the firearms alleged in the indictment. If\\nthe district court properly analogized to Ginn, then there\\nwould be little probability that the district court would\\nresolve the issue of Green’s acceptance of responsibility in\\nGreen’s favor on remand.\\n\\n We conclude it was error to rely on Ginn to determine\\nwhether the acceptance-of-responsibility reduction was\\nappropriate. Ginn and Garrido, read together, make a\\ndefendant ineligible for an acceptance-of-responsibility\\nreduction where he has not accepted responsibility for all\\ncounts of which he was convicted, excluding counts that\\ncannot be grouped under the Guidelines. Ginn, 87 F.3d at\\n370; Garrido, 596 F.3d at 619. That rule has little\\napplicability to this case. Here, the government charged\\nGreen with only one count. Three guns were (superfluously)\\ncharged in that single count, and Green pleaded guilty to that\\ncount. Green was not convicted of possessing three guns,\\nand he had no obligation to admit to possessing all three to\\naccept responsibility for the only crime of which he was\\nconvicted. See Vance, 62 F.3d at 1158; see also Guidelines\\nManual § 3E1.1 cmt. n.1(A).\\n\\f UNITED STATES V. GREEN 15\\n\\n As the rule set forth in Ginn and Garrido regarding\\nacceptance-of-responsibility reductions for defendants\\nconvicted of multiple counts is neither binding nor relevant\\nin this case, the district court will on remand have to\\nreconsider whether the acceptance-of-responsibility\\nreduction is appropriate. In making that determination,\\nGreen’s posture concerning the two guns found in the safe\\nmay have some bearing. The two guns in the safe were\\nfound by a preponderance of the evidence to be in Green’s\\npossession and so to be related conduct. See Guidelines\\nManual §§ 1B1.3, 3E1.1 cmt. n.1(A). “A defendant is not\\nrequired to volunteer, or affirmatively admit, relevant\\nconduct beyond the offense of conviction in order to obtain\\na[n acceptance-of-responsibility] reduction.” Id. § 3E1.1\\ncmt. n.1(A). But “a defendant who falsely denies, or\\nfrivolously contests, relevant conduct that the court\\ndetermines to be true has acted in a manner inconsistent with\\nacceptance of responsibility.” Id.; see also id. § 1B1.3.\\n\\n The district court specifically found that Green did not\\nfalsely deny possession of these two guns. But it did not\\ndetermine whether Green “frivolously contest[ed]”\\npossession of these guns. Thus, on remand the district court\\ncan consider whether Green has “frivolously contest[ed]”\\npossession of the two other guns, see United States v.\\nRamos-Medina, 706 F.3d 932, 942 (9th Cir. 2013), or\\nwhether he has otherwise “acted in a manner inconsistent\\nwith acceptance of responsibility.” Guidelines Manual\\n§ 3E1.1 cmt. n.1(A); see also id. § 3E1.1 cmt. n.3. If he has\\nengaged in such inconsistent conduct, the Guidelines direct\\nthat the court should consider whether such inconsistent\\nconduct outweighs the affirmative evidence indicating\\nGreen has accepted responsibility, including his decision to\\nenter “a plea of guilty prior to the commencement of trial”\\n\\f16 UNITED STATES V. GREEN\\n\\nand “admit[] the conduct comprising the offense of\\nconviction.” Id. § 3E1.1 cmt. n.3.\\n\\n Under these standards, the district court could conclude,\\ntaking into account the allocution, that Green meets the\\nacceptance-of-responsibility requirements with regard to\\nrelated conduct. There is at least a reasonable probability\\nthat after allocution, the district court could determine an\\nacceptance-of-responsibility reduction appropriate. And\\nbecause a decision to grant an acceptance-of-responsibility\\nreduction would likely lead to a less severe sentence for\\nGreen, the failure to consider Green’s allocution—coupled\\nwith the district court’s faulty reliance on Ginn—affected\\nGreen’s substantial rights and seriously affected the fairness,\\nintegrity, or public reputation of judicial proceedings.\\n\\n * * *\\n\\n We hold that the sentencing court erred by concluding\\nthat it could not first hear from the defendant before\\ndetermining whether a reduction for acceptance of\\nresponsibility was warranted under the Sentencing\\nGuidelines. We also conclude that this misapprehension was\\nplain error and so vacate the sentence and remand for\\nresentencing.\\n\\n Sentence VACATED and REMANDED for\\nresentencing.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4088,"cells":{"id":{"kind":"number","value":4654477,"string":"4,654,477"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Tommie Harris v. K. Harris"},"decision_date":{"kind":"string","value":"2019-08-21"},"docket_number":{"kind":"string","value":"16-55083"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Ronald M. Gould and Jacqueline H. Nguyen, Circuit Judges, and Algenon L. Marbley, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"NGUYEN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n TOMMIE LEE HARRIS, No. 16-55083\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:15-cv-03104-\\n ODW-E\\n K. HARRIS, Correctional Officer,\\n individual and official capacity,\\n Defendant-Appellee. OPINION\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Otis D. Wright II, District Judge, Presiding\\n\\n Argued and Submitted February 6, 2019\\n Pasadena, California\\n\\n Filed August 21, 2019\\n\\n Before: Ronald M. Gould and Jacqueline H. Nguyen,\\n Circuit Judges, and Algenon L. Marbley, * District Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n\\n\\n *\\n The Honorable Algenon L. Marbley, United States District Judge\\nfor the Southern District of Ohio, sitting by designation.\\n\\f2 HARRIS V. HARRIS\\n\\n SUMMARY **\\n\\n\\n Prisoner Civil Rights\\n\\n The panel reversed the district court’s revocation of a\\nstate prisoner’s in forma pauperis status on the ground that\\nhe had three prior strikes under the Prison Litigation Reform\\nAct, 28 U.S.C. § 1915(g), and remanded.\\n\\n One of plaintiff’s prior cases was dismissed because,\\nafter concluding that he failed to state a federal claim, the\\ndistrict court declined to exercise supplemental jurisdiction\\nover the remaining state claims. Another was dismissed\\nbecause plaintiff failed to serve one defendant, and several\\nother defendants enjoyed quasi-judicial immunity.\\n\\n The panel held that because the prior cases were not\\ndismissed on grounds enumerated in § 1915(g), they did not\\nqualify as strikes. Following the D.C. Circuit’s decision in\\nFourstar v. Garden City Grp., Inc., 875 F.3d 1147, 1152\\n(D.C. Cir. 2017), the panel first held that a dismissal based\\non a district court’s decision not to exercise supplemental\\njurisdiction is not an enumerated ground under § 1915(g).\\nThe panel further held that dismissal due to a failure to serve\\nis plainly not a dismissal on the ground that the suit was\\nfrivolous, malicious, or failed to state a claim. It is therefore\\nnot a strike under § 1915(g). Finally, the panel held that the\\nlanguage and structure of the Prison Litigation Reform Act\\nmake clear that immunity-based dismissals generally do not\\nfall within § 1915(g).\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f HARRIS V. HARRIS 3\\n\\n COUNSEL\\n\\nDaniel A. Arellano (argued), Ballard Spahr LLP, Phoenix,\\nArizona, for Plaintiff-Appellant.\\n\\nTodd Grabarsky (argued), Deputy Attorney General;\\nThomas S. Patterson and Misha D. Igra, Supervising Deputy\\nAttorneys General; Monica N. Anderson, Senior Assistant\\nAttorney General; Xavier Becerra, Attorney General; Office\\nof the Attorney General, Los Angeles, California; for\\nDefendant-Appellee.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Tommie Lee Harris, a state prisoner, appeals the district\\ncourt’s decision revoking his in forma pauperis (“IFP”)\\nstatus on the ground that he had three prior strikes under the\\nPrison Litigation Reform Act (“PLRA”), 28 U.S.C.\\n§ 1915(g). One of Harris’s prior cases was dismissed\\nbecause, after concluding that he failed to state a federal\\nclaim, the district court declined to exercise supplemental\\njurisdiction over the remaining state claims. Another was\\ndismissed because Harris failed to serve one defendant, and\\nseveral other defendants enjoyed quasi-judicial immunity.\\nWe hold that because these cases were not dismissed on\\ngrounds enumerated in § 1915(g), they do not qualify as\\nstrikes. We therefore reverse and remand.\\n\\n I.\\n\\n Harris filed the current lawsuit against a correctional\\nofficer under 42 U.S.C. § 1983 for use of excessive force.\\nHarris also filed for IFP status. The district court found that\\n\\f4 HARRIS V. HARRIS\\n\\nHarris had already accrued three strikes and revoked his IFP\\nstatus. Harris appeals. He admits that he has two strikes 1\\nunder the PLRA but argues that two of his other prior\\nlawsuits are not strikes, Harris v. Bick, No. 2:98-cv-01197-\\nLKK-DAD (E.D. Cal. Nov. 17, 1998) and Harris v. Nielsen,\\nNo. 2:98-mc-00225-WBS-GGH (PC) (E.D. Cal. Apr. 27,\\n2001).\\n\\nA. Harris v. Bick\\n\\n In Harris v. Bick, Harris sued various prison medical\\nprofessionals, including Dr. Bick, alleging he received\\ninadequate medical care in violation of state law and the\\nEighth Amendment. After Harris attempted to voluntarily\\ndismiss Dr. Bick, the district court found that the allegations\\nagainst the remaining defendants failed to state an Eighth\\nAmendment claim, and “in the absence of a cognizable\\nfederal claim, the court [declined] to exercise supplemental\\njurisdiction over [Harris’s] state contract and debtor-creditor\\nclaims.” On appeal, we affirmed the district court’s\\ndismissal on the Eighth Amendment claims and its decision\\nnot to exercise supplemental jurisdiction.\\n\\nB. Harris v. Nielsen\\n\\n In Harris v. Nielsen, Harris’s claims against various\\nmedical professionals arose out of their participation in\\npreparing mental health reports submitted in connection with\\nhis parole revocation hearings. Harris named defendants E.\\nTitus, J. Choy, G. Phelps, J. Karuzas, and C. Carter. The\\ndistrict court dismissed one defendant, Carter, because the\\n 1\\n Harris concedes that Harris v. Geraghty, No. 98-CV-861-GEB-\\nJFM (E.D. Cal. May 25, 1999) (dismissed for failure to state a claim),\\nand Harris v. Nielsen, No. 01-15006 (9th Cir. Jun. 19, 2001) (appeal\\ndismissed as frivolous), count as strikes under the PLRA. We agree.\\n\\f HARRIS V. HARRIS 5\\n\\nUnited States Marshal was unable to serve him despite\\nattempts at two workplaces. The court also dismissed three\\ndefendants, Titus, Choy, and Phelps, because, as court-\\nappointed psychologists, psychiatrists, and counselors at the\\nparole hearings, they were protected by quasi-judicial\\nimmunity. The district court granted defendant Karuzas’s\\nmotion for judgment on the pleadings for the same reason.\\n\\n The question before us is whether these two suits, Bick\\nand Nielsen, qualify as strikes under § 1915(g). 2\\n\\n II.\\n\\n We interpret § 1915(g) de novo. Andrews v. King,\\n398 F.3d 1113, 1118 (9th Cir. 2005). The denial of IFP\\nstatus is appealable as a final judgment. Id. We have\\njurisdiction under 28 U.S.C. § 1291.\\n\\n III.\\n\\nA. The PLRA’s Section 1915(g)\\n\\n Litigants who qualify for IFP status are excused from\\nprepaying court fees and costs. The PLRA’s “three strikes”\\nprovision, designed to discourage vexatious and voluminous\\nprisoner litigation, bars a prisoner from bringing a civil\\naction or an appeal IFP if the prisoner has three prior actions\\nthat were “dismissed on the grounds that it is frivolous,\\nmalicious, or fails to state a claim upon which relief may be\\ngranted, unless the prisoner is under imminent danger of\\nserious physical injury.” 28 U.S.C. § 1915(g) (emphasis\\nadded).\\n\\n 2\\n We GRANT the pending motions for judicial notice (Docket Nos.\\n21, 37, and 47).\\n\\f6 HARRIS V. HARRIS\\n\\n “[I]n a statutory construction case, analysis must begin\\nwith the language of the statute itself; when the statute is\\nclear, judicial inquiry into its meaning, in all but the most\\nextraordinary circumstance, is finished.” Talamantes v.\\nLeyva, 575 F.3d 1021, 1023 (9th Cir. 2009) (alterations\\nomitted) (quoting United States v. Carter, 421 F.3d 909, 911\\n(9th Cir. 2005)). Here, the statutory language is clear—if a\\ncase was not dismissed on one of the specific enumerated\\ngrounds, it does not count as a strike under § 1915(g).\\n\\n A defendant challenging a plaintiff’s IFP status bears the\\ninitial burden of showing through documentary evidence\\nthat a plaintiff had three prior strikes. King, 398 F.3d\\nat 1118–20. If a defendant presents a prima facie case, then\\n“the burden shifts to the plaintiff to persuade the court that\\n§ 1915(g) does not apply.” Id. at 1116. In determining\\nwhether a prior dismissal counts as a strike, “we should look\\nto the substance of the dismissed lawsuit,” and not to how\\nthe district court labelled or styled the dismissal. El-Shaddai\\nv. Zamora, 833 F.3d 1036, 1047 (9th Cir. 2016) (citing King,\\n398 F.3d at 1122 n.12) (finding that out of eleven prior suits,\\nonly one counted as a strike); see also Knapp v. Hogan, C.O.,\\n738 F.3d 1106, 1109 (9th Cir. 2013) (stating that we look to\\n“the dismissing court’s action and the reasons underlying\\nit . . . . [T]he procedural mechanism or Rule by which the\\ndismissal is accomplished, while informative, is not\\ndispositive.” (internal citation omitted)).\\n\\n We must also evaluate whether all the claims in a given\\nsuit satisfy the enumerated grounds for strikes, and partial\\ndismissals of even one claim for a non-qualifying reason will\\nsave an entire case from constituting a strike. See\\nWashington v. L.A. Cty. Sheriff’s Dep’t, 833 F.3d 1048, 1057\\n(9th Cir. 2016). In other words, “[w]hen we are presented\\nwith multiple claims within a single action, we assess a\\n\\f HARRIS V. HARRIS 7\\n\\nPLRA strike only when the ‘case as a whole’ is dismissed\\nfor a qualifying reason.” Id. (quoting Andrews v. Cervantes,\\n493 F.3d 1047, 1054 (9th Cir. 2007), as amended (July 5,\\n2007)).\\n\\nB. A Dismissal Based on the Court’s Refusal to Exercise\\n Supplemental Jurisdiction Does Not Count as a\\n Strike\\n\\n Harris argues that the dismissal of Harris v. Bick was due\\nto the district court’s refusal to exercise supplemental\\njurisdiction over his state-law claims after dismissal of his\\nEighth Amendment claim, and thus this case does not qualify\\nas a strike. 3 We agree. Dismissal based on a district court’s\\ndecision not to exercise supplemental jurisdiction is not an\\nenumerated ground under § 1915(g). As then-Judge\\nKavanaugh explained, “[w]hen a district court has declined\\nto exercise supplemental jurisdiction over state-law claims,\\nthe court has not dismissed the state-law claims for failure to\\nstate a claim, nor has the court dismissed the state-law claims\\nas frivolous or malicious.” Fourstar v. Garden City Grp.,\\nInc., 875 F.3d 1147, 1152 (D.C. Cir. 2017). Because the\\nlanguage of § 1915(g) is plain and unambiguous, a dismissal\\non a ground other than frivolousness, maliciousness, or\\nfailure to state a claim will not qualify as a strike. See id.\\nat 1151–52.\\n\\n Defendant argues that policy considerations counsel in\\nfavor of counting Bick as a strike. Otherwise, Defendant\\nurges, a prisoner can easily “strike-proof” a meritless lawsuit\\n\\n 3\\n Harris also argues that his attempt to voluntarily dismiss Dr. Bick\\nis an independent ground for not counting this case as a strike. The\\nparties dispute whether his attempt was sufficient but, in any event, we\\nneed not decide this issue.\\n\\f8 HARRIS V. HARRIS\\n\\nby adding state-law claims, knowing that federal courts are\\nunlikely to expend limited resources to exercise\\nsupplemental jurisdiction over such claims in the absence of\\na cognizable federal claim. Of course, courts are free to\\nreach state-law claims and dismiss them on an enumerated\\nstrike ground. But even if the policy concerns are warranted,\\nwe must still strictly construe the plain language of the\\nstatute. “It is not a judge’s job to add to or otherwise re-mold\\nstatutory text to try to meet a statute’s perceived policy\\nobjectives. Instead, we must apply the statute as written.”\\nFourstar, 875 F.3d at 1152. We follow the D.C. Circuit and\\nhold that a dismissal due to the district court’s decision not\\nto exercise supplemental jurisdiction over state-law claims\\ndoes not qualify the case as a strike under the PLRA.\\n\\nC. Neither a Dismissal Due to a Failure to Serve Nor a\\n Dismissal Based on Quasi-Judicial Immunity\\n Qualifies as a Strike\\n\\n Harris argues that Harris v. Nielsen does not qualify as a\\nstrike. The district court dismissed one defendant, Carter,\\nbecause the United States Marshal was unable to serve him\\ndespite multiple attempts at two workplaces. A dismissal\\ndue to a failure to serve is plainly not a dismissal on the\\nground that the suit was frivolous, malicious, or failed to\\nstate a claim. It is therefore not a strike under § 1915(g).\\nDefendant resists a plain reading of the statute and argues\\nthat a dismissal of one defendant for failure to serve should\\nnot prevent a case from qualifying as a strike where other\\nclaims were dismissed for frivolousness or failure to state a\\nclaim. But we evaluate the “case as a whole” and dismissal\\nof even one claim for an unenumerated reason saves the\\nentire case from counting as a strike. Washington, 833 F.3d\\nat 1057. Therefore, Nielsen does not qualify as a strike\\nbecause one defendant was dismissed for failure to serve.\\n\\f HARRIS V. HARRIS 9\\n\\n Nielsen does not qualify as a strike for an additional,\\nindependent reason. The district court dismissed the\\nremaining defendants because it found them entitled to\\nquasi-judicial immunity. Defendant argues that quasi-\\njudicial immunity falls within the enumerated strike ground\\nof failure to state a claim. We have not addressed this issue,\\nbut the Eighth Circuit has rejected a similar argument,\\nholding that “[d]ismissals based on immunity are not among\\nthe types of dismissals listed as strikes in section 1915(g).”\\nCastillo-Alvarez v. Krukow, 768 F.3d 1219, 1220 (8th Cir.\\n2014) (per curiam). We agree.\\n\\n The language and structure of the PLRA make clear that\\nimmunity-based dismissals generally do not fall within\\n§ 1915(g). Section 1915 sets out the IFP application\\nprocedures and rules regarding payment of court fees in\\nsubsections (a) through (d). Under § 1915(e)(2), regardless\\nof the filing fee payment, “the court shall [sua sponte]\\ndismiss the case at any time” if the case:\\n\\n (i) is frivolous or malicious;\\n\\n (ii) fails to state a claim on which relief\\n may be granted; or\\n\\n (iii) seeks monetary relief against a\\n defendant who is immune from such\\n relief,\\n\\n28 U.S.C. § 1915(e)(2)(B) (emphasis added).\\n\\n Yet the strike provision, subsection (g), which follows\\nsubsection (e), omits the immunity language as a ground for\\na strike. “[W]here Congress includes particular language in\\none section of a statute but omits it in another section of the\\nsame Act, it is generally presumed that Congress acts\\n\\f10 HARRIS V. HARRIS\\n\\nintentionally and purposefully in the disparate inclusion or\\nexclusion.” Russello v. United States, 464 U.S. 16, 23\\n(1983) (quoting United States v. Wong Kim Bo, 472 F.2d\\n720, 722 (5th Cir. 1972)).\\n\\n The PLRA also expressly lists immunity-based\\ndismissals in three other sections, further demonstrating that\\nCongress intentionally excluded immunity from the strike\\nprovision. For example, in § 1915A, Congress implemented\\na screening procedure for civil claims made by incarcerated\\nplaintiffs against “a governmental entity or officer or\\nemployee of a governmental entity.” 28 U.S.C. § 1915A(a).\\nWhen listing the grounds for dismissal while screening a\\ncomplaint, Congress included the three enumerated strike\\ngrounds and then added “or [] seeks monetary relief from a\\ndefendant who is immune from such relief,” 28 U.S.C.\\n§ 1915A(b), mirroring the language in 28 U.S.C.\\n§ 1915(e)(2)(B) discussed above. Congress intended for the\\nscreening procedure to address immunity because it\\nexpressly listed immunity after frivolousness,\\nmaliciousness, and failure to state a claim. See also\\n42 U.S.C. § 1997e(c) (stating that the court may dismiss a\\ncase regarding prison conditions or in which administrative\\nremedies have not been exhausted if it is “frivolous,\\nmalicious, fails to state a claim upon which relief can be\\ngranted, or seeks monetary relief from a defendant who is\\nimmune from such relief” in subsections 1997e(c)(1) and\\n(2) (emphasis added)). In sum, Congress’ omission of\\nimmunity-based dismissal from the strike provision in\\n§ 1915(g) evidences its intent generally not to include this\\ndismissal ground as a strike.\\n\\n To be sure, there are rare cases where an affirmative\\ndefense, such as immunity, may be so clear on the face of\\nthe complaint that dismissal may qualify as a strike for\\n\\f HARRIS V. HARRIS 11\\n\\nfailure to state a claim. See, e.g., El-Shaddai, 833 F.3d\\nat 1044 (recognizing the rarity of such dismissals and\\ndeclining to find a strike when the affirmative defense of\\nadministrative exhaustion was not clear on the face of the\\ncomplaint); Albino v. Baca, 747 F.3d 1162, 1169 (9th Cir.\\n2014) (en banc) (discussing the “rare cases” when failure to\\nexhaust is clear from the face of the complaint); Hafed v.\\nFed. Bureau of Prisons, 635 F.3d 1172, 1178 (10th Cir.\\n2011) (holding that a prior dismissal on immunity grounds\\nwas a strike where the prisoner “affirmatively asserted facts”\\ndemonstrating immunity (emphasis in original)), abrogated\\non other grounds as recognized by Carr v. Zwally, 760\\nF. App’x 550, 558 (10th Cir. 2019). There are also cases\\nwhere an affirmative defense like immunity is so obvious\\nthat the suit is frivolous, and dismissal counts as a strike. See\\nMills v. Fischer, 645 F.3d 176, 177 (2d Cir. 2011) (holding\\nthat a dismissal for judicial immunity was a strike because\\nthe claim was frivolous). But these are exceptional cases\\nwhere the affirmative defense is readily apparent without\\nresort to any additional information outside the four corners\\nof the complaint. Such will rarely be the case with\\nimmunity-based defenses. See, e.g., Milstein v. Cooley, 257\\nF.3d 1004 (9th Cir. 2001) (discussing at length the\\ncomplexities of prosecutorial immunity within the context of\\na Rule 12(b)(6) dismissal). And here, that is decidedly not\\nthe case. The district court moved beyond the Nielsen\\ncomplaint not only to consider the defendants’ roles, titles,\\nand actions, but to grapple with the legal analysis entitling\\nthe defendants to quasi-judicial immunity. Because the\\ncourt dismissed Harris’s Nielsen complaint on immunity\\ngrounds rather than the grounds that it was frivolous,\\nmalicious, or failed to state a claim, we conclude that Nielsen\\nis not a strike under § 1915(g).\\n\\f12 HARRIS V. HARRIS\\n\\n IV.\\n\\n The “denial of [IFP] status effectively, if not physically,\\ndenies many indigent prisoners access to the courts.”\\nSimone Schonenberger, Access Denied: The Prison\\nLitigation Reform Act, 86 Ky. L.J. 457, 474 (1998). In\\n§ 1915(g), Congress said what it meant, and we will construe\\nits language strictly and narrowly. “Our task is to give effect\\nto the will of Congress, and where its will has been expressed\\nin reasonably plain terms, that language must ordinarily be\\nregarded as conclusive.” Griffin v. Oceanic Contractors,\\nInc., 458 U.S. 564, 570 (1982) (internal quotation marks\\nomitted) (quoting Consumer Prod. Safety Comm’n v. GTE\\nSylvania, Inc., 447 U.S. 102, 108 (1980)). Unless an\\nincarcerated litigant has accrued three strikes on grounds\\nplainly enumerated in § 1915(g), she is entitled to IFP status.\\nDismissals for supplemental jurisdiction, failure to serve,\\nand quasi-judicial immunity are not grounds giving rise to\\nstrikes under § 1915(g) of the PLRA. With only two strikes\\nfrom prior litigation, Harris may proceed in this suit with IFP\\nstatus as long as he is otherwise eligible for IFP status.\\n\\n REVERSED AND REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4431730/\", \"author_raw\": \"NGUYEN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"RONALD M GOULD"},"judge2":{"kind":"string","value":"JACQUELINE H NGUYEN"},"judge3":{"kind":"string","value":"ALGENON L MARBLEY"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4654477/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Tommie Harris v. K. Harris\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-21\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n TOMMIE LEE HARRIS, No. 16-55083\n Plaintiff-Appellant,\n D.C. No.\n v. 2:15-cv-03104-\n ODW-E\n K. HARRIS, Correctional Officer,\n individual and official capacity,\n Defendant-Appellee. OPINION\n\n Appeal from the United States District Court\n for the Central District of California\n Otis D. Wright II, District Judge, Presiding\n\n Argued and Submitted February 6, 2019\n Pasadena, California\n\n Filed August 21, 2019\n\n Before: Ronald M. Gould and Jacqueline H. Nguyen,\n Circuit Judges, and Algenon L. Marbley, * District Judge.\n\n Opinion by Judge Nguyen\n\n\n\n\n *\n The Honorable Algenon L. Marbley, United States District Judge\nfor the Southern District of Ohio, sitting by designation.\n\f2 HARRIS V. HARRIS\n\n SUMMARY **\n\n\n Prisoner Civil Rights\n\n The panel reversed the district court’s revocation of a\nstate prisoner’s in forma pauperis status on the ground that\nhe had three prior strikes under the Prison Litigation Reform\nAct, 28 U.S.C. § 1915(g), and remanded.\n\n One of plaintiff’s prior cases was dismissed because,\nafter concluding that he failed to state a federal claim, the\ndistrict court declined to exercise supplemental jurisdiction\nover the remaining state claims. Another was dismissed\nbecause plaintiff failed to serve one defendant, and several\nother defendants enjoyed quasi-judicial immunity.\n\n The panel held that because the prior cases were not\ndismissed on grounds enumerated in § 1915(g), they did not\nqualify as strikes. Following the D.C. Circuit’s decision in\nFourstar v. Garden City Grp., Inc., 875 F.3d 1147, 1152\n(D.C. Cir. 2017), the panel first held that a dismissal based\non a district court’s decision not to exercise supplemental\njurisdiction is not an enumerated ground under § 1915(g).\nThe panel further held that dismissal due to a failure to serve\nis plainly not a dismissal on the ground that the suit was\nfrivolous, malicious, or failed to state a claim. It is therefore\nnot a strike under § 1915(g). Finally, the panel held that the\nlanguage and structure of the Prison Litigation Reform Act\nmake clear that immunity-based dismissals generally do not\nfall within § 1915(g).\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f HARRIS V. HARRIS 3\n\n COUNSEL\n\nDaniel A. Arellano (argued), Ballard Spahr LLP, Phoenix,\nArizona, for Plaintiff-Appellant.\n\nTodd Grabarsky (argued), Deputy Attorney General;\nThomas S. Patterson and Misha D. Igra, Supervising Deputy\nAttorneys General; Monica N. Anderson, Senior Assistant\nAttorney General; Xavier Becerra, Attorney General; Office\nof the Attorney General, Los Angeles, California; for\nDefendant-Appellee.\n\n\n OPINION\n\nNGUYEN, Circuit Judge:\n\n Tommie Lee Harris, a state prisoner, appeals the district\ncourt’s decision revoking his in forma pauperis (“IFP”)\nstatus on the ground that he had three prior strikes under the\nPrison Litigation Reform Act (“PLRA”), 28 U.S.C.\n§ 1915(g). One of Harris’s prior cases was dismissed\nbecause, after concluding that he failed to state a federal\nclaim, the district court declined to exercise supplemental\njurisdiction over the remaining state claims. Another was\ndismissed because Harris failed to serve one defendant, and\nseveral other defendants enjoyed quasi-judicial immunity.\nWe hold that because these cases were not dismissed on\ngrounds enumerated in § 1915(g), they do not qualify as\nstrikes. We therefore reverse and remand.\n\n I.\n\n Harris filed the current lawsuit against a correctional\nofficer under 42 U.S.C. § 1983 for use of excessive force.\nHarris also filed for IFP status. The district court found that\n\f4 HARRIS V. HARRIS\n\nHarris had already accrued three strikes and revoked his IFP\nstatus. Harris appeals. He admits that he has two strikes 1\nunder the PLRA but argues that two of his other prior\nlawsuits are not strikes, Harris v. Bick, No. 2:98-cv-01197-\nLKK-DAD (E.D. Cal. Nov. 17, 1998) and Harris v. Nielsen,\nNo. 2:98-mc-00225-WBS-GGH (PC) (E.D. Cal. Apr. 27,\n2001).\n\nA. Harris v. Bick\n\n In Harris v. Bick, Harris sued various prison medical\nprofessionals, including Dr. Bick, alleging he received\ninadequate medical care in violation of state law and the\nEighth Amendment. After Harris attempted to voluntarily\ndismiss Dr. Bick, the district court found that the allegations\nagainst the remaining defendants failed to state an Eighth\nAmendment claim, and “in the absence of a cognizable\nfederal claim, the court [declined] to exercise supplemental\njurisdiction over [Harris’s] state contract and debtor-creditor\nclaims.” On appeal, we affirmed the district court’s\ndismissal on the Eighth Amendment claims and its decision\nnot to exercise supplemental jurisdiction.\n\nB. Harris v. Nielsen\n\n In Harris v. Nielsen, Harris’s claims against various\nmedical professionals arose out of their participation in\npreparing mental health reports submitted in connection with\nhis parole revocation hearings. Harris named defendants E.\nTitus, J. Choy, G. Phelps, J. Karuzas, and C. Carter. The\ndistrict court dismissed one defendant, Carter, because the\n 1\n Harris concedes that Harris v. Geraghty, No. 98-CV-861-GEB-\nJFM (E.D. Cal. May 25, 1999) (dismissed for failure to state a claim),\nand Harris v. Nielsen, No. 01-15006 (9th Cir. Jun. 19, 2001) (appeal\ndismissed as frivolous), count as strikes under the PLRA. We agree.\n\f HARRIS V. HARRIS 5\n\nUnited States Marshal was unable to serve him despite\nattempts at two workplaces. The court also dismissed three\ndefendants, Titus, Choy, and Phelps, because, as court-\nappointed psychologists, psychiatrists, and counselors at the\nparole hearings, they were protected by quasi-judicial\nimmunity. The district court granted defendant Karuzas’s\nmotion for judgment on the pleadings for the same reason.\n\n The question before us is whether these two suits, Bick\nand Nielsen, qualify as strikes under § 1915(g). 2\n\n II.\n\n We interpret § 1915(g) de novo. Andrews v. King,\n398 F.3d 1113, 1118 (9th Cir. 2005). The denial of IFP\nstatus is appealable as a final judgment. Id. We have\njurisdiction under 28 U.S.C. § 1291.\n\n III.\n\nA. The PLRA’s Section 1915(g)\n\n Litigants who qualify for IFP status are excused from\nprepaying court fees and costs. The PLRA’s “three strikes”\nprovision, designed to discourage vexatious and voluminous\nprisoner litigation, bars a prisoner from bringing a civil\naction or an appeal IFP if the prisoner has three prior actions\nthat were “dismissed on the grounds that it is frivolous,\nmalicious, or fails to state a claim upon which relief may be\ngranted, unless the prisoner is under imminent danger of\nserious physical injury.” 28 U.S.C. § 1915(g) (emphasis\nadded).\n\n 2\n We GRANT the pending motions for judicial notice (Docket Nos.\n21, 37, and 47).\n\f6 HARRIS V. HARRIS\n\n “[I]n a statutory construction case, analysis must begin\nwith the language of the statute itself; when the statute is\nclear, judicial inquiry into its meaning, in all but the most\nextraordinary circumstance, is finished.” Talamantes v.\nLeyva, 575 F.3d 1021, 1023 (9th Cir. 2009) (alterations\nomitted) (quoting United States v. Carter, 421 F.3d 909, 911\n(9th Cir. 2005)). Here, the statutory language is clear—if a\ncase was not dismissed on one of the specific enumerated\ngrounds, it does not count as a strike under § 1915(g).\n\n A defendant challenging a plaintiff’s IFP status bears the\ninitial burden of showing through documentary evidence\nthat a plaintiff had three prior strikes. King, 398 F.3d\nat 1118–20. If a defendant presents a prima facie case, then\n“the burden shifts to the plaintiff to persuade the court that\n§ 1915(g) does not apply.” Id. at 1116. In determining\nwhether a prior dismissal counts as a strike, “we should look\nto the substance of the dismissed lawsuit,” and not to how\nthe district court labelled or styled the dismissal. El-Shaddai\nv. Zamora, 833 F.3d 1036, 1047 (9th Cir. 2016) (citing King,\n398 F.3d at 1122 n.12) (finding that out of eleven prior suits,\nonly one counted as a strike); see also Knapp v. Hogan, C.O.,\n738 F.3d 1106, 1109 (9th Cir. 2013) (stating that we look to\n“the dismissing court’s action and the reasons underlying\nit . . . . [T]he procedural mechanism or Rule by which the\ndismissal is accomplished, while informative, is not\ndispositive.” (internal citation omitted)).\n\n We must also evaluate whether all the claims in a given\nsuit satisfy the enumerated grounds for strikes, and partial\ndismissals of even one claim for a non-qualifying reason will\nsave an entire case from constituting a strike. See\nWashington v. L.A. Cty. Sheriff’s Dep’t, 833 F.3d 1048, 1057\n(9th Cir. 2016). In other words, “[w]hen we are presented\nwith multiple claims within a single action, we assess a\n\f HARRIS V. HARRIS 7\n\nPLRA strike only when the ‘case as a whole’ is dismissed\nfor a qualifying reason.” Id. (quoting Andrews v. Cervantes,\n493 F.3d 1047, 1054 (9th Cir. 2007), as amended (July 5,\n2007)).\n\nB. A Dismissal Based on the Court’s Refusal to Exercise\n Supplemental Jurisdiction Does Not Count as a\n Strike\n\n Harris argues that the dismissal of Harris v. Bick was due\nto the district court’s refusal to exercise supplemental\njurisdiction over his state-law claims after dismissal of his\nEighth Amendment claim, and thus this case does not qualify\nas a strike. 3 We agree. Dismissal based on a district court’s\ndecision not to exercise supplemental jurisdiction is not an\nenumerated ground under § 1915(g). As then-Judge\nKavanaugh explained, “[w]hen a district court has declined\nto exercise supplemental jurisdiction over state-law claims,\nthe court has not dismissed the state-law claims for failure to\nstate a claim, nor has the court dismissed the state-law claims\nas frivolous or malicious.” Fourstar v. Garden City Grp.,\nInc., 875 F.3d 1147, 1152 (D.C. Cir. 2017). Because the\nlanguage of § 1915(g) is plain and unambiguous, a dismissal\non a ground other than frivolousness, maliciousness, or\nfailure to state a claim will not qualify as a strike. See id.\nat 1151–52.\n\n Defendant argues that policy considerations counsel in\nfavor of counting Bick as a strike. Otherwise, Defendant\nurges, a prisoner can easily “strike-proof” a meritless lawsuit\n\n 3\n Harris also argues that his attempt to voluntarily dismiss Dr. Bick\nis an independent ground for not counting this case as a strike. The\nparties dispute whether his attempt was sufficient but, in any event, we\nneed not decide this issue.\n\f8 HARRIS V. HARRIS\n\nby adding state-law claims, knowing that federal courts are\nunlikely to expend limited resources to exercise\nsupplemental jurisdiction over such claims in the absence of\na cognizable federal claim. Of course, courts are free to\nreach state-law claims and dismiss them on an enumerated\nstrike ground. But even if the policy concerns are warranted,\nwe must still strictly construe the plain language of the\nstatute. “It is not a judge’s job to add to or otherwise re-mold\nstatutory text to try to meet a statute’s perceived policy\nobjectives. Instead, we must apply the statute as written.”\nFourstar, 875 F.3d at 1152. We follow the D.C. Circuit and\nhold that a dismissal due to the district court’s decision not\nto exercise supplemental jurisdiction over state-law claims\ndoes not qualify the case as a strike under the PLRA.\n\nC. Neither a Dismissal Due to a Failure to Serve Nor a\n Dismissal Based on Quasi-Judicial Immunity\n Qualifies as a Strike\n\n Harris argues that Harris v. Nielsen does not qualify as a\nstrike. The district court dismissed one defendant, Carter,\nbecause the United States Marshal was unable to serve him\ndespite multiple attempts at two workplaces. A dismissal\ndue to a failure to serve is plainly not a dismissal on the\nground that the suit was frivolous, malicious, or failed to\nstate a claim. It is therefore not a strike under § 1915(g).\nDefendant resists a plain reading of the statute and argues\nthat a dismissal of one defendant for failure to serve should\nnot prevent a case from qualifying as a strike where other\nclaims were dismissed for frivolousness or failure to state a\nclaim. But we evaluate the “case as a whole” and dismissal\nof even one claim for an unenumerated reason saves the\nentire case from counting as a strike. Washington, 833 F.3d\nat 1057. Therefore, Nielsen does not qualify as a strike\nbecause one defendant was dismissed for failure to serve.\n\f HARRIS V. HARRIS 9\n\n Nielsen does not qualify as a strike for an additional,\nindependent reason. The district court dismissed the\nremaining defendants because it found them entitled to\nquasi-judicial immunity. Defendant argues that quasi-\njudicial immunity falls within the enumerated strike ground\nof failure to state a claim. We have not addressed this issue,\nbut the Eighth Circuit has rejected a similar argument,\nholding that “[d]ismissals based on immunity are not among\nthe types of dismissals listed as strikes in section 1915(g).”\nCastillo-Alvarez v. Krukow, 768 F.3d 1219, 1220 (8th Cir.\n2014) (per curiam). We agree.\n\n The language and structure of the PLRA make clear that\nimmunity-based dismissals generally do not fall within\n§ 1915(g). Section 1915 sets out the IFP application\nprocedures and rules regarding payment of court fees in\nsubsections (a) through (d). Under § 1915(e)(2), regardless\nof the filing fee payment, “the court shall [sua sponte]\ndismiss the case at any time” if the case:\n\n (i) is frivolous or malicious;\n\n (ii) fails to state a claim on which relief\n may be granted; or\n\n (iii) seeks monetary relief against a\n defendant who is immune from such\n relief,\n\n28 U.S.C. § 1915(e)(2)(B) (emphasis added).\n\n Yet the strike provision, subsection (g), which follows\nsubsection (e), omits the immunity language as a ground for\na strike. “[W]here Congress includes particular language in\none section of a statute but omits it in another section of the\nsame Act, it is generally presumed that Congress acts\n\f10 HARRIS V. HARRIS\n\nintentionally and purposefully in the disparate inclusion or\nexclusion.” Russello v. United States, 464 U.S. 16, 23\n(1983) (quoting United States v. Wong Kim Bo, 472 F.2d\n720, 722 (5th Cir. 1972)).\n\n The PLRA also expressly lists immunity-based\ndismissals in three other sections, further demonstrating that\nCongress intentionally excluded immunity from the strike\nprovision. For example, in § 1915A, Congress implemented\na screening procedure for civil claims made by incarcerated\nplaintiffs against “a governmental entity or officer or\nemployee of a governmental entity.” 28 U.S.C. § 1915A(a).\nWhen listing the grounds for dismissal while screening a\ncomplaint, Congress included the three enumerated strike\ngrounds and then added “or [] seeks monetary relief from a\ndefendant who is immune from such relief,” 28 U.S.C.\n§ 1915A(b), mirroring the language in 28 U.S.C.\n§ 1915(e)(2)(B) discussed above. Congress intended for the\nscreening procedure to address immunity because it\nexpressly listed immunity after frivolousness,\nmaliciousness, and failure to state a claim. See also\n42 U.S.C. § 1997e(c) (stating that the court may dismiss a\ncase regarding prison conditions or in which administrative\nremedies have not been exhausted if it is “frivolous,\nmalicious, fails to state a claim upon which relief can be\ngranted, or seeks monetary relief from a defendant who is\nimmune from such relief” in subsections 1997e(c)(1) and\n(2) (emphasis added)). In sum, Congress’ omission of\nimmunity-based dismissal from the strike provision in\n§ 1915(g) evidences its intent generally not to include this\ndismissal ground as a strike.\n\n To be sure, there are rare cases where an affirmative\ndefense, such as immunity, may be so clear on the face of\nthe complaint that dismissal may qualify as a strike for\n\f HARRIS V. HARRIS 11\n\nfailure to state a claim. See, e.g., El-Shaddai, 833 F.3d\nat 1044 (recognizing the rarity of such dismissals and\ndeclining to find a strike when the affirmative defense of\nadministrative exhaustion was not clear on the face of the\ncomplaint); Albino v. Baca, 747 F.3d 1162, 1169 (9th Cir.\n2014) (en banc) (discussing the “rare cases” when failure to\nexhaust is clear from the face of the complaint); Hafed v.\nFed. Bureau of Prisons, 635 F.3d 1172, 1178 (10th Cir.\n2011) (holding that a prior dismissal on immunity grounds\nwas a strike where the prisoner “affirmatively asserted facts”\ndemonstrating immunity (emphasis in original)), abrogated\non other grounds as recognized by Carr v. Zwally, 760\nF. App’x 550, 558 (10th Cir. 2019). There are also cases\nwhere an affirmative defense like immunity is so obvious\nthat the suit is frivolous, and dismissal counts as a strike. See\nMills v. Fischer, 645 F.3d 176, 177 (2d Cir. 2011) (holding\nthat a dismissal for judicial immunity was a strike because\nthe claim was frivolous). But these are exceptional cases\nwhere the affirmative defense is readily apparent without\nresort to any additional information outside the four corners\nof the complaint. Such will rarely be the case with\nimmunity-based defenses. See, e.g., Milstein v. Cooley, 257\nF.3d 1004 (9th Cir. 2001) (discussing at length the\ncomplexities of prosecutorial immunity within the context of\na Rule 12(b)(6) dismissal). And here, that is decidedly not\nthe case. The district court moved beyond the Nielsen\ncomplaint not only to consider the defendants’ roles, titles,\nand actions, but to grapple with the legal analysis entitling\nthe defendants to quasi-judicial immunity. Because the\ncourt dismissed Harris’s Nielsen complaint on immunity\ngrounds rather than the grounds that it was frivolous,\nmalicious, or failed to state a claim, we conclude that Nielsen\nis not a strike under § 1915(g).\n\f12 HARRIS V. HARRIS\n\n IV.\n\n The “denial of [IFP] status effectively, if not physically,\ndenies many indigent prisoners access to the courts.”\nSimone Schonenberger, Access Denied: The Prison\nLitigation Reform Act, 86 Ky. L.J. 457, 474 (1998). In\n§ 1915(g), Congress said what it meant, and we will construe\nits language strictly and narrowly. “Our task is to give effect\nto the will of Congress, and where its will has been expressed\nin reasonably plain terms, that language must ordinarily be\nregarded as conclusive.” Griffin v. Oceanic Contractors,\nInc., 458 U.S. 564, 570 (1982) (internal quotation marks\nomitted) (quoting Consumer Prod. Safety Comm’n v. GTE\nSylvania, Inc., 447 U.S. 102, 108 (1980)). Unless an\nincarcerated litigant has accrued three strikes on grounds\nplainly enumerated in § 1915(g), she is entitled to IFP status.\nDismissals for supplemental jurisdiction, failure to serve,\nand quasi-judicial immunity are not grounds giving rise to\nstrikes under § 1915(g) of the PLRA. With only two strikes\nfrom prior litigation, Harris may proceed in this suit with IFP\nstatus as long as he is otherwise eligible for IFP status.\n\n REVERSED AND REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Tommie Harris v. K. Harris\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-21\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n TOMMIE LEE HARRIS, No. 16-55083\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:15-cv-03104-\\n ODW-E\\n K. HARRIS, Correctional Officer,\\n individual and official capacity,\\n Defendant-Appellee. OPINION\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Otis D. Wright II, District Judge, Presiding\\n\\n Argued and Submitted February 6, 2019\\n Pasadena, California\\n\\n Filed August 21, 2019\\n\\n Before: Ronald M. Gould and Jacqueline H. Nguyen,\\n Circuit Judges, and Algenon L. Marbley, * District Judge.\\n\\n Opinion by Judge Nguyen\\n\\n\\n\\n\\n *\\n The Honorable Algenon L. Marbley, United States District Judge\\nfor the Southern District of Ohio, sitting by designation.\\n\\f2 HARRIS V. HARRIS\\n\\n SUMMARY **\\n\\n\\n Prisoner Civil Rights\\n\\n The panel reversed the district court’s revocation of a\\nstate prisoner’s in forma pauperis status on the ground that\\nhe had three prior strikes under the Prison Litigation Reform\\nAct, 28 U.S.C. § 1915(g), and remanded.\\n\\n One of plaintiff’s prior cases was dismissed because,\\nafter concluding that he failed to state a federal claim, the\\ndistrict court declined to exercise supplemental jurisdiction\\nover the remaining state claims. Another was dismissed\\nbecause plaintiff failed to serve one defendant, and several\\nother defendants enjoyed quasi-judicial immunity.\\n\\n The panel held that because the prior cases were not\\ndismissed on grounds enumerated in § 1915(g), they did not\\nqualify as strikes. Following the D.C. Circuit’s decision in\\nFourstar v. Garden City Grp., Inc., 875 F.3d 1147, 1152\\n(D.C. Cir. 2017), the panel first held that a dismissal based\\non a district court’s decision not to exercise supplemental\\njurisdiction is not an enumerated ground under § 1915(g).\\nThe panel further held that dismissal due to a failure to serve\\nis plainly not a dismissal on the ground that the suit was\\nfrivolous, malicious, or failed to state a claim. It is therefore\\nnot a strike under § 1915(g). Finally, the panel held that the\\nlanguage and structure of the Prison Litigation Reform Act\\nmake clear that immunity-based dismissals generally do not\\nfall within § 1915(g).\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f HARRIS V. HARRIS 3\\n\\n COUNSEL\\n\\nDaniel A. Arellano (argued), Ballard Spahr LLP, Phoenix,\\nArizona, for Plaintiff-Appellant.\\n\\nTodd Grabarsky (argued), Deputy Attorney General;\\nThomas S. Patterson and Misha D. Igra, Supervising Deputy\\nAttorneys General; Monica N. Anderson, Senior Assistant\\nAttorney General; Xavier Becerra, Attorney General; Office\\nof the Attorney General, Los Angeles, California; for\\nDefendant-Appellee.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n Tommie Lee Harris, a state prisoner, appeals the district\\ncourt’s decision revoking his in forma pauperis (“IFP”)\\nstatus on the ground that he had three prior strikes under the\\nPrison Litigation Reform Act (“PLRA”), 28 U.S.C.\\n§ 1915(g). One of Harris’s prior cases was dismissed\\nbecause, after concluding that he failed to state a federal\\nclaim, the district court declined to exercise supplemental\\njurisdiction over the remaining state claims. Another was\\ndismissed because Harris failed to serve one defendant, and\\nseveral other defendants enjoyed quasi-judicial immunity.\\nWe hold that because these cases were not dismissed on\\ngrounds enumerated in § 1915(g), they do not qualify as\\nstrikes. We therefore reverse and remand.\\n\\n I.\\n\\n Harris filed the current lawsuit against a correctional\\nofficer under 42 U.S.C. § 1983 for use of excessive force.\\nHarris also filed for IFP status. The district court found that\\n\\f4 HARRIS V. HARRIS\\n\\nHarris had already accrued three strikes and revoked his IFP\\nstatus. Harris appeals. He admits that he has two strikes 1\\nunder the PLRA but argues that two of his other prior\\nlawsuits are not strikes, Harris v. Bick, No. 2:98-cv-01197-\\nLKK-DAD (E.D. Cal. Nov. 17, 1998) and Harris v. Nielsen,\\nNo. 2:98-mc-00225-WBS-GGH (PC) (E.D. Cal. Apr. 27,\\n2001).\\n\\nA. Harris v. Bick\\n\\n In Harris v. Bick, Harris sued various prison medical\\nprofessionals, including Dr. Bick, alleging he received\\ninadequate medical care in violation of state law and the\\nEighth Amendment. After Harris attempted to voluntarily\\ndismiss Dr. Bick, the district court found that the allegations\\nagainst the remaining defendants failed to state an Eighth\\nAmendment claim, and “in the absence of a cognizable\\nfederal claim, the court [declined] to exercise supplemental\\njurisdiction over [Harris’s] state contract and debtor-creditor\\nclaims.” On appeal, we affirmed the district court’s\\ndismissal on the Eighth Amendment claims and its decision\\nnot to exercise supplemental jurisdiction.\\n\\nB. Harris v. Nielsen\\n\\n In Harris v. Nielsen, Harris’s claims against various\\nmedical professionals arose out of their participation in\\npreparing mental health reports submitted in connection with\\nhis parole revocation hearings. Harris named defendants E.\\nTitus, J. Choy, G. Phelps, J. Karuzas, and C. Carter. The\\ndistrict court dismissed one defendant, Carter, because the\\n 1\\n Harris concedes that Harris v. Geraghty, No. 98-CV-861-GEB-\\nJFM (E.D. Cal. May 25, 1999) (dismissed for failure to state a claim),\\nand Harris v. Nielsen, No. 01-15006 (9th Cir. Jun. 19, 2001) (appeal\\ndismissed as frivolous), count as strikes under the PLRA. We agree.\\n\\f HARRIS V. HARRIS 5\\n\\nUnited States Marshal was unable to serve him despite\\nattempts at two workplaces. The court also dismissed three\\ndefendants, Titus, Choy, and Phelps, because, as court-\\nappointed psychologists, psychiatrists, and counselors at the\\nparole hearings, they were protected by quasi-judicial\\nimmunity. The district court granted defendant Karuzas’s\\nmotion for judgment on the pleadings for the same reason.\\n\\n The question before us is whether these two suits, Bick\\nand Nielsen, qualify as strikes under § 1915(g). 2\\n\\n II.\\n\\n We interpret § 1915(g) de novo. Andrews v. King,\\n398 F.3d 1113, 1118 (9th Cir. 2005). The denial of IFP\\nstatus is appealable as a final judgment. Id. We have\\njurisdiction under 28 U.S.C. § 1291.\\n\\n III.\\n\\nA. The PLRA’s Section 1915(g)\\n\\n Litigants who qualify for IFP status are excused from\\nprepaying court fees and costs. The PLRA’s “three strikes”\\nprovision, designed to discourage vexatious and voluminous\\nprisoner litigation, bars a prisoner from bringing a civil\\naction or an appeal IFP if the prisoner has three prior actions\\nthat were “dismissed on the grounds that it is frivolous,\\nmalicious, or fails to state a claim upon which relief may be\\ngranted, unless the prisoner is under imminent danger of\\nserious physical injury.” 28 U.S.C. § 1915(g) (emphasis\\nadded).\\n\\n 2\\n We GRANT the pending motions for judicial notice (Docket Nos.\\n21, 37, and 47).\\n\\f6 HARRIS V. HARRIS\\n\\n “[I]n a statutory construction case, analysis must begin\\nwith the language of the statute itself; when the statute is\\nclear, judicial inquiry into its meaning, in all but the most\\nextraordinary circumstance, is finished.” Talamantes v.\\nLeyva, 575 F.3d 1021, 1023 (9th Cir. 2009) (alterations\\nomitted) (quoting United States v. Carter, 421 F.3d 909, 911\\n(9th Cir. 2005)). Here, the statutory language is clear—if a\\ncase was not dismissed on one of the specific enumerated\\ngrounds, it does not count as a strike under § 1915(g).\\n\\n A defendant challenging a plaintiff’s IFP status bears the\\ninitial burden of showing through documentary evidence\\nthat a plaintiff had three prior strikes. King, 398 F.3d\\nat 1118–20. If a defendant presents a prima facie case, then\\n“the burden shifts to the plaintiff to persuade the court that\\n§ 1915(g) does not apply.” Id. at 1116. In determining\\nwhether a prior dismissal counts as a strike, “we should look\\nto the substance of the dismissed lawsuit,” and not to how\\nthe district court labelled or styled the dismissal. El-Shaddai\\nv. Zamora, 833 F.3d 1036, 1047 (9th Cir. 2016) (citing King,\\n398 F.3d at 1122 n.12) (finding that out of eleven prior suits,\\nonly one counted as a strike); see also Knapp v. Hogan, C.O.,\\n738 F.3d 1106, 1109 (9th Cir. 2013) (stating that we look to\\n“the dismissing court’s action and the reasons underlying\\nit . . . . [T]he procedural mechanism or Rule by which the\\ndismissal is accomplished, while informative, is not\\ndispositive.” (internal citation omitted)).\\n\\n We must also evaluate whether all the claims in a given\\nsuit satisfy the enumerated grounds for strikes, and partial\\ndismissals of even one claim for a non-qualifying reason will\\nsave an entire case from constituting a strike. See\\nWashington v. L.A. Cty. Sheriff’s Dep’t, 833 F.3d 1048, 1057\\n(9th Cir. 2016). In other words, “[w]hen we are presented\\nwith multiple claims within a single action, we assess a\\n\\f HARRIS V. HARRIS 7\\n\\nPLRA strike only when the ‘case as a whole’ is dismissed\\nfor a qualifying reason.” Id. (quoting Andrews v. Cervantes,\\n493 F.3d 1047, 1054 (9th Cir. 2007), as amended (July 5,\\n2007)).\\n\\nB. A Dismissal Based on the Court’s Refusal to Exercise\\n Supplemental Jurisdiction Does Not Count as a\\n Strike\\n\\n Harris argues that the dismissal of Harris v. Bick was due\\nto the district court’s refusal to exercise supplemental\\njurisdiction over his state-law claims after dismissal of his\\nEighth Amendment claim, and thus this case does not qualify\\nas a strike. 3 We agree. Dismissal based on a district court’s\\ndecision not to exercise supplemental jurisdiction is not an\\nenumerated ground under § 1915(g). As then-Judge\\nKavanaugh explained, “[w]hen a district court has declined\\nto exercise supplemental jurisdiction over state-law claims,\\nthe court has not dismissed the state-law claims for failure to\\nstate a claim, nor has the court dismissed the state-law claims\\nas frivolous or malicious.” Fourstar v. Garden City Grp.,\\nInc., 875 F.3d 1147, 1152 (D.C. Cir. 2017). Because the\\nlanguage of § 1915(g) is plain and unambiguous, a dismissal\\non a ground other than frivolousness, maliciousness, or\\nfailure to state a claim will not qualify as a strike. See id.\\nat 1151–52.\\n\\n Defendant argues that policy considerations counsel in\\nfavor of counting Bick as a strike. Otherwise, Defendant\\nurges, a prisoner can easily “strike-proof” a meritless lawsuit\\n\\n 3\\n Harris also argues that his attempt to voluntarily dismiss Dr. Bick\\nis an independent ground for not counting this case as a strike. The\\nparties dispute whether his attempt was sufficient but, in any event, we\\nneed not decide this issue.\\n\\f8 HARRIS V. HARRIS\\n\\nby adding state-law claims, knowing that federal courts are\\nunlikely to expend limited resources to exercise\\nsupplemental jurisdiction over such claims in the absence of\\na cognizable federal claim. Of course, courts are free to\\nreach state-law claims and dismiss them on an enumerated\\nstrike ground. But even if the policy concerns are warranted,\\nwe must still strictly construe the plain language of the\\nstatute. “It is not a judge’s job to add to or otherwise re-mold\\nstatutory text to try to meet a statute’s perceived policy\\nobjectives. Instead, we must apply the statute as written.”\\nFourstar, 875 F.3d at 1152. We follow the D.C. Circuit and\\nhold that a dismissal due to the district court’s decision not\\nto exercise supplemental jurisdiction over state-law claims\\ndoes not qualify the case as a strike under the PLRA.\\n\\nC. Neither a Dismissal Due to a Failure to Serve Nor a\\n Dismissal Based on Quasi-Judicial Immunity\\n Qualifies as a Strike\\n\\n Harris argues that Harris v. Nielsen does not qualify as a\\nstrike. The district court dismissed one defendant, Carter,\\nbecause the United States Marshal was unable to serve him\\ndespite multiple attempts at two workplaces. A dismissal\\ndue to a failure to serve is plainly not a dismissal on the\\nground that the suit was frivolous, malicious, or failed to\\nstate a claim. It is therefore not a strike under § 1915(g).\\nDefendant resists a plain reading of the statute and argues\\nthat a dismissal of one defendant for failure to serve should\\nnot prevent a case from qualifying as a strike where other\\nclaims were dismissed for frivolousness or failure to state a\\nclaim. But we evaluate the “case as a whole” and dismissal\\nof even one claim for an unenumerated reason saves the\\nentire case from counting as a strike. Washington, 833 F.3d\\nat 1057. Therefore, Nielsen does not qualify as a strike\\nbecause one defendant was dismissed for failure to serve.\\n\\f HARRIS V. HARRIS 9\\n\\n Nielsen does not qualify as a strike for an additional,\\nindependent reason. The district court dismissed the\\nremaining defendants because it found them entitled to\\nquasi-judicial immunity. Defendant argues that quasi-\\njudicial immunity falls within the enumerated strike ground\\nof failure to state a claim. We have not addressed this issue,\\nbut the Eighth Circuit has rejected a similar argument,\\nholding that “[d]ismissals based on immunity are not among\\nthe types of dismissals listed as strikes in section 1915(g).”\\nCastillo-Alvarez v. Krukow, 768 F.3d 1219, 1220 (8th Cir.\\n2014) (per curiam). We agree.\\n\\n The language and structure of the PLRA make clear that\\nimmunity-based dismissals generally do not fall within\\n§ 1915(g). Section 1915 sets out the IFP application\\nprocedures and rules regarding payment of court fees in\\nsubsections (a) through (d). Under § 1915(e)(2), regardless\\nof the filing fee payment, “the court shall [sua sponte]\\ndismiss the case at any time” if the case:\\n\\n (i) is frivolous or malicious;\\n\\n (ii) fails to state a claim on which relief\\n may be granted; or\\n\\n (iii) seeks monetary relief against a\\n defendant who is immune from such\\n relief,\\n\\n28 U.S.C. § 1915(e)(2)(B) (emphasis added).\\n\\n Yet the strike provision, subsection (g), which follows\\nsubsection (e), omits the immunity language as a ground for\\na strike. “[W]here Congress includes particular language in\\none section of a statute but omits it in another section of the\\nsame Act, it is generally presumed that Congress acts\\n\\f10 HARRIS V. HARRIS\\n\\nintentionally and purposefully in the disparate inclusion or\\nexclusion.” Russello v. United States, 464 U.S. 16, 23\\n(1983) (quoting United States v. Wong Kim Bo, 472 F.2d\\n720, 722 (5th Cir. 1972)).\\n\\n The PLRA also expressly lists immunity-based\\ndismissals in three other sections, further demonstrating that\\nCongress intentionally excluded immunity from the strike\\nprovision. For example, in § 1915A, Congress implemented\\na screening procedure for civil claims made by incarcerated\\nplaintiffs against “a governmental entity or officer or\\nemployee of a governmental entity.” 28 U.S.C. § 1915A(a).\\nWhen listing the grounds for dismissal while screening a\\ncomplaint, Congress included the three enumerated strike\\ngrounds and then added “or [] seeks monetary relief from a\\ndefendant who is immune from such relief,” 28 U.S.C.\\n§ 1915A(b), mirroring the language in 28 U.S.C.\\n§ 1915(e)(2)(B) discussed above. Congress intended for the\\nscreening procedure to address immunity because it\\nexpressly listed immunity after frivolousness,\\nmaliciousness, and failure to state a claim. See also\\n42 U.S.C. § 1997e(c) (stating that the court may dismiss a\\ncase regarding prison conditions or in which administrative\\nremedies have not been exhausted if it is “frivolous,\\nmalicious, fails to state a claim upon which relief can be\\ngranted, or seeks monetary relief from a defendant who is\\nimmune from such relief” in subsections 1997e(c)(1) and\\n(2) (emphasis added)). In sum, Congress’ omission of\\nimmunity-based dismissal from the strike provision in\\n§ 1915(g) evidences its intent generally not to include this\\ndismissal ground as a strike.\\n\\n To be sure, there are rare cases where an affirmative\\ndefense, such as immunity, may be so clear on the face of\\nthe complaint that dismissal may qualify as a strike for\\n\\f HARRIS V. HARRIS 11\\n\\nfailure to state a claim. See, e.g., El-Shaddai, 833 F.3d\\nat 1044 (recognizing the rarity of such dismissals and\\ndeclining to find a strike when the affirmative defense of\\nadministrative exhaustion was not clear on the face of the\\ncomplaint); Albino v. Baca, 747 F.3d 1162, 1169 (9th Cir.\\n2014) (en banc) (discussing the “rare cases” when failure to\\nexhaust is clear from the face of the complaint); Hafed v.\\nFed. Bureau of Prisons, 635 F.3d 1172, 1178 (10th Cir.\\n2011) (holding that a prior dismissal on immunity grounds\\nwas a strike where the prisoner “affirmatively asserted facts”\\ndemonstrating immunity (emphasis in original)), abrogated\\non other grounds as recognized by Carr v. Zwally, 760\\nF. App’x 550, 558 (10th Cir. 2019). There are also cases\\nwhere an affirmative defense like immunity is so obvious\\nthat the suit is frivolous, and dismissal counts as a strike. See\\nMills v. Fischer, 645 F.3d 176, 177 (2d Cir. 2011) (holding\\nthat a dismissal for judicial immunity was a strike because\\nthe claim was frivolous). But these are exceptional cases\\nwhere the affirmative defense is readily apparent without\\nresort to any additional information outside the four corners\\nof the complaint. Such will rarely be the case with\\nimmunity-based defenses. See, e.g., Milstein v. Cooley, 257\\nF.3d 1004 (9th Cir. 2001) (discussing at length the\\ncomplexities of prosecutorial immunity within the context of\\na Rule 12(b)(6) dismissal). And here, that is decidedly not\\nthe case. The district court moved beyond the Nielsen\\ncomplaint not only to consider the defendants’ roles, titles,\\nand actions, but to grapple with the legal analysis entitling\\nthe defendants to quasi-judicial immunity. Because the\\ncourt dismissed Harris’s Nielsen complaint on immunity\\ngrounds rather than the grounds that it was frivolous,\\nmalicious, or failed to state a claim, we conclude that Nielsen\\nis not a strike under § 1915(g).\\n\\f12 HARRIS V. HARRIS\\n\\n IV.\\n\\n The “denial of [IFP] status effectively, if not physically,\\ndenies many indigent prisoners access to the courts.”\\nSimone Schonenberger, Access Denied: The Prison\\nLitigation Reform Act, 86 Ky. L.J. 457, 474 (1998). In\\n§ 1915(g), Congress said what it meant, and we will construe\\nits language strictly and narrowly. “Our task is to give effect\\nto the will of Congress, and where its will has been expressed\\nin reasonably plain terms, that language must ordinarily be\\nregarded as conclusive.” Griffin v. Oceanic Contractors,\\nInc., 458 U.S. 564, 570 (1982) (internal quotation marks\\nomitted) (quoting Consumer Prod. Safety Comm’n v. GTE\\nSylvania, Inc., 447 U.S. 102, 108 (1980)). Unless an\\nincarcerated litigant has accrued three strikes on grounds\\nplainly enumerated in § 1915(g), she is entitled to IFP status.\\nDismissals for supplemental jurisdiction, failure to serve,\\nand quasi-judicial immunity are not grounds giving rise to\\nstrikes under § 1915(g) of the PLRA. With only two strikes\\nfrom prior litigation, Harris may proceed in this suit with IFP\\nstatus as long as he is otherwise eligible for IFP status.\\n\\n REVERSED AND REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4089,"cells":{"id":{"kind":"number","value":4654478,"string":"4,654,478"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Hooman Panah v. Kevin Chappell"},"decision_date":{"kind":"string","value":"2019-08-21"},"docket_number":{"kind":"string","value":"13-99010"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Kim McLane Wardlaw, Jacqueline H. Nguyen, and John B. Owens, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"OWENS, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nHOOMAN ASHKAN PANAH, No. 13-99010\\n Petitioner-Appellant,\\n D.C. No.\\n v. 2:05-cv-07606-\\n RGK\\nKEVIN CHAPPELL, Warden of\\nCalifornia State Prison at San\\nQuentin, OPINION\\n Respondent-Appellee.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n R. Gary Klausner, District Judge, Presiding\\n\\n Argued and Submitted June 26, 2019\\n Pasadena, California\\n\\n Filed August 21, 2019\\n\\nBefore: Kim McLane Wardlaw, Jacqueline H. Nguyen,\\n and John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Owens\\n\\f2 PANAH V. CHAPPELL\\n\\n SUMMARY *\\n\\n\\n Habeas Corpus\\n\\n The panel affirmed the district court’s denial of Hooman\\nPanah’s habeas corpus petition challenging his State of\\nCalifornia conviction and sentence for the first-degree\\nmurder and sexual assault of an eight-year-old girl.\\n\\n The district court granted a certificate of appealability as\\nto Panah’s claim brought pursuant to Napue v. Illinois, 360\\nU.S. 264 (1959), in which Panah, relying on post-conviction\\nDNA reports, contended that he was prejudiced by the\\nState’s presentation of serology testimony which, he argued,\\nthe State knew was false and misleading. The panel held that\\nthe California Supreme Court reasonably rejected this claim.\\nThe panel held that even assuming there was no reasonable\\nbasis for the state court to deny the claim as to the first two\\nNapue requirements – that the testimony was false or\\nmisleading, and that the State knew or should have known\\nthat – the panel could not say that it would be unreasonable\\nto conclude that the testimony did not satisfy the third\\nrequirement – materiality. Observing that even setting aside\\nthe serology testimony, the case against Panah was\\ndevastating, the panel held that the California Supreme\\nCourt would not have erred in finding no reasonable\\nlikelihood that the testimony could have affected the verdict.\\n\\n The panel expanded the certificate of appealability to\\nencompass Panah’s claim that his trial counsel rendered\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f PANAH V. CHAPPELL 3\\n\\nineffective assistance by failing to conduct a reasonable\\ninvestigation and therefore not rebutting the State’s serology\\nand pathology evidence. The panel expressed concern with\\ncounsel’s lack of pre-trial investigation, but held that even\\nassuming counsel’s performance was deficient, it could not\\nsay – in light of the overwhelming evidence of Panah’s guilt\\nand the deference owed the state court judgment – that the\\nCalifornia Supreme Court would have erred in finding no\\nreasonable probability that, but for counsel’s unprofessional\\nerrors, the result of the proceeding would have been\\ndifferent.\\n\\n\\n COUNSEL\\n\\nJoseph A. Trigilio (argued), Mark R. Drozdowski, and Susel\\nB. Carrillo-Orellana, Deputy Federal Public Defenders;\\nHilary Potashner, Federal Public Defender; Office of the\\nFederal Public Defender, Los Angeles, California; Firdaus\\nF. Dordi, Los Angeles, California; for Petitioner-Appellant.\\n\\nToni R. Johns Estaville (argued), Ana R. Duarte, A. Scott\\nHayward, and Dana M. Ali, Deputy Attorneys General;\\nLance E. Winters, Senior Assistant Attorney General;\\nGerald A. Engler, Chief Assistant Attorney General; Xavier\\nBecerra, Attorney General; Office of the Attorney General,\\nLos Angeles, California; for Respondent-Appellee.\\n\\f4 PANAH V. CHAPPELL\\n\\n OPINION\\n\\nOWENS, Circuit Judge:\\n\\n California state prisoner Hooman Panah appeals from\\nthe district court’s denial of his habeas corpus petition\\nchallenging his conviction and sentence for the first-degree\\nmurder and sexual assault of eight-year-old Nicole Parker.\\nWe have jurisdiction under 28 U.S.C. § 1291, and we affirm.\\n\\n\\n\\n\\n In the early afternoon of November 20, 1993, Parker\\nwent missing from her father’s apartment complex, where\\nPanah also lived with his mother. While searching for her in\\nthe complex, Parker’s father knocked on Panah’s door and\\nasked if Panah had seen her. Panah responded something\\nlike, “oh, is she missing.” He then offered to help Parker’s\\nfather look for her, “persistent[ly]” suggesting they search\\noutside the apartment complex. Soon after, the police\\narrived and conducted a door-to-door search for Parker,\\nincluding Panah’s apartment. The police did not find Parker\\nor any clues as to her whereabouts.\\n\\n That day, Panah reported to work in the mid-afternoon.\\nAround 5:30 pm, his mother, who was with two police\\nofficers, called Panah. The officers asked him if he knew\\nParker or had seen her that day. He responded that he knew\\nher only “vaguely” and denied having seen her that day.\\nShortly after the officers’ inquiry – hours before his shift\\nended – Panah left work without telling anyone. He later\\ncalled his manager to say that he would not return “because\\nsome people that he knew [were] trying to get him in trouble\\nand would [his manager] please inform his mother to get out\\n\\f PANAH V. CHAPPELL 5\\n\\nof town.” Panah also paged his co-worker, Rauni Campbell,\\nasking for help. He told her that he “d[id] something very\\nbad,” “so big” that she would find out.\\n\\n The next morning, Panah showed up without warning at\\nCampbell’s apartment. His wrists were cut, and he requested\\nsleeping pills, which she helped him buy. Campbell asked\\nPanah if he had anything to do with “the little girl that was\\nmissing from his apartment complex.” He said yes. She\\nthen asked him whether Parker was still alive. He said no.\\nAt this point, Campbell surreptitiously called the police.\\nWhen they arrived, Panah tried to evade arrest but was\\neventually caught and taken to the hospital. At the hospital,\\nunder the influence of drugs and reportedly in a psychotic\\nstate, Panah told police, in response to questions about\\nParker, that he “liked her very much, even [to] carry her\\nskeleton remains around.”\\n\\n Later that evening, the police, armed with a search\\nwarrant, returned to Panah’s apartment. In his bedroom\\ncloset, they found Parker’s naked body, wrapped in a\\nbedsheet and stuffed in a suitcase. The police then gathered\\nevidence from Panah’s bedroom, including examining his\\nbed and Parker’s body for evidence of sexual assault.\\n\\n Panah was indicted on charges of first-degree murder\\nwith special circumstances alleging that the murder occurred\\nduring a kidnapping, sodomy, lewd acts on a person under\\nfourteen years old, and oral copulation of a person under\\nfourteen years old. He was also charged with the substantive\\ncounts of kidnapping, sodomy by force, lewd acts on a\\nperson under fourteen years old, penetration of genital or\\nanal openings by a foreign object with a person under\\nfourteen years old, and oral copulation of a person under\\nfourteen years old. Panah pled not guilty.\\n\\f6 PANAH V. CHAPPELL\\n\\n\\n\\n Panah was initially represented by a family friend,\\nSyamak Shafi-Nia, who had limited criminal law\\nexperience. But prior to trial, the court appointed Robert\\nSheahen, a veteran criminal lawyer, as lead counsel, and\\nallowed Shafi-Nia to stay on as second counsel. Sheahen\\nhad requested this appointment, promising the court that he\\nwould facilitate a settlement, which would “save[] a great\\ndeal of time and the taxpayers would be saved a great deal\\nof money” by avoiding “an extremely costly trial.”\\n\\n In July 1994, several months before trial, the State\\nnotified the court and defense that it had ordered DNA\\ntesting on evidence found at the crime scene. While\\nawaiting the test results in September, the court urged\\nSheahen to “find a DNA expert to assist you” and “see if\\nthere’s any basis for questioning the results.” In October,\\ntwo months before trial, the State shared the DNA test results\\nwith the defense. Again, the court advised Sheahen to retain\\nan expert, to which Sheahen responded, “That will be taken\\ncare of.”\\n\\n However, as trial approached, the State decided not to\\nintroduce the DNA evidence. The court pressed defense\\ncounsel why he had not yet independently tested the DNA.\\nCounsel explained that doing so “would put us in the\\nposition of confirming the prosecution results,” and that he\\ninstead planned to argue that the State’s “failure to do DNA\\ntesting should be held against” them. The court approved of\\nthis strategy, calling counsel’s “tactics . . . very sound in this\\nparticular case.”\\n\\f PANAH V. CHAPPELL 7\\n\\n\\n\\n Panah’s trial began on December 5, 1994. With jury\\nselection set to begin, Sheahen notified the court that Shafi-\\nNia was no longer able to serve as second counsel but did\\nnot request a continuance. Accordingly, Panah began jury\\nselection with just one lawyer. Shortly after, the court\\nappointed new second counsel to replace Shafi-Nia, but\\nsecond counsel was required to familiarize himself with the\\ncase during trial.\\n\\n\\n\\n The State’s theory was felony murder. It emphasized the\\nabundance of circumstantial evidence against Panah and\\nfocused on “Parker’s body bloody and battered,” which was\\n“tied up in a sheet inside a zipped suitcase” in Panah’s closet.\\nIt also highlighted Panah’s incriminating behavior soon after\\nParker went missing, including that Panah was “anxious” to\\nencourage Parker’s father to search outside the apartment\\ncomplex; “had fled” work hours before his shift ended after\\nreceiving a call about Parker from his mother and police; and\\nmade numerous admissions about his involvement in\\nParker’s murder. Panah’s manager and Campbell testified\\nabout his statements on the day of and after Parker’s\\ndisappearance.\\n\\n The State also presented forensic evidence as part of its\\ncase-in-chief.\\n\\n\\n\\n The coroner, Dr. Heuser, testified that Parker’s physical\\ninjuries occurred premortem. Describing the violent nature\\nof the assault, she explained that Parker suffered “blunt\\nforce” injuries, including bruising on her forehead, eye,\\n\\f8 PANAH V. CHAPPELL\\n\\nneck, lip, arms, and buttocks; scratches on her inner thighs;\\nand brain swelling. Dr. Heuser also testified to Parker’s\\nsexual assault injuries. Parker had bruising, as well as signs\\nof bleeding and tears, in the vaginal and rectal areas. Her\\nvaginal opening was “very widely” open – most likely\\nconsistent with digital penetration. Her anal opening was\\nalso “widely open and very lax looking,” “consistent with\\nthe insertion of a penis into her rectum.”\\n\\n Dr. Heuser also testified that Parker’s death was due to\\n“traumatic injuries,” either the result of “manual\\nstrangulation” or force to Parker’s rectum.\\n\\n\\n\\n Serologist William Moore testified about stains found on\\nthree items in Panah’s bedroom: (1) the bedsheet Parker was\\nwrapped in, (2) a robe found on Panah’s bed, 1 and (3) a\\ntissue from a wastebasket in Panah’s bathroom.\\nPreliminarily, Moore noted that Panah’s blood type was B\\nand Parker’s was A. Because his testing discovered that\\nstains on each item contained a mixture of A and B antigens,\\nMoore posited that this was consistent with a mixture of\\nParker’s and Panah’s bodily fluids and thus sexual contact\\nbetween them.\\n\\n As to the bedsheet, Moore described the stains as\\n“indicative of a mixture of physiological fluids” – blood,\\nsemen, and amylase (a constituent of saliva and other bodily\\nfluids) – that included both A and B antigens. Because the\\nbloodstain was “consistent” with Parker’s type A blood, he\\nsurmised the other stains were consistent with semen from\\nPanah and saliva from Parker. He also noted that the pattern\\n\\n 1\\n The district court’s opinion calls the robe a kimono.\\n\\f PANAH V. CHAPPELL 9\\n\\nof the stains “could . . . be consistent with the spewing of\\nsemen across the bed sheet.”\\n\\n Moore similarly testified that the tissue “bore semen\\nstains, and high amylase activity,” likely from a mix of\\nParker’s and Panah’s bodily fluids. Again, he remarked that\\nthis stain “could be consistent with the product of an oral\\ncopulation.” Lastly, he testified that Panah’s robe had a\\nlarge stain with a mix of A and B antigens. Because the\\nrobe’s bloodstain was consistent with Parker, he\\nhypothesized that the B antigens came from Panah’s saliva.\\n\\n Moore also briefly testified about evidence collected\\nfrom the sexual assault examination, although he did not\\nconduct it. He acknowledged that the oral and anal swabs\\nhad not produced any signs of semen, nor was “the presence\\nof semen conclusively” found anywhere on Parker’s body.\\n\\n On cross-examination, Moore admitted that he could not\\n“establish any certainty” that either Parker or Panah was a\\ncontributor to the stains because of the relatively high\\nstatistical frequency of the A and B antigens matching other\\npeople. Moore further testified that his theory would not\\nbear out if the A and B antigens on the three items came from\\none person with AB blood type. 2\\n\\n\\n\\n The defense consisted of testimony from Dr. John\\nPalmer, the emergency room doctor who treated Panah the\\nday after Parker went missing, and several character\\nwitnesses. Dr. Palmer reported that Panah was “acutely\\n 2\\n After the State presented its case, the court granted the defense’s\\nmotion for acquittal on the substantive charges of kidnapping and the\\nspecial circumstance allegation of kidnapping.\\n\\f10 PANAH V. CHAPPELL\\n\\npsychotic” and suicidal when brought into the emergency\\nroom that day.\\n\\n\\n\\n The State summarized its evidence: “We have blood\\ntyping that matches. We have the body in his suitcase in his\\ncloset, and we have statements he makes that show\\nknowledge before the body was found. We have his\\ninvolvement in the crime clearly established.” The State\\nfocused on where Parker’s body was found: “You have a\\nbody in his closet, in his suitcase. There isn’t a whole lot\\nmore you need to do after that in terms of looking and\\ninvestigating outside of the obvious, which is that Mr. Panah\\nis the person involved.” It also focused on Panah’s\\nstatements after Parker went missing: “Those aren’t crazed\\nremarks. Those are the remarks of an individual who is\\ntelling exactly what happened.” The State then reminded the\\njury of Moore’s testimony, arguing it was “not a harebrained\\nprosecution theory.” It particularly used his testimony to\\nprove the alleged oral copulation:\\n\\n We think the evidence that was presented to\\n you is very consistent with the fact that he\\n ejaculated in her mouth, that he allowed her\\n to spit it out in a Kleenex, because we have\\n the evidence of semen of his blood type, high\\n amylase content, indicating a saliva which\\n matches her blood type on the Kleenex, as\\n well as having a spattering on the bed sheet\\n of a mixture of semen and saliva – again the\\n high amylase indicating saliva – of his type B\\n and her type A.\\n\\nIt also said that “the one possible inference that can be\\ndrawn” from Moore’s testimony about the robe is that the B\\n\\f PANAH V. CHAPPELL 11\\n\\nantigens came from “the saliva of the defendant” during the\\nsodomy.\\n\\n Acknowledging that none of this was “conclusive\\nevidence,” the State argued that, “when taken with\\neverything else[, this] would indicate that there had been an\\nact of oral copulation, that there was ejaculate in Nicole\\nParker’s mouth.” The State also responded to the defense’s\\nassertion that its case was weak because of the lack of DNA,\\nclaiming that DNA testing is “usually ordered in a situation\\nwhere you don’t have other types of proof available. In this\\nsituation we have the proof available.”\\n\\n In its closing, the defense questioned the reliability of the\\nserology evidence, calling Moore’s theory “hogwash,” and\\ninsisted that the stains proved nothing. Rather, counsel\\nhighlighted the lack of DNA evidence, which “could tell us\\nwho’s the source of this stuff . . . [and] whether it, in fact,\\ncould be traced to the deceased or whether it could be traced\\nto any number of other people.”\\n\\n\\n\\n The jury convicted Panah of first-degree murder and the\\nother felonies. The jury also found true the special\\ncircumstance allegations that the murder was committed\\nwhile engaged in the crime of sodomy and lewd acts on a\\nperson under the age of fourteen. The jury did not find true\\nthe special circumstance allegation that the murder occurred\\nin the commission of oral copulation. After the penalty\\nphase, the jury returned a death sentence.\\n\\n\\n\\n On March 14, 2005, the California Supreme Court\\naffirmed Panah’s conviction and sentence, People v. Panah,\\n\\f12 PANAH V. CHAPPELL\\n\\n107 P.3d 790 (Cal. 2005), and the United States Supreme\\nCourt subsequently denied certiorari, Panah v. California,\\n546 U.S. 1216 (2006). About a year later, the California\\nSupreme Court summarily denied Panah’s first habeas\\npetition. After filing a protective habeas petition in the\\nDistrict Court for the Central District of California, Panah\\nfiled a second state habeas petition and a first amended\\npetition in the district court. The district court stayed\\nproceedings during the pendency of Panah’s state habeas\\nproceedings. On March 16, 2011, the California Supreme\\nCourt again summarily denied Panah’s second state habeas\\npetition, and the district court lifted the stay on Panah’s\\nfederal habeas proceedings.\\n\\n In his habeas petitions, Panah provided new evidence,\\nincluding two reports detailing post-conviction DNA testing\\non the stains that Moore testified about at trial. The first\\nreport (“Calandro Report”), prepared in 2004, disagreed\\nwith much of Moore’s testimony. But because the Calandro\\nReport yielded several inconclusive results, additional\\ntesting was conducted two years later, leading to the second\\nreport (“Inman Report”). Both reports doubted the\\nfoundation of Moore’s mixture theory. The Calandro Report\\ncalled “Mr. Moore’s approach . . . biased and indefensible,”\\nand the Inman Report wrote that “[n]o biological evidence\\nexists to support the hypothesis that a mixture of biological\\nfluids from Mr. Panah and Ms. Parker was present on the\\ntissue, bedsheet, or kimono.”\\n\\n More specifically, both reports “contradict[ed]” Moore’s\\ntestimony about the tissue. While Moore testified that Panah\\nand Parker were both possible contributors to the tissue stain,\\nthe reports eliminated any possibility that Parker was a\\nsource. On appeal to this court, the State concedes that\\nMoore’s tissue testimony was false.\\n\\f PANAH V. CHAPPELL 13\\n\\n However, the post-conviction testing produced\\ninconclusive results regarding the bedsheet and robe stains.\\nNeither report could definitively eliminate Parker as a\\ncontributor to several stains on the bedsheet. While they\\nconclusively found that two stains were consistent only with\\nPanah, they could not conclusively rule Parker out as a\\ncontributor to the three other bedsheet stains. Although this\\nleft open the possibility that Moore’s mixture theory was\\ncorrect, the reports opined that this at minimum refuted his\\nassumption that “Ms. Parker ‘spit out’ ejaculate onto the bed\\nsheet” because one would then expect to “detect Ms.\\nParker’s DNA in significant quantities on the bed sheet.” As\\nfor the robe, the reports agreed with Moore that the\\nbloodstain was consistent with Parker’s blood type. But,\\nunlike Moore’s serology testimony, they found no trace of\\nPanah’s DNA on the robe.\\n\\n Panah’s habeas petitions also included declarations from\\nhis three trial lawyers. Each declaration acknowledged there\\nhad been almost no pre-trial investigation and only a limited\\npenalty-phase investigation, nor were any experts retained to\\nindependently analyze the State’s serology and pathology\\nevidence. Instead, “all of [defense counsel’s] efforts had\\ngone into the aborted settlement.”\\n\\n On November 14, 2013, the district court denied Panah’s\\npetition. As for Panah’s Napue v. Illinois, 360 U.S. 264\\n(1959), claim, the district court held that, even if post-\\nconviction DNA testing rendered a portion of Moore’s\\ntestimony false, the California Supreme Court could have\\nreasonably concluded that it did not render all of the\\ntestimony false and that his testimony was immaterial in\\nlight of the other evidence. In this discussion, the district\\ncourt also rejected Panah’s claim of ineffective assistance for\\nfailure to investigate the State’s forensic evidence because\\n\\f14 PANAH V. CHAPPELL\\n\\nthe California Supreme Court reasonably could have\\nconcluded that Panah was not prejudiced.\\n\\n The district court granted a certificate of appealability\\n(“COA”) for Panah’s Napue claim, discussed below in\\nsection III. On appeal, Panah has raised a number of\\nuncertified issues in his opening brief, which we treat as a\\nrequest to expand the COA. 9th Cir. R. 22-1(e). After\\nasking the State to respond to several of the uncertified\\nissues, we expand the COA to encompass Panah’s guilt-\\nphase ineffective assistance claim, addressed below in\\nsection IV, but deny Panah’s request to expand the COA as\\nto the other uncertified claims. We evaluate Panah’s two\\ncertified claims in turn.\\n\\n\\n\\n We review the district court’s denial of habeas relief de\\nnovo. Lewis v. Mayle, 391 F.3d 989, 995 (9th Cir. 2004).\\n\\n Because Panah filed his federal habeas petition after\\nApril 24, 1996, it is subject to the Antiterrorism and\\nEffective Death Penalty Act of 1996 (“AEDPA”). See\\nMurray v. Schriro, 745 F.3d 984, 996 (9th Cir. 2014). Under\\nAEDPA, we may grant relief only if the adjudication\\n“resulted in a decision that was contrary to, or involved an\\nunreasonable application of, clearly established Federal law,\\nas determined by the Supreme Court of the United States,”\\n28 U.S.C. § 2254(d)(1), or “was based on an unreasonable\\ndetermination of the facts in light of the evidence presented\\nin the State court proceeding,” id. § 2254(d)(2).\\n\\n Although we typically “look through” a summary\\ndisposition to the last reasoned state court decision, Ylst v.\\nNunnemaker, 501 U.S. 797, 806 (1991), here there is no\\nreasoned state court decision addressing either certified\\n\\f PANAH V. CHAPPELL 15\\n\\nclaim. Therefore, we independently review the record to\\ndetermine whether the California Supreme Court had any\\nreasonable basis to deny Panah relief. See Reis-Campos v.\\nBiter, 832 F.3d 968, 974 (9th Cir. 2016) (citing Harrington\\nv. Richter, 562 U.S. 86, 102 (2011)).\\n\\n\\n\\n Relying on the two post-conviction DNA reports, Panah\\ncontends that he was prejudiced by the State’s presentation\\nof Moore’s serology testimony, which he argues the State\\nknew was false or misleading.\\n\\n In Napue, the Supreme Court held “that a conviction\\nobtained through use of false evidence, known to be such by\\nrepresentatives of the State, must fall under the Fourteenth\\nAmendment.” 360 U.S. at 269. Nonetheless, a Napue claim\\nsucceeds only if three elements are satisfied. See United\\nStates v. Zuno-Arce, 339 F.3d 886, 889 (9th Cir. 2003).\\nFirst, the testimony or evidence in question must have been\\nfalse or misleading. See id.; see also Alcorta v. Texas,\\n355 U.S. 28, 31 (1957) (considering whether “testimony,\\ntaken as a whole, gave the jury [a] false impression”).\\nSecond, the State must have known or should have known\\nthat it was false or misleading. See Zuno-Arce, 339 F.3d\\nat 889; see also Maxwell v. Roe, 628 F.3d 486, 506 (9th Cir.\\n2010) (“[E]ven false evidence presented in good faith . . .\\nhardly comports with fundamental fairness.” (quoting\\nKillian v. Poole, 282 F.3d 1204, 1209 (9th Cir. 2002))). And\\nthird, because “Napue does not create a ‘per se rule of\\nreversal,’” the testimony or evidence in question must be\\nmaterial. Sivak v. Hardison, 658 F.3d 898, 912 (9th Cir.\\n2011) (quoting Jackson v. Brown, 513 F.3d 1057, 1076 (9th\\nCir. 2008)); see also Zuno-Arce, 339 F.3d at 889.\\n\\f16 PANAH V. CHAPPELL\\n\\n Materiality under Napue requires a “lesser showing of\\nharm . . . than under ordinary harmless error review.” Dow\\nv. Virga, 729 F.3d 1041, 1048 (9th Cir. 2013). But, after\\nweighing the effect of alleged Napue violations collectively,\\nsee Phillips v. Ornoski, 673 F.3d 1168, 1189 (9th Cir. 2012),\\nthere still needs to be a “reasonable likelihood that the false\\ntestimony could have affected the judgment of the jury.”\\nHayes v. Brown, 399 F.3d 972, 985 (9th Cir. 2005) (en banc)\\n(quoting Belmontes v. Woodford, 350 F.3d 861, 881 (9th Cir.\\n2003)). Thus, a Napue claim fails if, absent the false\\ntestimony or evidence, the petitioner still “received a fair\\ntrial, understood as a trial resulting in a verdict worthy of\\nconfidence.” Hayes, 399 F.3d at 984 (quoting Hall v. Dir.\\nof Corr., 343 F.3d 976, 984 (9th Cir. 2003) (per curiam)).\\n\\n Even if we assume there was no reasonable basis for the\\nstate court to deny Panah’s claim as to the first two Napue\\nrequirements, we cannot say that it would be unreasonable\\nto conclude that Moore’s testimony was immaterial. See\\nTowery v. Schriro, 641 F.3d 300, 308 (9th Cir. 2010) (not\\nreviewing all three requirements because petitioner’s\\n“argument fails at the second Napue prong”). The State\\npresented a powerful case of Panah’s guilt, with substantial\\nevidence linking him to Parker’s murder and sexual assault.\\nMoore’s testimony was just one – and not a crucial – piece\\nof that presentation. Because the “verdict” is still reasonably\\n“worthy of confidence,” Phillips, 673 F.3d at 1189 (quoting\\nSivak, 658 F.3d at 912), we hold that the California Supreme\\nCourt would not have erred in finding no “reasonable\\nlikelihood that [Moore’s testimony] could have affected the\\njudgment of the jury.” Hayes, 399 F.3d at 985 (quoting\\nBelmontes, 350 F.3d at 881); see also Phillips, 673 F.3d\\nat 1190 (“[T]he prosecution’s Napue violations, although\\n‘pernicious’ and ‘reprehensible,’ were not material to\\n\\f PANAH V. CHAPPELL 17\\n\\n[petitioner’s] conviction of first-degree murder.” (quoting\\nHayes, 399 F.3d at 981)).\\n\\n Even setting aside Moore’s testimony, the case against\\nPanah was devastating. Parker’s naked body was found in a\\nsuitcase in Panah’s bedroom closet. Blood stains matching\\nParker’s blood type – according to both Moore and the post-\\nconviction reports – were found on Panah’s robe. Moreover,\\nPanah’s behavior on the day of and after Parker went missing\\nwas highly suspicious. Hours after she disappeared, Panah\\ntried to divert her father away from where her body was later\\nfound. Then, after the police called Panah at work to ask if\\nhe had seen Parker, he left without explanation, later telling\\nhis manager and Campbell that he was in serious trouble.\\nThe next morning, Panah even confided in Campbell that he\\nwas involved in Parker’s death. See Arizona v. Fulminante,\\n499 U.S. 279, 296 (1991) (“A confession is like no other\\nevidence. Indeed, ‘the defendant’s own confession is\\nprobably the most probative and damaging evidence that can\\nbe admitted against him.’” (quoting Bruton v. United States,\\n391 U.S. 123, 139 (1968) (White, J., dissenting))). And then,\\nparticularly incriminating in light of his statements, Panah\\nattempted suicide the morning after Parker’s disappearance.\\n\\n The jury then heard Dr. Heuser’s impactful testimony\\nabout the crime itself. She described Parker’s extensive\\ninjuries from the sexual assault, including significant trauma\\nto Parker’s vaginal and rectal areas, indicative of digital and\\npenile penetration. In addition, Dr. Heuser described the\\nviolent nature of the assault: Parker was hit with “blunt\\nforce,” consistent with her “head striking . . . a wall or a\\nfloor” or being hit with a “fist.” As a result, Parker’s brain\\nwas swollen from this “significant impact,” and she had\\nbruises – some caused by “manual strangulation” – along her\\nface, neck, arms, buttocks, and legs, and scratches on her\\n\\f18 PANAH V. CHAPPELL\\n\\nlegs. Dr. Heuser cited strangulation and sodomy as “the\\nmost lethal injuries.”\\n\\n Although Dr. Heuser’s testimony did not directly\\nimplicate Panah, it was nonetheless critical. It established\\nthat a crime – and a brutal one – took place. When added to\\nthe evidence the jury heard about where Parker’s body was\\nfound and Panah’s statements, it was more than sufficient for\\nthe jury to render a guilty verdict. And, while Panah\\ncontends that Moore’s testimony was prejudicial because of\\nits at-times graphic descriptions, particularly of oral\\ncopulation, Dr. Heuser’s testimony offered an even more\\ngraphic and detailed description of the entire sexual assault\\nand murder. As the State itself said, Dr. Heuser’s testimony\\nis “probably the most telling evidence of what happened.”\\n\\n The state court also reasonably could have found\\nMoore’s testimony to be an immaterial part of the State’s\\ncase because it offered the jury, at most, hypotheticals and\\nwavering findings. Unlike Dr. Heuser’s straightforward\\ntestimony about Parker’s injuries, Moore acknowledged that\\nhis findings rested on a number of assumptions, such as that\\nthe A and B antigens came from two people rather than one\\nwith AB blood type. And even if he was correct that the A\\nand B antigens came from two people, Moore neither could\\ndefinitively say they came from Parker and Panah, nor could\\nhe even narrow the pool of possible matches to less than\\nhundreds of thousands of people. He said frankly on the\\nstand: “I cannot establish any certainty based on\\nconventional serology. I can only demonstrate consistency.”\\nFor this reason, Moore’s testimony was couched in\\ninconclusive terms, such as “could have originated from” or\\nwere “consistent” with. The State even acknowledged this\\nweakness in Moore’s findings in closing argument:\\n\\f PANAH V. CHAPPELL 19\\n\\n Now the question is, did a person with AB\\n blood leave . . . body fluids such as blood,\\n semen[,] and saliva, on the sheets, on the\\n toilet paper, on the robe. That is one\\n interpretation. The other interpretation, of\\n course, is that you have two separate people,\\n one of whom has type A, and one has type B.\\n\\n Therefore, at best, Moore’s testimony required the jury\\nto draw its own inferences. This was not the case of\\nimpactful expert testimony telling the jury that there was a\\none-in-a-million chance the evidence matched anyone but\\nPanah. Rather, any effect Moore’s findings may have had\\non the jury – which was reasonably none – was fully\\ndependent on the other weighty evidence presented by the\\nState. For instance, without having found Parker’s body in\\nPanah’s bedroom, no juror could have reasonably inferred\\nthat Parker was the A antigen contributor.\\n\\n The jury’s verdict removes any lingering doubt about the\\nmateriality of Moore’s testimony. If the State needed\\nMoore’s testimony at all, it was to prove the special\\ncircumstance allegation and substantive charge of oral\\ncopulation. In its closing, for instance, the State referred to\\nMoore’s findings of mixed bodily fluids on the tissue and\\nbedsheet to prove this sexual act. Yet, the jury did not find\\ntrue the special circumstance allegation that Parker’s murder\\nwas committed while in the commission of oral copulation.\\nAlthough the jury did find Panah guilty of the felony of oral\\ncopulation, the verdict is still telling. A reasonable\\ninterpretation of the jury’s rejection of this special\\ncircumstance is that the jury was not entirely persuaded by\\nMoore’s mixture theory. In contrast, this highlights the\\neffectiveness of Dr. Heuser’s testimony. The State relied on\\nher findings to prove the special circumstance allegations\\n\\f20 PANAH V. CHAPPELL\\n\\nand substantive charges of sodomy and lewd acts, which the\\njury found to be true.\\n\\n Still, Panah contends that Moore’s testimony was\\nmaterial because it was the only evidence identifying him as\\nthe perpetrator. Although creative, this argument makes\\nlittle headway. It ignores the substantial evidence tying\\nPanah to Parker’s murder and sexual assault. This was not a\\ncase where the police had no leads on a suspect. Nor was it\\na case where the prosecution needed Moore’s serology\\nevidence to place Panah at the crime scene. Rather, as the\\nState emphasized in its closing: “You have a body in his\\ncloset, in his suitcase. There isn’t a whole lot more you need\\nto do after that in terms of looking and investigating outside\\nof the obvious, which is that Mr. Panah is the person\\ninvolved.” And, as previously mentioned, Panah’s own\\nadmissions linked him to the assault. Thus, even without\\nMoore’s testimony, the State had no difficulty proving\\nidentity.\\n\\n For these reasons, Panah’s reliance on Miller v. Pate,\\n386 U.S. 1 (1967), is unavailing. In Miller, the petitioner\\nwas also charged with the murder and sexual assault of a\\nyoung girl without any eyewitnesses to the crime. Id. at 2–\\n3. The State’s only evidence linking petitioner to the victim\\nwas male underwear, which an expert said had blood stains\\non it matching the victim, found near the crime scene. Id.\\nat 3–4. Post-conviction testing proved the underwear stains\\nwere paint, not blood, and that the State had known this at\\ntrial. Id. at 5–6. The Supreme Court reversed petitioner’s\\nconviction because, while the State successfully used the\\nunderwear as “an important link in the chain of\\ncircumstantial evidence against the petitioner,” in reality it\\nwas “virtually valueless as evidence.” Id. at 4, 6.\\n\\f PANAH V. CHAPPELL 21\\n\\n Miller is clearly distinguishable. First, although the\\nparties agree that the tissue testimony was false, post-\\nconviction testing did not render the majority of Moore’s\\ntestimony false. Neither post-conviction report conclusively\\nrefuted his findings as to the bedsheet or robe stains.\\nTherefore, unlike in Miller, the evidence in question here did\\nnot become “virtually valueless” to convict. Id. at 6.\\nSecond, Moore’s testimony was not the “important link” in\\nproving Panah’s guilt. Id. at 4. The State did not even need\\nMoore’s testimony to convict Panah. This case differs\\nsignificantly from Miller because here the state court could\\nreasonably rely on an abundance of other evidence to still\\nhave confidence in the conviction.\\n\\n Panah’s reliance on Alcorta, 355 U.S. 28, also is\\nmisplaced. There, the Court overturned petitioner’s\\nconviction because the false “testimony was seriously\\nprejudicial to petitioner” and “tended squarely to refute his\\nclaim.” Id. at 31. Had the false testimony been corrected,\\npetitioner’s defense would have been corroborated. But,\\nhere, even if Moore’s testimony had been corrected or\\nentirely excluded, the jury would not have heard a\\nsignificantly different presentation of evidence. At most,\\nalthough we think unlikely, the State’s case may have\\nbecome marginally weaker.\\n\\n Rather, this case is quite similar to Sivak, 658 F.3d 898,\\nin which we rejected a guilt-phase Napue claim because the\\ncourt had “full confidence that the jury would still have\\nconvicted.” Id. at 913. There, we held: “[E]ven if the jury\\ndisbelieved [the false testimony] entirely . . . there still is no\\n‘reasonable likelihood that the false testimony could have\\naffected the judgment of the jury.’” Id. at 914 (quoting\\nJackson, 513 F.3d at 1076). As in Sivak, we think the\\nCalifornia Supreme Court reasonably still could have had\\n\\f22 PANAH V. CHAPPELL\\n\\n“full confidence” that the jury would have returned the same\\nverdict even in absence of Moore’s testimony. Id. at 913.\\n\\n In sum, we do not think Moore’s testimony was critical\\nin convicting Panah. Excluding Moore’s testimony, the\\nState’s case was still devastating and largely unchallenged.\\nMoore’s testimony was certainly not “the centerpiece of the\\nprosecution’s case.” Hayes, 399 F.3d at 985. Rather, in light\\nof the overwhelming evidence against Panah and the jury’s\\nrejection of the oral copulation special circumstance, it is\\nreasonable to conclude that his testimony had essentially no\\neffect on the jury’s decision making. Cf. Dow, 729 F.3d\\nat 1049–50 (concluding that false testimony was material\\nbecause “[t]he evidence against [petitioner] was weak”);\\nMaxwell, 628 F.3d at 508 (holding that false testimony was\\nmaterial because it came from a “‘make-or-break’ witness\\nfor the State” and there was a “paucity of other evidence”);\\nHall, 343 F.3d at 984 (reversing under Napue “[i]n light of\\nthe already scant evidence on which the conviction was\\nbased”). As such, we conclude that the California Supreme\\nCourt reasonably rejected Panah’s Napue claim.\\n\\n\\n\\n We next turn to Panah’s guilt-phase ineffective\\nassistance claim. To prevail, Panah must show that his\\ncounsel’s performance “fell below an objective standard of\\nreasonableness” and “a reasonable probability that, but for\\ncounsel’s unprofessional errors, the result of the proceeding\\nwould have been different.” Strickland v. Washington,\\n466 U.S. 668, 688, 694 (1984). “The likelihood of a\\ndifferent result,” however, “must be substantial, not just\\nconceivable.” Richter, 562 U.S. at 112. This already\\nimposing standard becomes doubly difficult to satisfy once\\nAEDPA deference is tacked on. See Cullen v. Pinholster,\\n563 U.S. 170, 189 (2011) (“[T]he benchmark for judging any\\n\\f PANAH V. CHAPPELL 23\\n\\nclaim of ineffectiveness [under Strickland] must be whether\\ncounsel’s conduct so undermined the proper functioning of\\nthe adversarial process that the trial cannot be relied on as\\nhaving produced a just result.” (quoting Strickland, 466 U.S.\\nat 686)).\\n\\n Panah argues that his trial counsel was ineffective for\\nfailing to conduct a reasonable investigation and therefore\\nnot rebutting the State’s serology and pathology evidence.\\nIn support of his claim, Panah notes that his counsel never\\nretained an expert to independently analyze the pathology\\nand serology evidence or to testify at the guilt phase.\\nInstead, defense counsel essentially seemed to accept the\\nState’s evidence as true. In post-conviction proceedings,\\ncounsel acknowledged that his inordinate focus on\\nsettlement resulted in too little, if even any, pre-trial\\ninvestigation.\\n\\n While we are instructed to “avoid the temptation to\\nsecond-guess [counsel’s] performance,” Mayfield v.\\nWoodford, 270 F.3d 915, 927 (9th Cir. 2001) (en banc), we\\nare concerned with defense counsel’s lack of pre-trial\\ninvestigation. See Duncan v. Ornoski, 528 F.3d 1222, 1235\\n(9th Cir. 2008) (“[L]awyers [have] considerable discretion\\nto make strategic decisions about what to investigate, but\\nonly after those lawyers ‘have gathered sufficient evidence\\nupon which to base their tactical choices.’” (quoting\\nJennings v. Woodford, 290 F.3d 1006, 1014 (9th Cir.\\n2002))). But we “need not determine whether counsel’s\\nperformance was deficient before examining the prejudice\\nsuffered by the defendant as a result of the alleged\\ndeficiencies.” Strickland, 466 U.S. at 697. Here, even\\nassuming counsel’s performance was deficient, we cannot\\nsay – in light of the overwhelming evidence of Panah’s guilt\\nand the deference we owe the state court judgment – that the\\n\\f24 PANAH V. CHAPPELL\\n\\nCalifornia Supreme Court would have erred in finding no\\n“reasonable probability that, but for counsel’s\\nunprofessional errors, the result of the proceeding would\\nhave been different.” Id. at 694.\\n\\n After weighing counsel’s deficiencies cumulatively with\\n“the strength of the government’s case,” Rios v. Rocha,\\n299 F.3d 796, 808–09 (9th Cir. 2002) (quoting Eggleston v.\\nUnited States, 798 F.2d 374, 376 (9th Cir. 1986)), we believe\\nthe state court reasonably rejected Panah’s assertion that the\\ntrial’s outcome “would have been dramatically different”\\nhad counsel’s performance not been deficient. Our\\nreasoning on prejudice bears significant similarities to why\\nwe reject Panah’s Napue claim. We do not wish to harp on\\nwhat was detailed in the preceding section, but the State had\\na uniquely strong case against Panah. Parker’s body was\\nfound in his bedroom; Panah’s behavior the day of and after\\nher disappearance was incriminating; Panah admitted his\\nown involvement in her death; and Parker’s serious physical\\ninjuries, including to her genitalia, were well-documented.\\nDefense counsel even acknowledged that this evidence was\\nthe foundation of the State’s case: “The critical pieces of\\nevidence are obviously that the child’s body is found in Mr.\\nPanah’s closet, her naked body with a considerable amount\\nof blood. There is evidence . . . that the child was beaten.”\\n\\n It is, therefore, inconceivable, even had defense counsel\\nindependently investigated the serology and pathology\\nevidence, that the jury would have reached a different\\nverdict. See Williams v. Filson, 908 F.3d 546, 570 (9th Cir.\\n2018) (“We have long recognized . . . that ‘prejudice\\nresulting from ineffective assistance of counsel must be\\n“considered collectively, not item by item.”’” (quoting Doe\\nv. Ayers, 782 F.3d 425, 460 n.62 (9th Cir. 2015))). Even if\\nthe weaknesses in Moore’s and Dr. Heuser’s findings that\\n\\f PANAH V. CHAPPELL 25\\n\\ncame to light post-conviction were raised at trial, that would\\nhave done nothing to reasonably change the outcome. Not\\nonly was there no “strong, unequivocal, exculpatory\\nevidence available,” Rios, 299 F.3d at 813, but there was\\nnothing to substantively challenge the serology or pathology\\nevidence. It is true that counsel could have told the jury that\\nMoore’s findings as to the tissue were wrong. But counsel\\ncould not have refuted Moore’s findings as to the bedsheet\\nand robe stains, and even a different outcome on the felony\\nof oral copulation would not affect Panah’s guilty verdict\\nand death sentence. Also, Panah’s contention that he was\\nprejudiced by counsel’s failure to rebut Moore’s mixture\\ntheory because it put him at the crime scene fails to\\nacknowledge that finding Parker’s body in his bedroom\\nalone was sufficient to do that. Therefore, whatever rebuttal\\nof the State’s expert witnesses that Panah believes he was\\ndeprived of and thus prejudiced by would not have overcome\\nthe other significant evidence of guilt.\\n\\n Moreover, for however deficient defense counsel’s\\ninvestigation was, the state court also could have reasonably\\nfound no prejudice because counsel adequately challenged\\nthe State’s expert witnesses on the stand. See Richter,\\n562 U.S. at 111 (“In many instances cross-examination will\\nbe sufficient to expose defects in an expert’s presentation.”).\\nDuring Moore’s cross-examination, defense counsel\\nattempted to cast doubt on his findings. His questions\\npushed Moore to acknowledge the high statistical\\nprobability of persons other than Panah and Parker\\ncontributing to the stains, and that Moore could not even\\n“determine when th[e] stains were deposited.” Counsel also\\nremarked that Moore offered nothing more than\\n“inconclusive blood typing,” hypothesized that Moore\\nsimply “construct[ed] some sort of theory whereby you can\\nlink that blood to Mr. Panah or to the deceased,” and\\n\\f26 PANAH V. CHAPPELL\\n\\nquestioned why the jury was “not offered DNA evidence.”\\nNotably, even one of the post-conviction reports described\\nthe cross-examination of Moore as “reasonably successful.”\\n\\n Similarly, we do not think there was anything in the\\ndefense’s questioning of Dr. Heuser that would have\\nchanged the outcome. There was nothing to rebut her\\ndetailed testimony about Parker’s extensive injuries. Panah\\ninstead alleges that he was prejudiced by counsel’s failure to\\nrebut Dr. Heuser’s assessment of the cause and time of death.\\nBut at trial, counsel did flag concerns with this part of her\\ntestimony. Thus, we agree with the district court’s\\nconclusion that, “in light of the ‘setting of a sexual assault,’”\\nfurther challenging Dr. Heuser’s testimony on the cause of\\ndeath “would have been no more palatable to the jury.”\\n\\n As a result, the state court would not have unreasonably\\ndetermined that counsel’s casting-doubt strategy was\\nappropriate, even effective, and thus found a lack of\\nprejudice. With little to do about the State’s formidable\\nevidence against Panah, counsel still sought to inform the\\njury of weaknesses in the experts’ testimony. See id. (“When\\ndefense counsel does not have a solid case, the best strategy\\ncan be to say that there is too much doubt about the State’s\\ntheory for a jury to convict.”). And here the strategy even\\nseems to have been somewhat successful, as the jury rejected\\nthe special circumstance allegation of oral copulation. Like\\nin the Napue analysis, this is probative evidence that the jury\\ndid not give persuasive weight to the serology testimony,\\npresumably because of defense counsel’s strategy and cross-\\nexamination.\\n\\n Our analysis is further guided by Richter, in which the\\nSupreme Court rejected a similar ineffective assistance\\nclaim based on counsel’s failure to present an expert to rebut\\nthe State’s forensic evidence. Id. at 111–13. The Court’s\\n\\f PANAH V. CHAPPELL 27\\n\\nreasoning is almost entirely applicable here. Holding that\\nthe petitioner was not prejudiced, the Court concluded that\\nthere was little chance of a different outcome because the\\npost-conviction evidence did not exonerate petitioner and\\nbecause some of petitioner’s post-conviction evidentiary\\nconcerns were already raised by counsel at trial before the\\njury. The Court went on to determine – as is also applicable\\nhere – that the petitioner also did not prove any likelihood of\\na different outcome because he had done nothing to rebut the\\nother “sufficient conventional circumstantial evidence\\npointing to [his] guilt.” 3 Id. at 113.\\n\\n We do not hold that counsel should not have done more.\\nBut, based on the particular facts before us, we recognize\\nthere was “nothing more than a theoretical possibility” of a\\ndifferent verdict. Id. at 112. The evidence of Panah’s guilt\\nwas so strong that there remained an “ample basis for the\\nCalifornia Supreme Court to think any real possibility of\\n[Panah’s] being acquitted was eclipsed by the remaining\\nevidence pointing to guilt.” Id. at 113. Therefore, we affirm\\nthe district court’s denial of Panah’s ineffective assistance\\nclaim.\\n\\n AFFIRMED.\\n\\n\\n\\n\\n 3\\n Panah’s attempt to analogize his case to several out-of-circuit cases\\nfalls short. For instance, in Elmore v. Ozmint, 661 F.3d 783, 871–72 (4th\\nCir. 2011), the Fourth Circuit held that the petitioner was prejudiced by\\ncounsel’s failure to investigate any of the prosecution’s forensic\\nevidence. But, there, “[t]he case was a real ‘who-done-it,’” and, with a\\nproper “investigation, the jury undeniably would have seen a drastically\\ndifferent – and significantly weaker – prosecution case.” Id. at 861, 870.\\nNone of that is true here.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4431731/\", \"author_raw\": \"OWENS, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"KIM MCLANE WARDLAW"},"judge2":{"kind":"string","value":"JACQUELINE H NGUYEN"},"judge3":{"kind":"string","value":"JOHN B OWENS"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4654478/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Hooman Panah v. Kevin Chappell\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-21\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nHOOMAN ASHKAN PANAH, No. 13-99010\n Petitioner-Appellant,\n D.C. No.\n v. 2:05-cv-07606-\n RGK\nKEVIN CHAPPELL, Warden of\nCalifornia State Prison at San\nQuentin, OPINION\n Respondent-Appellee.\n\n Appeal from the United States District Court\n for the Central District of California\n R. Gary Klausner, District Judge, Presiding\n\n Argued and Submitted June 26, 2019\n Pasadena, California\n\n Filed August 21, 2019\n\nBefore: Kim McLane Wardlaw, Jacqueline H. Nguyen,\n and John B. Owens, Circuit Judges.\n\n Opinion by Judge Owens\n\f2 PANAH V. CHAPPELL\n\n SUMMARY *\n\n\n Habeas Corpus\n\n The panel affirmed the district court’s denial of Hooman\nPanah’s habeas corpus petition challenging his State of\nCalifornia conviction and sentence for the first-degree\nmurder and sexual assault of an eight-year-old girl.\n\n The district court granted a certificate of appealability as\nto Panah’s claim brought pursuant to Napue v. Illinois, 360\nU.S. 264 (1959), in which Panah, relying on post-conviction\nDNA reports, contended that he was prejudiced by the\nState’s presentation of serology testimony which, he argued,\nthe State knew was false and misleading. The panel held that\nthe California Supreme Court reasonably rejected this claim.\nThe panel held that even assuming there was no reasonable\nbasis for the state court to deny the claim as to the first two\nNapue requirements – that the testimony was false or\nmisleading, and that the State knew or should have known\nthat – the panel could not say that it would be unreasonable\nto conclude that the testimony did not satisfy the third\nrequirement – materiality. Observing that even setting aside\nthe serology testimony, the case against Panah was\ndevastating, the panel held that the California Supreme\nCourt would not have erred in finding no reasonable\nlikelihood that the testimony could have affected the verdict.\n\n The panel expanded the certificate of appealability to\nencompass Panah’s claim that his trial counsel rendered\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f PANAH V. CHAPPELL 3\n\nineffective assistance by failing to conduct a reasonable\ninvestigation and therefore not rebutting the State’s serology\nand pathology evidence. The panel expressed concern with\ncounsel’s lack of pre-trial investigation, but held that even\nassuming counsel’s performance was deficient, it could not\nsay – in light of the overwhelming evidence of Panah’s guilt\nand the deference owed the state court judgment – that the\nCalifornia Supreme Court would have erred in finding no\nreasonable probability that, but for counsel’s unprofessional\nerrors, the result of the proceeding would have been\ndifferent.\n\n\n COUNSEL\n\nJoseph A. Trigilio (argued), Mark R. Drozdowski, and Susel\nB. Carrillo-Orellana, Deputy Federal Public Defenders;\nHilary Potashner, Federal Public Defender; Office of the\nFederal Public Defender, Los Angeles, California; Firdaus\nF. Dordi, Los Angeles, California; for Petitioner-Appellant.\n\nToni R. Johns Estaville (argued), Ana R. Duarte, A. Scott\nHayward, and Dana M. Ali, Deputy Attorneys General;\nLance E. Winters, Senior Assistant Attorney General;\nGerald A. Engler, Chief Assistant Attorney General; Xavier\nBecerra, Attorney General; Office of the Attorney General,\nLos Angeles, California; for Respondent-Appellee.\n\f4 PANAH V. CHAPPELL\n\n OPINION\n\nOWENS, Circuit Judge:\n\n California state prisoner Hooman Panah appeals from\nthe district court’s denial of his habeas corpus petition\nchallenging his conviction and sentence for the first-degree\nmurder and sexual assault of eight-year-old Nicole Parker.\nWe have jurisdiction under 28 U.S.C. § 1291, and we affirm.\n\n\n\n\n In the early afternoon of November 20, 1993, Parker\nwent missing from her father’s apartment complex, where\nPanah also lived with his mother. While searching for her in\nthe complex, Parker’s father knocked on Panah’s door and\nasked if Panah had seen her. Panah responded something\nlike, “oh, is she missing.” He then offered to help Parker’s\nfather look for her, “persistent[ly]” suggesting they search\noutside the apartment complex. Soon after, the police\narrived and conducted a door-to-door search for Parker,\nincluding Panah’s apartment. The police did not find Parker\nor any clues as to her whereabouts.\n\n That day, Panah reported to work in the mid-afternoon.\nAround 5:30 pm, his mother, who was with two police\nofficers, called Panah. The officers asked him if he knew\nParker or had seen her that day. He responded that he knew\nher only “vaguely” and denied having seen her that day.\nShortly after the officers’ inquiry – hours before his shift\nended – Panah left work without telling anyone. He later\ncalled his manager to say that he would not return “because\nsome people that he knew [were] trying to get him in trouble\nand would [his manager] please inform his mother to get out\n\f PANAH V. CHAPPELL 5\n\nof town.” Panah also paged his co-worker, Rauni Campbell,\nasking for help. He told her that he “d[id] something very\nbad,” “so big” that she would find out.\n\n The next morning, Panah showed up without warning at\nCampbell’s apartment. His wrists were cut, and he requested\nsleeping pills, which she helped him buy. Campbell asked\nPanah if he had anything to do with “the little girl that was\nmissing from his apartment complex.” He said yes. She\nthen asked him whether Parker was still alive. He said no.\nAt this point, Campbell surreptitiously called the police.\nWhen they arrived, Panah tried to evade arrest but was\neventually caught and taken to the hospital. At the hospital,\nunder the influence of drugs and reportedly in a psychotic\nstate, Panah told police, in response to questions about\nParker, that he “liked her very much, even [to] carry her\nskeleton remains around.”\n\n Later that evening, the police, armed with a search\nwarrant, returned to Panah’s apartment. In his bedroom\ncloset, they found Parker’s naked body, wrapped in a\nbedsheet and stuffed in a suitcase. The police then gathered\nevidence from Panah’s bedroom, including examining his\nbed and Parker’s body for evidence of sexual assault.\n\n Panah was indicted on charges of first-degree murder\nwith special circumstances alleging that the murder occurred\nduring a kidnapping, sodomy, lewd acts on a person under\nfourteen years old, and oral copulation of a person under\nfourteen years old. He was also charged with the substantive\ncounts of kidnapping, sodomy by force, lewd acts on a\nperson under fourteen years old, penetration of genital or\nanal openings by a foreign object with a person under\nfourteen years old, and oral copulation of a person under\nfourteen years old. Panah pled not guilty.\n\f6 PANAH V. CHAPPELL\n\n\n\n Panah was initially represented by a family friend,\nSyamak Shafi-Nia, who had limited criminal law\nexperience. But prior to trial, the court appointed Robert\nSheahen, a veteran criminal lawyer, as lead counsel, and\nallowed Shafi-Nia to stay on as second counsel. Sheahen\nhad requested this appointment, promising the court that he\nwould facilitate a settlement, which would “save[] a great\ndeal of time and the taxpayers would be saved a great deal\nof money” by avoiding “an extremely costly trial.”\n\n In July 1994, several months before trial, the State\nnotified the court and defense that it had ordered DNA\ntesting on evidence found at the crime scene. While\nawaiting the test results in September, the court urged\nSheahen to “find a DNA expert to assist you” and “see if\nthere’s any basis for questioning the results.” In October,\ntwo months before trial, the State shared the DNA test results\nwith the defense. Again, the court advised Sheahen to retain\nan expert, to which Sheahen responded, “That will be taken\ncare of.”\n\n However, as trial approached, the State decided not to\nintroduce the DNA evidence. The court pressed defense\ncounsel why he had not yet independently tested the DNA.\nCounsel explained that doing so “would put us in the\nposition of confirming the prosecution results,” and that he\ninstead planned to argue that the State’s “failure to do DNA\ntesting should be held against” them. The court approved of\nthis strategy, calling counsel’s “tactics . . . very sound in this\nparticular case.”\n\f PANAH V. CHAPPELL 7\n\n\n\n Panah’s trial began on December 5, 1994. With jury\nselection set to begin, Sheahen notified the court that Shafi-\nNia was no longer able to serve as second counsel but did\nnot request a continuance. Accordingly, Panah began jury\nselection with just one lawyer. Shortly after, the court\nappointed new second counsel to replace Shafi-Nia, but\nsecond counsel was required to familiarize himself with the\ncase during trial.\n\n\n\n The State’s theory was felony murder. It emphasized the\nabundance of circumstantial evidence against Panah and\nfocused on “Parker’s body bloody and battered,” which was\n“tied up in a sheet inside a zipped suitcase” in Panah’s closet.\nIt also highlighted Panah’s incriminating behavior soon after\nParker went missing, including that Panah was “anxious” to\nencourage Parker’s father to search outside the apartment\ncomplex; “had fled” work hours before his shift ended after\nreceiving a call about Parker from his mother and police; and\nmade numerous admissions about his involvement in\nParker’s murder. Panah’s manager and Campbell testified\nabout his statements on the day of and after Parker’s\ndisappearance.\n\n The State also presented forensic evidence as part of its\ncase-in-chief.\n\n\n\n The coroner, Dr. Heuser, testified that Parker’s physical\ninjuries occurred premortem. Describing the violent nature\nof the assault, she explained that Parker suffered “blunt\nforce” injuries, including bruising on her forehead, eye,\n\f8 PANAH V. CHAPPELL\n\nneck, lip, arms, and buttocks; scratches on her inner thighs;\nand brain swelling. Dr. Heuser also testified to Parker’s\nsexual assault injuries. Parker had bruising, as well as signs\nof bleeding and tears, in the vaginal and rectal areas. Her\nvaginal opening was “very widely” open – most likely\nconsistent with digital penetration. Her anal opening was\nalso “widely open and very lax looking,” “consistent with\nthe insertion of a penis into her rectum.”\n\n Dr. Heuser also testified that Parker’s death was due to\n“traumatic injuries,” either the result of “manual\nstrangulation” or force to Parker’s rectum.\n\n\n\n Serologist William Moore testified about stains found on\nthree items in Panah’s bedroom: (1) the bedsheet Parker was\nwrapped in, (2) a robe found on Panah’s bed, 1 and (3) a\ntissue from a wastebasket in Panah’s bathroom.\nPreliminarily, Moore noted that Panah’s blood type was B\nand Parker’s was A. Because his testing discovered that\nstains on each item contained a mixture of A and B antigens,\nMoore posited that this was consistent with a mixture of\nParker’s and Panah’s bodily fluids and thus sexual contact\nbetween them.\n\n As to the bedsheet, Moore described the stains as\n“indicative of a mixture of physiological fluids” – blood,\nsemen, and amylase (a constituent of saliva and other bodily\nfluids) – that included both A and B antigens. Because the\nbloodstain was “consistent” with Parker’s type A blood, he\nsurmised the other stains were consistent with semen from\nPanah and saliva from Parker. He also noted that the pattern\n\n 1\n The district court’s opinion calls the robe a kimono.\n\f PANAH V. CHAPPELL 9\n\nof the stains “could . . . be consistent with the spewing of\nsemen across the bed sheet.”\n\n Moore similarly testified that the tissue “bore semen\nstains, and high amylase activity,” likely from a mix of\nParker’s and Panah’s bodily fluids. Again, he remarked that\nthis stain “could be consistent with the product of an oral\ncopulation.” Lastly, he testified that Panah’s robe had a\nlarge stain with a mix of A and B antigens. Because the\nrobe’s bloodstain was consistent with Parker, he\nhypothesized that the B antigens came from Panah’s saliva.\n\n Moore also briefly testified about evidence collected\nfrom the sexual assault examination, although he did not\nconduct it. He acknowledged that the oral and anal swabs\nhad not produced any signs of semen, nor was “the presence\nof semen conclusively” found anywhere on Parker’s body.\n\n On cross-examination, Moore admitted that he could not\n“establish any certainty” that either Parker or Panah was a\ncontributor to the stains because of the relatively high\nstatistical frequency of the A and B antigens matching other\npeople. Moore further testified that his theory would not\nbear out if the A and B antigens on the three items came from\none person with AB blood type. 2\n\n\n\n The defense consisted of testimony from Dr. John\nPalmer, the emergency room doctor who treated Panah the\nday after Parker went missing, and several character\nwitnesses. Dr. Palmer reported that Panah was “acutely\n 2\n After the State presented its case, the court granted the defense’s\nmotion for acquittal on the substantive charges of kidnapping and the\nspecial circumstance allegation of kidnapping.\n\f10 PANAH V. CHAPPELL\n\npsychotic” and suicidal when brought into the emergency\nroom that day.\n\n\n\n The State summarized its evidence: “We have blood\ntyping that matches. We have the body in his suitcase in his\ncloset, and we have statements he makes that show\nknowledge before the body was found. We have his\ninvolvement in the crime clearly established.” The State\nfocused on where Parker’s body was found: “You have a\nbody in his closet, in his suitcase. There isn’t a whole lot\nmore you need to do after that in terms of looking and\ninvestigating outside of the obvious, which is that Mr. Panah\nis the person involved.” It also focused on Panah’s\nstatements after Parker went missing: “Those aren’t crazed\nremarks. Those are the remarks of an individual who is\ntelling exactly what happened.” The State then reminded the\njury of Moore’s testimony, arguing it was “not a harebrained\nprosecution theory.” It particularly used his testimony to\nprove the alleged oral copulation:\n\n We think the evidence that was presented to\n you is very consistent with the fact that he\n ejaculated in her mouth, that he allowed her\n to spit it out in a Kleenex, because we have\n the evidence of semen of his blood type, high\n amylase content, indicating a saliva which\n matches her blood type on the Kleenex, as\n well as having a spattering on the bed sheet\n of a mixture of semen and saliva – again the\n high amylase indicating saliva – of his type B\n and her type A.\n\nIt also said that “the one possible inference that can be\ndrawn” from Moore’s testimony about the robe is that the B\n\f PANAH V. CHAPPELL 11\n\nantigens came from “the saliva of the defendant” during the\nsodomy.\n\n Acknowledging that none of this was “conclusive\nevidence,” the State argued that, “when taken with\neverything else[, this] would indicate that there had been an\nact of oral copulation, that there was ejaculate in Nicole\nParker’s mouth.” The State also responded to the defense’s\nassertion that its case was weak because of the lack of DNA,\nclaiming that DNA testing is “usually ordered in a situation\nwhere you don’t have other types of proof available. In this\nsituation we have the proof available.”\n\n In its closing, the defense questioned the reliability of the\nserology evidence, calling Moore’s theory “hogwash,” and\ninsisted that the stains proved nothing. Rather, counsel\nhighlighted the lack of DNA evidence, which “could tell us\nwho’s the source of this stuff . . . [and] whether it, in fact,\ncould be traced to the deceased or whether it could be traced\nto any number of other people.”\n\n\n\n The jury convicted Panah of first-degree murder and the\nother felonies. The jury also found true the special\ncircumstance allegations that the murder was committed\nwhile engaged in the crime of sodomy and lewd acts on a\nperson under the age of fourteen. The jury did not find true\nthe special circumstance allegation that the murder occurred\nin the commission of oral copulation. After the penalty\nphase, the jury returned a death sentence.\n\n\n\n On March 14, 2005, the California Supreme Court\naffirmed Panah’s conviction and sentence, People v. Panah,\n\f12 PANAH V. CHAPPELL\n\n107 P.3d 790 (Cal. 2005), and the United States Supreme\nCourt subsequently denied certiorari, Panah v. California,\n546 U.S. 1216 (2006). About a year later, the California\nSupreme Court summarily denied Panah’s first habeas\npetition. After filing a protective habeas petition in the\nDistrict Court for the Central District of California, Panah\nfiled a second state habeas petition and a first amended\npetition in the district court. The district court stayed\nproceedings during the pendency of Panah’s state habeas\nproceedings. On March 16, 2011, the California Supreme\nCourt again summarily denied Panah’s second state habeas\npetition, and the district court lifted the stay on Panah’s\nfederal habeas proceedings.\n\n In his habeas petitions, Panah provided new evidence,\nincluding two reports detailing post-conviction DNA testing\non the stains that Moore testified about at trial. The first\nreport (“Calandro Report”), prepared in 2004, disagreed\nwith much of Moore’s testimony. But because the Calandro\nReport yielded several inconclusive results, additional\ntesting was conducted two years later, leading to the second\nreport (“Inman Report”). Both reports doubted the\nfoundation of Moore’s mixture theory. The Calandro Report\ncalled “Mr. Moore’s approach . . . biased and indefensible,”\nand the Inman Report wrote that “[n]o biological evidence\nexists to support the hypothesis that a mixture of biological\nfluids from Mr. Panah and Ms. Parker was present on the\ntissue, bedsheet, or kimono.”\n\n More specifically, both reports “contradict[ed]” Moore’s\ntestimony about the tissue. While Moore testified that Panah\nand Parker were both possible contributors to the tissue stain,\nthe reports eliminated any possibility that Parker was a\nsource. On appeal to this court, the State concedes that\nMoore’s tissue testimony was false.\n\f PANAH V. CHAPPELL 13\n\n However, the post-conviction testing produced\ninconclusive results regarding the bedsheet and robe stains.\nNeither report could definitively eliminate Parker as a\ncontributor to several stains on the bedsheet. While they\nconclusively found that two stains were consistent only with\nPanah, they could not conclusively rule Parker out as a\ncontributor to the three other bedsheet stains. Although this\nleft open the possibility that Moore’s mixture theory was\ncorrect, the reports opined that this at minimum refuted his\nassumption that “Ms. Parker ‘spit out’ ejaculate onto the bed\nsheet” because one would then expect to “detect Ms.\nParker’s DNA in significant quantities on the bed sheet.” As\nfor the robe, the reports agreed with Moore that the\nbloodstain was consistent with Parker’s blood type. But,\nunlike Moore’s serology testimony, they found no trace of\nPanah’s DNA on the robe.\n\n Panah’s habeas petitions also included declarations from\nhis three trial lawyers. Each declaration acknowledged there\nhad been almost no pre-trial investigation and only a limited\npenalty-phase investigation, nor were any experts retained to\nindependently analyze the State’s serology and pathology\nevidence. Instead, “all of [defense counsel’s] efforts had\ngone into the aborted settlement.”\n\n On November 14, 2013, the district court denied Panah’s\npetition. As for Panah’s Napue v. Illinois, 360 U.S. 264\n(1959), claim, the district court held that, even if post-\nconviction DNA testing rendered a portion of Moore’s\ntestimony false, the California Supreme Court could have\nreasonably concluded that it did not render all of the\ntestimony false and that his testimony was immaterial in\nlight of the other evidence. In this discussion, the district\ncourt also rejected Panah’s claim of ineffective assistance for\nfailure to investigate the State’s forensic evidence because\n\f14 PANAH V. CHAPPELL\n\nthe California Supreme Court reasonably could have\nconcluded that Panah was not prejudiced.\n\n The district court granted a certificate of appealability\n(“COA”) for Panah’s Napue claim, discussed below in\nsection III. On appeal, Panah has raised a number of\nuncertified issues in his opening brief, which we treat as a\nrequest to expand the COA. 9th Cir. R. 22-1(e). After\nasking the State to respond to several of the uncertified\nissues, we expand the COA to encompass Panah’s guilt-\nphase ineffective assistance claim, addressed below in\nsection IV, but deny Panah’s request to expand the COA as\nto the other uncertified claims. We evaluate Panah’s two\ncertified claims in turn.\n\n\n\n We review the district court’s denial of habeas relief de\nnovo. Lewis v. Mayle, 391 F.3d 989, 995 (9th Cir. 2004).\n\n Because Panah filed his federal habeas petition after\nApril 24, 1996, it is subject to the Antiterrorism and\nEffective Death Penalty Act of 1996 (“AEDPA”). See\nMurray v. Schriro, 745 F.3d 984, 996 (9th Cir. 2014). Under\nAEDPA, we may grant relief only if the adjudication\n“resulted in a decision that was contrary to, or involved an\nunreasonable application of, clearly established Federal law,\nas determined by the Supreme Court of the United States,”\n28 U.S.C. § 2254(d)(1), or “was based on an unreasonable\ndetermination of the facts in light of the evidence presented\nin the State court proceeding,” id. § 2254(d)(2).\n\n Although we typically “look through” a summary\ndisposition to the last reasoned state court decision, Ylst v.\nNunnemaker, 501 U.S. 797, 806 (1991), here there is no\nreasoned state court decision addressing either certified\n\f PANAH V. CHAPPELL 15\n\nclaim. Therefore, we independently review the record to\ndetermine whether the California Supreme Court had any\nreasonable basis to deny Panah relief. See Reis-Campos v.\nBiter, 832 F.3d 968, 974 (9th Cir. 2016) (citing Harrington\nv. Richter, 562 U.S. 86, 102 (2011)).\n\n\n\n Relying on the two post-conviction DNA reports, Panah\ncontends that he was prejudiced by the State’s presentation\nof Moore’s serology testimony, which he argues the State\nknew was false or misleading.\n\n In Napue, the Supreme Court held “that a conviction\nobtained through use of false evidence, known to be such by\nrepresentatives of the State, must fall under the Fourteenth\nAmendment.” 360 U.S. at 269. Nonetheless, a Napue claim\nsucceeds only if three elements are satisfied. See United\nStates v. Zuno-Arce, 339 F.3d 886, 889 (9th Cir. 2003).\nFirst, the testimony or evidence in question must have been\nfalse or misleading. See id.; see also Alcorta v. Texas,\n355 U.S. 28, 31 (1957) (considering whether “testimony,\ntaken as a whole, gave the jury [a] false impression”).\nSecond, the State must have known or should have known\nthat it was false or misleading. See Zuno-Arce, 339 F.3d\nat 889; see also Maxwell v. Roe, 628 F.3d 486, 506 (9th Cir.\n2010) (“[E]ven false evidence presented in good faith . . .\nhardly comports with fundamental fairness.” (quoting\nKillian v. Poole, 282 F.3d 1204, 1209 (9th Cir. 2002))). And\nthird, because “Napue does not create a ‘per se rule of\nreversal,’” the testimony or evidence in question must be\nmaterial. Sivak v. Hardison, 658 F.3d 898, 912 (9th Cir.\n2011) (quoting Jackson v. Brown, 513 F.3d 1057, 1076 (9th\nCir. 2008)); see also Zuno-Arce, 339 F.3d at 889.\n\f16 PANAH V. CHAPPELL\n\n Materiality under Napue requires a “lesser showing of\nharm . . . than under ordinary harmless error review.” Dow\nv. Virga, 729 F.3d 1041, 1048 (9th Cir. 2013). But, after\nweighing the effect of alleged Napue violations collectively,\nsee Phillips v. Ornoski, 673 F.3d 1168, 1189 (9th Cir. 2012),\nthere still needs to be a “reasonable likelihood that the false\ntestimony could have affected the judgment of the jury.”\nHayes v. Brown, 399 F.3d 972, 985 (9th Cir. 2005) (en banc)\n(quoting Belmontes v. Woodford, 350 F.3d 861, 881 (9th Cir.\n2003)). Thus, a Napue claim fails if, absent the false\ntestimony or evidence, the petitioner still “received a fair\ntrial, understood as a trial resulting in a verdict worthy of\nconfidence.” Hayes, 399 F.3d at 984 (quoting Hall v. Dir.\nof Corr., 343 F.3d 976, 984 (9th Cir. 2003) (per curiam)).\n\n Even if we assume there was no reasonable basis for the\nstate court to deny Panah’s claim as to the first two Napue\nrequirements, we cannot say that it would be unreasonable\nto conclude that Moore’s testimony was immaterial. See\nTowery v. Schriro, 641 F.3d 300, 308 (9th Cir. 2010) (not\nreviewing all three requirements because petitioner’s\n“argument fails at the second Napue prong”). The State\npresented a powerful case of Panah’s guilt, with substantial\nevidence linking him to Parker’s murder and sexual assault.\nMoore’s testimony was just one – and not a crucial – piece\nof that presentation. Because the “verdict” is still reasonably\n“worthy of confidence,” Phillips, 673 F.3d at 1189 (quoting\nSivak, 658 F.3d at 912), we hold that the California Supreme\nCourt would not have erred in finding no “reasonable\nlikelihood that [Moore’s testimony] could have affected the\njudgment of the jury.” Hayes, 399 F.3d at 985 (quoting\nBelmontes, 350 F.3d at 881); see also Phillips, 673 F.3d\nat 1190 (“[T]he prosecution’s Napue violations, although\n‘pernicious’ and ‘reprehensible,’ were not material to\n\f PANAH V. CHAPPELL 17\n\n[petitioner’s] conviction of first-degree murder.” (quoting\nHayes, 399 F.3d at 981)).\n\n Even setting aside Moore’s testimony, the case against\nPanah was devastating. Parker’s naked body was found in a\nsuitcase in Panah’s bedroom closet. Blood stains matching\nParker’s blood type – according to both Moore and the post-\nconviction reports – were found on Panah’s robe. Moreover,\nPanah’s behavior on the day of and after Parker went missing\nwas highly suspicious. Hours after she disappeared, Panah\ntried to divert her father away from where her body was later\nfound. Then, after the police called Panah at work to ask if\nhe had seen Parker, he left without explanation, later telling\nhis manager and Campbell that he was in serious trouble.\nThe next morning, Panah even confided in Campbell that he\nwas involved in Parker’s death. See Arizona v. Fulminante,\n499 U.S. 279, 296 (1991) (“A confession is like no other\nevidence. Indeed, ‘the defendant’s own confession is\nprobably the most probative and damaging evidence that can\nbe admitted against him.’” (quoting Bruton v. United States,\n391 U.S. 123, 139 (1968) (White, J., dissenting))). And then,\nparticularly incriminating in light of his statements, Panah\nattempted suicide the morning after Parker’s disappearance.\n\n The jury then heard Dr. Heuser’s impactful testimony\nabout the crime itself. She described Parker’s extensive\ninjuries from the sexual assault, including significant trauma\nto Parker’s vaginal and rectal areas, indicative of digital and\npenile penetration. In addition, Dr. Heuser described the\nviolent nature of the assault: Parker was hit with “blunt\nforce,” consistent with her “head striking . . . a wall or a\nfloor” or being hit with a “fist.” As a result, Parker’s brain\nwas swollen from this “significant impact,” and she had\nbruises – some caused by “manual strangulation” – along her\nface, neck, arms, buttocks, and legs, and scratches on her\n\f18 PANAH V. CHAPPELL\n\nlegs. Dr. Heuser cited strangulation and sodomy as “the\nmost lethal injuries.”\n\n Although Dr. Heuser’s testimony did not directly\nimplicate Panah, it was nonetheless critical. It established\nthat a crime – and a brutal one – took place. When added to\nthe evidence the jury heard about where Parker’s body was\nfound and Panah’s statements, it was more than sufficient for\nthe jury to render a guilty verdict. And, while Panah\ncontends that Moore’s testimony was prejudicial because of\nits at-times graphic descriptions, particularly of oral\ncopulation, Dr. Heuser’s testimony offered an even more\ngraphic and detailed description of the entire sexual assault\nand murder. As the State itself said, Dr. Heuser’s testimony\nis “probably the most telling evidence of what happened.”\n\n The state court also reasonably could have found\nMoore’s testimony to be an immaterial part of the State’s\ncase because it offered the jury, at most, hypotheticals and\nwavering findings. Unlike Dr. Heuser’s straightforward\ntestimony about Parker’s injuries, Moore acknowledged that\nhis findings rested on a number of assumptions, such as that\nthe A and B antigens came from two people rather than one\nwith AB blood type. And even if he was correct that the A\nand B antigens came from two people, Moore neither could\ndefinitively say they came from Parker and Panah, nor could\nhe even narrow the pool of possible matches to less than\nhundreds of thousands of people. He said frankly on the\nstand: “I cannot establish any certainty based on\nconventional serology. I can only demonstrate consistency.”\nFor this reason, Moore’s testimony was couched in\ninconclusive terms, such as “could have originated from” or\nwere “consistent” with. The State even acknowledged this\nweakness in Moore’s findings in closing argument:\n\f PANAH V. CHAPPELL 19\n\n Now the question is, did a person with AB\n blood leave . . . body fluids such as blood,\n semen[,] and saliva, on the sheets, on the\n toilet paper, on the robe. That is one\n interpretation. The other interpretation, of\n course, is that you have two separate people,\n one of whom has type A, and one has type B.\n\n Therefore, at best, Moore’s testimony required the jury\nto draw its own inferences. This was not the case of\nimpactful expert testimony telling the jury that there was a\none-in-a-million chance the evidence matched anyone but\nPanah. Rather, any effect Moore’s findings may have had\non the jury – which was reasonably none – was fully\ndependent on the other weighty evidence presented by the\nState. For instance, without having found Parker’s body in\nPanah’s bedroom, no juror could have reasonably inferred\nthat Parker was the A antigen contributor.\n\n The jury’s verdict removes any lingering doubt about the\nmateriality of Moore’s testimony. If the State needed\nMoore’s testimony at all, it was to prove the special\ncircumstance allegation and substantive charge of oral\ncopulation. In its closing, for instance, the State referred to\nMoore’s findings of mixed bodily fluids on the tissue and\nbedsheet to prove this sexual act. Yet, the jury did not find\ntrue the special circumstance allegation that Parker’s murder\nwas committed while in the commission of oral copulation.\nAlthough the jury did find Panah guilty of the felony of oral\ncopulation, the verdict is still telling. A reasonable\ninterpretation of the jury’s rejection of this special\ncircumstance is that the jury was not entirely persuaded by\nMoore’s mixture theory. In contrast, this highlights the\neffectiveness of Dr. Heuser’s testimony. The State relied on\nher findings to prove the special circumstance allegations\n\f20 PANAH V. CHAPPELL\n\nand substantive charges of sodomy and lewd acts, which the\njury found to be true.\n\n Still, Panah contends that Moore’s testimony was\nmaterial because it was the only evidence identifying him as\nthe perpetrator. Although creative, this argument makes\nlittle headway. It ignores the substantial evidence tying\nPanah to Parker’s murder and sexual assault. This was not a\ncase where the police had no leads on a suspect. Nor was it\na case where the prosecution needed Moore’s serology\nevidence to place Panah at the crime scene. Rather, as the\nState emphasized in its closing: “You have a body in his\ncloset, in his suitcase. There isn’t a whole lot more you need\nto do after that in terms of looking and investigating outside\nof the obvious, which is that Mr. Panah is the person\ninvolved.” And, as previously mentioned, Panah’s own\nadmissions linked him to the assault. Thus, even without\nMoore’s testimony, the State had no difficulty proving\nidentity.\n\n For these reasons, Panah’s reliance on Miller v. Pate,\n386 U.S. 1 (1967), is unavailing. In Miller, the petitioner\nwas also charged with the murder and sexual assault of a\nyoung girl without any eyewitnesses to the crime. Id. at 2–\n3. The State’s only evidence linking petitioner to the victim\nwas male underwear, which an expert said had blood stains\non it matching the victim, found near the crime scene. Id.\nat 3–4. Post-conviction testing proved the underwear stains\nwere paint, not blood, and that the State had known this at\ntrial. Id. at 5–6. The Supreme Court reversed petitioner’s\nconviction because, while the State successfully used the\nunderwear as “an important link in the chain of\ncircumstantial evidence against the petitioner,” in reality it\nwas “virtually valueless as evidence.” Id. at 4, 6.\n\f PANAH V. CHAPPELL 21\n\n Miller is clearly distinguishable. First, although the\nparties agree that the tissue testimony was false, post-\nconviction testing did not render the majority of Moore’s\ntestimony false. Neither post-conviction report conclusively\nrefuted his findings as to the bedsheet or robe stains.\nTherefore, unlike in Miller, the evidence in question here did\nnot become “virtually valueless” to convict. Id. at 6.\nSecond, Moore’s testimony was not the “important link” in\nproving Panah’s guilt. Id. at 4. The State did not even need\nMoore’s testimony to convict Panah. This case differs\nsignificantly from Miller because here the state court could\nreasonably rely on an abundance of other evidence to still\nhave confidence in the conviction.\n\n Panah’s reliance on Alcorta, 355 U.S. 28, also is\nmisplaced. There, the Court overturned petitioner’s\nconviction because the false “testimony was seriously\nprejudicial to petitioner” and “tended squarely to refute his\nclaim.” Id. at 31. Had the false testimony been corrected,\npetitioner’s defense would have been corroborated. But,\nhere, even if Moore’s testimony had been corrected or\nentirely excluded, the jury would not have heard a\nsignificantly different presentation of evidence. At most,\nalthough we think unlikely, the State’s case may have\nbecome marginally weaker.\n\n Rather, this case is quite similar to Sivak, 658 F.3d 898,\nin which we rejected a guilt-phase Napue claim because the\ncourt had “full confidence that the jury would still have\nconvicted.” Id. at 913. There, we held: “[E]ven if the jury\ndisbelieved [the false testimony] entirely . . . there still is no\n‘reasonable likelihood that the false testimony could have\naffected the judgment of the jury.’” Id. at 914 (quoting\nJackson, 513 F.3d at 1076). As in Sivak, we think the\nCalifornia Supreme Court reasonably still could have had\n\f22 PANAH V. CHAPPELL\n\n“full confidence” that the jury would have returned the same\nverdict even in absence of Moore’s testimony. Id. at 913.\n\n In sum, we do not think Moore’s testimony was critical\nin convicting Panah. Excluding Moore’s testimony, the\nState’s case was still devastating and largely unchallenged.\nMoore’s testimony was certainly not “the centerpiece of the\nprosecution’s case.” Hayes, 399 F.3d at 985. Rather, in light\nof the overwhelming evidence against Panah and the jury’s\nrejection of the oral copulation special circumstance, it is\nreasonable to conclude that his testimony had essentially no\neffect on the jury’s decision making. Cf. Dow, 729 F.3d\nat 1049–50 (concluding that false testimony was material\nbecause “[t]he evidence against [petitioner] was weak”);\nMaxwell, 628 F.3d at 508 (holding that false testimony was\nmaterial because it came from a “‘make-or-break’ witness\nfor the State” and there was a “paucity of other evidence”);\nHall, 343 F.3d at 984 (reversing under Napue “[i]n light of\nthe already scant evidence on which the conviction was\nbased”). As such, we conclude that the California Supreme\nCourt reasonably rejected Panah’s Napue claim.\n\n\n\n We next turn to Panah’s guilt-phase ineffective\nassistance claim. To prevail, Panah must show that his\ncounsel’s performance “fell below an objective standard of\nreasonableness” and “a reasonable probability that, but for\ncounsel’s unprofessional errors, the result of the proceeding\nwould have been different.” Strickland v. Washington,\n466 U.S. 668, 688, 694 (1984). “The likelihood of a\ndifferent result,” however, “must be substantial, not just\nconceivable.” Richter, 562 U.S. at 112. This already\nimposing standard becomes doubly difficult to satisfy once\nAEDPA deference is tacked on. See Cullen v. Pinholster,\n563 U.S. 170, 189 (2011) (“[T]he benchmark for judging any\n\f PANAH V. CHAPPELL 23\n\nclaim of ineffectiveness [under Strickland] must be whether\ncounsel’s conduct so undermined the proper functioning of\nthe adversarial process that the trial cannot be relied on as\nhaving produced a just result.” (quoting Strickland, 466 U.S.\nat 686)).\n\n Panah argues that his trial counsel was ineffective for\nfailing to conduct a reasonable investigation and therefore\nnot rebutting the State’s serology and pathology evidence.\nIn support of his claim, Panah notes that his counsel never\nretained an expert to independently analyze the pathology\nand serology evidence or to testify at the guilt phase.\nInstead, defense counsel essentially seemed to accept the\nState’s evidence as true. In post-conviction proceedings,\ncounsel acknowledged that his inordinate focus on\nsettlement resulted in too little, if even any, pre-trial\ninvestigation.\n\n While we are instructed to “avoid the temptation to\nsecond-guess [counsel’s] performance,” Mayfield v.\nWoodford, 270 F.3d 915, 927 (9th Cir. 2001) (en banc), we\nare concerned with defense counsel’s lack of pre-trial\ninvestigation. See Duncan v. Ornoski, 528 F.3d 1222, 1235\n(9th Cir. 2008) (“[L]awyers [have] considerable discretion\nto make strategic decisions about what to investigate, but\nonly after those lawyers ‘have gathered sufficient evidence\nupon which to base their tactical choices.’” (quoting\nJennings v. Woodford, 290 F.3d 1006, 1014 (9th Cir.\n2002))). But we “need not determine whether counsel’s\nperformance was deficient before examining the prejudice\nsuffered by the defendant as a result of the alleged\ndeficiencies.” Strickland, 466 U.S. at 697. Here, even\nassuming counsel’s performance was deficient, we cannot\nsay – in light of the overwhelming evidence of Panah’s guilt\nand the deference we owe the state court judgment – that the\n\f24 PANAH V. CHAPPELL\n\nCalifornia Supreme Court would have erred in finding no\n“reasonable probability that, but for counsel’s\nunprofessional errors, the result of the proceeding would\nhave been different.” Id. at 694.\n\n After weighing counsel’s deficiencies cumulatively with\n“the strength of the government’s case,” Rios v. Rocha,\n299 F.3d 796, 808–09 (9th Cir. 2002) (quoting Eggleston v.\nUnited States, 798 F.2d 374, 376 (9th Cir. 1986)), we believe\nthe state court reasonably rejected Panah’s assertion that the\ntrial’s outcome “would have been dramatically different”\nhad counsel’s performance not been deficient. Our\nreasoning on prejudice bears significant similarities to why\nwe reject Panah’s Napue claim. We do not wish to harp on\nwhat was detailed in the preceding section, but the State had\na uniquely strong case against Panah. Parker’s body was\nfound in his bedroom; Panah’s behavior the day of and after\nher disappearance was incriminating; Panah admitted his\nown involvement in her death; and Parker’s serious physical\ninjuries, including to her genitalia, were well-documented.\nDefense counsel even acknowledged that this evidence was\nthe foundation of the State’s case: “The critical pieces of\nevidence are obviously that the child’s body is found in Mr.\nPanah’s closet, her naked body with a considerable amount\nof blood. There is evidence . . . that the child was beaten.”\n\n It is, therefore, inconceivable, even had defense counsel\nindependently investigated the serology and pathology\nevidence, that the jury would have reached a different\nverdict. See Williams v. Filson, 908 F.3d 546, 570 (9th Cir.\n2018) (“We have long recognized . . . that ‘prejudice\nresulting from ineffective assistance of counsel must be\n“considered collectively, not item by item.”’” (quoting Doe\nv. Ayers, 782 F.3d 425, 460 n.62 (9th Cir. 2015))). Even if\nthe weaknesses in Moore’s and Dr. Heuser’s findings that\n\f PANAH V. CHAPPELL 25\n\ncame to light post-conviction were raised at trial, that would\nhave done nothing to reasonably change the outcome. Not\nonly was there no “strong, unequivocal, exculpatory\nevidence available,” Rios, 299 F.3d at 813, but there was\nnothing to substantively challenge the serology or pathology\nevidence. It is true that counsel could have told the jury that\nMoore’s findings as to the tissue were wrong. But counsel\ncould not have refuted Moore’s findings as to the bedsheet\nand robe stains, and even a different outcome on the felony\nof oral copulation would not affect Panah’s guilty verdict\nand death sentence. Also, Panah’s contention that he was\nprejudiced by counsel’s failure to rebut Moore’s mixture\ntheory because it put him at the crime scene fails to\nacknowledge that finding Parker’s body in his bedroom\nalone was sufficient to do that. Therefore, whatever rebuttal\nof the State’s expert witnesses that Panah believes he was\ndeprived of and thus prejudiced by would not have overcome\nthe other significant evidence of guilt.\n\n Moreover, for however deficient defense counsel’s\ninvestigation was, the state court also could have reasonably\nfound no prejudice because counsel adequately challenged\nthe State’s expert witnesses on the stand. See Richter,\n562 U.S. at 111 (“In many instances cross-examination will\nbe sufficient to expose defects in an expert’s presentation.”).\nDuring Moore’s cross-examination, defense counsel\nattempted to cast doubt on his findings. His questions\npushed Moore to acknowledge the high statistical\nprobability of persons other than Panah and Parker\ncontributing to the stains, and that Moore could not even\n“determine when th[e] stains were deposited.” Counsel also\nremarked that Moore offered nothing more than\n“inconclusive blood typing,” hypothesized that Moore\nsimply “construct[ed] some sort of theory whereby you can\nlink that blood to Mr. Panah or to the deceased,” and\n\f26 PANAH V. CHAPPELL\n\nquestioned why the jury was “not offered DNA evidence.”\nNotably, even one of the post-conviction reports described\nthe cross-examination of Moore as “reasonably successful.”\n\n Similarly, we do not think there was anything in the\ndefense’s questioning of Dr. Heuser that would have\nchanged the outcome. There was nothing to rebut her\ndetailed testimony about Parker’s extensive injuries. Panah\ninstead alleges that he was prejudiced by counsel’s failure to\nrebut Dr. Heuser’s assessment of the cause and time of death.\nBut at trial, counsel did flag concerns with this part of her\ntestimony. Thus, we agree with the district court’s\nconclusion that, “in light of the ‘setting of a sexual assault,’”\nfurther challenging Dr. Heuser’s testimony on the cause of\ndeath “would have been no more palatable to the jury.”\n\n As a result, the state court would not have unreasonably\ndetermined that counsel’s casting-doubt strategy was\nappropriate, even effective, and thus found a lack of\nprejudice. With little to do about the State’s formidable\nevidence against Panah, counsel still sought to inform the\njury of weaknesses in the experts’ testimony. See id. (“When\ndefense counsel does not have a solid case, the best strategy\ncan be to say that there is too much doubt about the State’s\ntheory for a jury to convict.”). And here the strategy even\nseems to have been somewhat successful, as the jury rejected\nthe special circumstance allegation of oral copulation. Like\nin the Napue analysis, this is probative evidence that the jury\ndid not give persuasive weight to the serology testimony,\npresumably because of defense counsel’s strategy and cross-\nexamination.\n\n Our analysis is further guided by Richter, in which the\nSupreme Court rejected a similar ineffective assistance\nclaim based on counsel’s failure to present an expert to rebut\nthe State’s forensic evidence. Id. at 111–13. The Court’s\n\f PANAH V. CHAPPELL 27\n\nreasoning is almost entirely applicable here. Holding that\nthe petitioner was not prejudiced, the Court concluded that\nthere was little chance of a different outcome because the\npost-conviction evidence did not exonerate petitioner and\nbecause some of petitioner’s post-conviction evidentiary\nconcerns were already raised by counsel at trial before the\njury. The Court went on to determine – as is also applicable\nhere – that the petitioner also did not prove any likelihood of\na different outcome because he had done nothing to rebut the\nother “sufficient conventional circumstantial evidence\npointing to [his] guilt.” 3 Id. at 113.\n\n We do not hold that counsel should not have done more.\nBut, based on the particular facts before us, we recognize\nthere was “nothing more than a theoretical possibility” of a\ndifferent verdict. Id. at 112. The evidence of Panah’s guilt\nwas so strong that there remained an “ample basis for the\nCalifornia Supreme Court to think any real possibility of\n[Panah’s] being acquitted was eclipsed by the remaining\nevidence pointing to guilt.” Id. at 113. Therefore, we affirm\nthe district court’s denial of Panah’s ineffective assistance\nclaim.\n\n AFFIRMED.\n\n\n\n\n 3\n Panah’s attempt to analogize his case to several out-of-circuit cases\nfalls short. For instance, in Elmore v. Ozmint, 661 F.3d 783, 871–72 (4th\nCir. 2011), the Fourth Circuit held that the petitioner was prejudiced by\ncounsel’s failure to investigate any of the prosecution’s forensic\nevidence. But, there, “[t]he case was a real ‘who-done-it,’” and, with a\nproper “investigation, the jury undeniably would have seen a drastically\ndifferent – and significantly weaker – prosecution case.” Id. at 861, 870.\nNone of that is true here.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Hooman Panah v. Kevin Chappell\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-21\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nHOOMAN ASHKAN PANAH, No. 13-99010\\n Petitioner-Appellant,\\n D.C. No.\\n v. 2:05-cv-07606-\\n RGK\\nKEVIN CHAPPELL, Warden of\\nCalifornia State Prison at San\\nQuentin, OPINION\\n Respondent-Appellee.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n R. Gary Klausner, District Judge, Presiding\\n\\n Argued and Submitted June 26, 2019\\n Pasadena, California\\n\\n Filed August 21, 2019\\n\\nBefore: Kim McLane Wardlaw, Jacqueline H. Nguyen,\\n and John B. Owens, Circuit Judges.\\n\\n Opinion by Judge Owens\\n\\f2 PANAH V. CHAPPELL\\n\\n SUMMARY *\\n\\n\\n Habeas Corpus\\n\\n The panel affirmed the district court’s denial of Hooman\\nPanah’s habeas corpus petition challenging his State of\\nCalifornia conviction and sentence for the first-degree\\nmurder and sexual assault of an eight-year-old girl.\\n\\n The district court granted a certificate of appealability as\\nto Panah’s claim brought pursuant to Napue v. Illinois, 360\\nU.S. 264 (1959), in which Panah, relying on post-conviction\\nDNA reports, contended that he was prejudiced by the\\nState’s presentation of serology testimony which, he argued,\\nthe State knew was false and misleading. The panel held that\\nthe California Supreme Court reasonably rejected this claim.\\nThe panel held that even assuming there was no reasonable\\nbasis for the state court to deny the claim as to the first two\\nNapue requirements – that the testimony was false or\\nmisleading, and that the State knew or should have known\\nthat – the panel could not say that it would be unreasonable\\nto conclude that the testimony did not satisfy the third\\nrequirement – materiality. Observing that even setting aside\\nthe serology testimony, the case against Panah was\\ndevastating, the panel held that the California Supreme\\nCourt would not have erred in finding no reasonable\\nlikelihood that the testimony could have affected the verdict.\\n\\n The panel expanded the certificate of appealability to\\nencompass Panah’s claim that his trial counsel rendered\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f PANAH V. CHAPPELL 3\\n\\nineffective assistance by failing to conduct a reasonable\\ninvestigation and therefore not rebutting the State’s serology\\nand pathology evidence. The panel expressed concern with\\ncounsel’s lack of pre-trial investigation, but held that even\\nassuming counsel’s performance was deficient, it could not\\nsay – in light of the overwhelming evidence of Panah’s guilt\\nand the deference owed the state court judgment – that the\\nCalifornia Supreme Court would have erred in finding no\\nreasonable probability that, but for counsel’s unprofessional\\nerrors, the result of the proceeding would have been\\ndifferent.\\n\\n\\n COUNSEL\\n\\nJoseph A. Trigilio (argued), Mark R. Drozdowski, and Susel\\nB. Carrillo-Orellana, Deputy Federal Public Defenders;\\nHilary Potashner, Federal Public Defender; Office of the\\nFederal Public Defender, Los Angeles, California; Firdaus\\nF. Dordi, Los Angeles, California; for Petitioner-Appellant.\\n\\nToni R. Johns Estaville (argued), Ana R. Duarte, A. Scott\\nHayward, and Dana M. Ali, Deputy Attorneys General;\\nLance E. Winters, Senior Assistant Attorney General;\\nGerald A. Engler, Chief Assistant Attorney General; Xavier\\nBecerra, Attorney General; Office of the Attorney General,\\nLos Angeles, California; for Respondent-Appellee.\\n\\f4 PANAH V. CHAPPELL\\n\\n OPINION\\n\\nOWENS, Circuit Judge:\\n\\n California state prisoner Hooman Panah appeals from\\nthe district court’s denial of his habeas corpus petition\\nchallenging his conviction and sentence for the first-degree\\nmurder and sexual assault of eight-year-old Nicole Parker.\\nWe have jurisdiction under 28 U.S.C. § 1291, and we affirm.\\n\\n\\n\\n\\n In the early afternoon of November 20, 1993, Parker\\nwent missing from her father’s apartment complex, where\\nPanah also lived with his mother. While searching for her in\\nthe complex, Parker’s father knocked on Panah’s door and\\nasked if Panah had seen her. Panah responded something\\nlike, “oh, is she missing.” He then offered to help Parker’s\\nfather look for her, “persistent[ly]” suggesting they search\\noutside the apartment complex. Soon after, the police\\narrived and conducted a door-to-door search for Parker,\\nincluding Panah’s apartment. The police did not find Parker\\nor any clues as to her whereabouts.\\n\\n That day, Panah reported to work in the mid-afternoon.\\nAround 5:30 pm, his mother, who was with two police\\nofficers, called Panah. The officers asked him if he knew\\nParker or had seen her that day. He responded that he knew\\nher only “vaguely” and denied having seen her that day.\\nShortly after the officers’ inquiry – hours before his shift\\nended – Panah left work without telling anyone. He later\\ncalled his manager to say that he would not return “because\\nsome people that he knew [were] trying to get him in trouble\\nand would [his manager] please inform his mother to get out\\n\\f PANAH V. CHAPPELL 5\\n\\nof town.” Panah also paged his co-worker, Rauni Campbell,\\nasking for help. He told her that he “d[id] something very\\nbad,” “so big” that she would find out.\\n\\n The next morning, Panah showed up without warning at\\nCampbell’s apartment. His wrists were cut, and he requested\\nsleeping pills, which she helped him buy. Campbell asked\\nPanah if he had anything to do with “the little girl that was\\nmissing from his apartment complex.” He said yes. She\\nthen asked him whether Parker was still alive. He said no.\\nAt this point, Campbell surreptitiously called the police.\\nWhen they arrived, Panah tried to evade arrest but was\\neventually caught and taken to the hospital. At the hospital,\\nunder the influence of drugs and reportedly in a psychotic\\nstate, Panah told police, in response to questions about\\nParker, that he “liked her very much, even [to] carry her\\nskeleton remains around.”\\n\\n Later that evening, the police, armed with a search\\nwarrant, returned to Panah’s apartment. In his bedroom\\ncloset, they found Parker’s naked body, wrapped in a\\nbedsheet and stuffed in a suitcase. The police then gathered\\nevidence from Panah’s bedroom, including examining his\\nbed and Parker’s body for evidence of sexual assault.\\n\\n Panah was indicted on charges of first-degree murder\\nwith special circumstances alleging that the murder occurred\\nduring a kidnapping, sodomy, lewd acts on a person under\\nfourteen years old, and oral copulation of a person under\\nfourteen years old. He was also charged with the substantive\\ncounts of kidnapping, sodomy by force, lewd acts on a\\nperson under fourteen years old, penetration of genital or\\nanal openings by a foreign object with a person under\\nfourteen years old, and oral copulation of a person under\\nfourteen years old. Panah pled not guilty.\\n\\f6 PANAH V. CHAPPELL\\n\\n\\n\\n Panah was initially represented by a family friend,\\nSyamak Shafi-Nia, who had limited criminal law\\nexperience. But prior to trial, the court appointed Robert\\nSheahen, a veteran criminal lawyer, as lead counsel, and\\nallowed Shafi-Nia to stay on as second counsel. Sheahen\\nhad requested this appointment, promising the court that he\\nwould facilitate a settlement, which would “save[] a great\\ndeal of time and the taxpayers would be saved a great deal\\nof money” by avoiding “an extremely costly trial.”\\n\\n In July 1994, several months before trial, the State\\nnotified the court and defense that it had ordered DNA\\ntesting on evidence found at the crime scene. While\\nawaiting the test results in September, the court urged\\nSheahen to “find a DNA expert to assist you” and “see if\\nthere’s any basis for questioning the results.” In October,\\ntwo months before trial, the State shared the DNA test results\\nwith the defense. Again, the court advised Sheahen to retain\\nan expert, to which Sheahen responded, “That will be taken\\ncare of.”\\n\\n However, as trial approached, the State decided not to\\nintroduce the DNA evidence. The court pressed defense\\ncounsel why he had not yet independently tested the DNA.\\nCounsel explained that doing so “would put us in the\\nposition of confirming the prosecution results,” and that he\\ninstead planned to argue that the State’s “failure to do DNA\\ntesting should be held against” them. The court approved of\\nthis strategy, calling counsel’s “tactics . . . very sound in this\\nparticular case.”\\n\\f PANAH V. CHAPPELL 7\\n\\n\\n\\n Panah’s trial began on December 5, 1994. With jury\\nselection set to begin, Sheahen notified the court that Shafi-\\nNia was no longer able to serve as second counsel but did\\nnot request a continuance. Accordingly, Panah began jury\\nselection with just one lawyer. Shortly after, the court\\nappointed new second counsel to replace Shafi-Nia, but\\nsecond counsel was required to familiarize himself with the\\ncase during trial.\\n\\n\\n\\n The State’s theory was felony murder. It emphasized the\\nabundance of circumstantial evidence against Panah and\\nfocused on “Parker’s body bloody and battered,” which was\\n“tied up in a sheet inside a zipped suitcase” in Panah’s closet.\\nIt also highlighted Panah’s incriminating behavior soon after\\nParker went missing, including that Panah was “anxious” to\\nencourage Parker’s father to search outside the apartment\\ncomplex; “had fled” work hours before his shift ended after\\nreceiving a call about Parker from his mother and police; and\\nmade numerous admissions about his involvement in\\nParker’s murder. Panah’s manager and Campbell testified\\nabout his statements on the day of and after Parker’s\\ndisappearance.\\n\\n The State also presented forensic evidence as part of its\\ncase-in-chief.\\n\\n\\n\\n The coroner, Dr. Heuser, testified that Parker’s physical\\ninjuries occurred premortem. Describing the violent nature\\nof the assault, she explained that Parker suffered “blunt\\nforce” injuries, including bruising on her forehead, eye,\\n\\f8 PANAH V. CHAPPELL\\n\\nneck, lip, arms, and buttocks; scratches on her inner thighs;\\nand brain swelling. Dr. Heuser also testified to Parker’s\\nsexual assault injuries. Parker had bruising, as well as signs\\nof bleeding and tears, in the vaginal and rectal areas. Her\\nvaginal opening was “very widely” open – most likely\\nconsistent with digital penetration. Her anal opening was\\nalso “widely open and very lax looking,” “consistent with\\nthe insertion of a penis into her rectum.”\\n\\n Dr. Heuser also testified that Parker’s death was due to\\n“traumatic injuries,” either the result of “manual\\nstrangulation” or force to Parker’s rectum.\\n\\n\\n\\n Serologist William Moore testified about stains found on\\nthree items in Panah’s bedroom: (1) the bedsheet Parker was\\nwrapped in, (2) a robe found on Panah’s bed, 1 and (3) a\\ntissue from a wastebasket in Panah’s bathroom.\\nPreliminarily, Moore noted that Panah’s blood type was B\\nand Parker’s was A. Because his testing discovered that\\nstains on each item contained a mixture of A and B antigens,\\nMoore posited that this was consistent with a mixture of\\nParker’s and Panah’s bodily fluids and thus sexual contact\\nbetween them.\\n\\n As to the bedsheet, Moore described the stains as\\n“indicative of a mixture of physiological fluids” – blood,\\nsemen, and amylase (a constituent of saliva and other bodily\\nfluids) – that included both A and B antigens. Because the\\nbloodstain was “consistent” with Parker’s type A blood, he\\nsurmised the other stains were consistent with semen from\\nPanah and saliva from Parker. He also noted that the pattern\\n\\n 1\\n The district court’s opinion calls the robe a kimono.\\n\\f PANAH V. CHAPPELL 9\\n\\nof the stains “could . . . be consistent with the spewing of\\nsemen across the bed sheet.”\\n\\n Moore similarly testified that the tissue “bore semen\\nstains, and high amylase activity,” likely from a mix of\\nParker’s and Panah’s bodily fluids. Again, he remarked that\\nthis stain “could be consistent with the product of an oral\\ncopulation.” Lastly, he testified that Panah’s robe had a\\nlarge stain with a mix of A and B antigens. Because the\\nrobe’s bloodstain was consistent with Parker, he\\nhypothesized that the B antigens came from Panah’s saliva.\\n\\n Moore also briefly testified about evidence collected\\nfrom the sexual assault examination, although he did not\\nconduct it. He acknowledged that the oral and anal swabs\\nhad not produced any signs of semen, nor was “the presence\\nof semen conclusively” found anywhere on Parker’s body.\\n\\n On cross-examination, Moore admitted that he could not\\n“establish any certainty” that either Parker or Panah was a\\ncontributor to the stains because of the relatively high\\nstatistical frequency of the A and B antigens matching other\\npeople. Moore further testified that his theory would not\\nbear out if the A and B antigens on the three items came from\\none person with AB blood type. 2\\n\\n\\n\\n The defense consisted of testimony from Dr. John\\nPalmer, the emergency room doctor who treated Panah the\\nday after Parker went missing, and several character\\nwitnesses. Dr. Palmer reported that Panah was “acutely\\n 2\\n After the State presented its case, the court granted the defense’s\\nmotion for acquittal on the substantive charges of kidnapping and the\\nspecial circumstance allegation of kidnapping.\\n\\f10 PANAH V. CHAPPELL\\n\\npsychotic” and suicidal when brought into the emergency\\nroom that day.\\n\\n\\n\\n The State summarized its evidence: “We have blood\\ntyping that matches. We have the body in his suitcase in his\\ncloset, and we have statements he makes that show\\nknowledge before the body was found. We have his\\ninvolvement in the crime clearly established.” The State\\nfocused on where Parker’s body was found: “You have a\\nbody in his closet, in his suitcase. There isn’t a whole lot\\nmore you need to do after that in terms of looking and\\ninvestigating outside of the obvious, which is that Mr. Panah\\nis the person involved.” It also focused on Panah’s\\nstatements after Parker went missing: “Those aren’t crazed\\nremarks. Those are the remarks of an individual who is\\ntelling exactly what happened.” The State then reminded the\\njury of Moore’s testimony, arguing it was “not a harebrained\\nprosecution theory.” It particularly used his testimony to\\nprove the alleged oral copulation:\\n\\n We think the evidence that was presented to\\n you is very consistent with the fact that he\\n ejaculated in her mouth, that he allowed her\\n to spit it out in a Kleenex, because we have\\n the evidence of semen of his blood type, high\\n amylase content, indicating a saliva which\\n matches her blood type on the Kleenex, as\\n well as having a spattering on the bed sheet\\n of a mixture of semen and saliva – again the\\n high amylase indicating saliva – of his type B\\n and her type A.\\n\\nIt also said that “the one possible inference that can be\\ndrawn” from Moore’s testimony about the robe is that the B\\n\\f PANAH V. CHAPPELL 11\\n\\nantigens came from “the saliva of the defendant” during the\\nsodomy.\\n\\n Acknowledging that none of this was “conclusive\\nevidence,” the State argued that, “when taken with\\neverything else[, this] would indicate that there had been an\\nact of oral copulation, that there was ejaculate in Nicole\\nParker’s mouth.” The State also responded to the defense’s\\nassertion that its case was weak because of the lack of DNA,\\nclaiming that DNA testing is “usually ordered in a situation\\nwhere you don’t have other types of proof available. In this\\nsituation we have the proof available.”\\n\\n In its closing, the defense questioned the reliability of the\\nserology evidence, calling Moore’s theory “hogwash,” and\\ninsisted that the stains proved nothing. Rather, counsel\\nhighlighted the lack of DNA evidence, which “could tell us\\nwho’s the source of this stuff . . . [and] whether it, in fact,\\ncould be traced to the deceased or whether it could be traced\\nto any number of other people.”\\n\\n\\n\\n The jury convicted Panah of first-degree murder and the\\nother felonies. The jury also found true the special\\ncircumstance allegations that the murder was committed\\nwhile engaged in the crime of sodomy and lewd acts on a\\nperson under the age of fourteen. The jury did not find true\\nthe special circumstance allegation that the murder occurred\\nin the commission of oral copulation. After the penalty\\nphase, the jury returned a death sentence.\\n\\n\\n\\n On March 14, 2005, the California Supreme Court\\naffirmed Panah’s conviction and sentence, People v. Panah,\\n\\f12 PANAH V. CHAPPELL\\n\\n107 P.3d 790 (Cal. 2005), and the United States Supreme\\nCourt subsequently denied certiorari, Panah v. California,\\n546 U.S. 1216 (2006). About a year later, the California\\nSupreme Court summarily denied Panah’s first habeas\\npetition. After filing a protective habeas petition in the\\nDistrict Court for the Central District of California, Panah\\nfiled a second state habeas petition and a first amended\\npetition in the district court. The district court stayed\\nproceedings during the pendency of Panah’s state habeas\\nproceedings. On March 16, 2011, the California Supreme\\nCourt again summarily denied Panah’s second state habeas\\npetition, and the district court lifted the stay on Panah’s\\nfederal habeas proceedings.\\n\\n In his habeas petitions, Panah provided new evidence,\\nincluding two reports detailing post-conviction DNA testing\\non the stains that Moore testified about at trial. The first\\nreport (“Calandro Report”), prepared in 2004, disagreed\\nwith much of Moore’s testimony. But because the Calandro\\nReport yielded several inconclusive results, additional\\ntesting was conducted two years later, leading to the second\\nreport (“Inman Report”). Both reports doubted the\\nfoundation of Moore’s mixture theory. The Calandro Report\\ncalled “Mr. Moore’s approach . . . biased and indefensible,”\\nand the Inman Report wrote that “[n]o biological evidence\\nexists to support the hypothesis that a mixture of biological\\nfluids from Mr. Panah and Ms. Parker was present on the\\ntissue, bedsheet, or kimono.”\\n\\n More specifically, both reports “contradict[ed]” Moore’s\\ntestimony about the tissue. While Moore testified that Panah\\nand Parker were both possible contributors to the tissue stain,\\nthe reports eliminated any possibility that Parker was a\\nsource. On appeal to this court, the State concedes that\\nMoore’s tissue testimony was false.\\n\\f PANAH V. CHAPPELL 13\\n\\n However, the post-conviction testing produced\\ninconclusive results regarding the bedsheet and robe stains.\\nNeither report could definitively eliminate Parker as a\\ncontributor to several stains on the bedsheet. While they\\nconclusively found that two stains were consistent only with\\nPanah, they could not conclusively rule Parker out as a\\ncontributor to the three other bedsheet stains. Although this\\nleft open the possibility that Moore’s mixture theory was\\ncorrect, the reports opined that this at minimum refuted his\\nassumption that “Ms. Parker ‘spit out’ ejaculate onto the bed\\nsheet” because one would then expect to “detect Ms.\\nParker’s DNA in significant quantities on the bed sheet.” As\\nfor the robe, the reports agreed with Moore that the\\nbloodstain was consistent with Parker’s blood type. But,\\nunlike Moore’s serology testimony, they found no trace of\\nPanah’s DNA on the robe.\\n\\n Panah’s habeas petitions also included declarations from\\nhis three trial lawyers. Each declaration acknowledged there\\nhad been almost no pre-trial investigation and only a limited\\npenalty-phase investigation, nor were any experts retained to\\nindependently analyze the State’s serology and pathology\\nevidence. Instead, “all of [defense counsel’s] efforts had\\ngone into the aborted settlement.”\\n\\n On November 14, 2013, the district court denied Panah’s\\npetition. As for Panah’s Napue v. Illinois, 360 U.S. 264\\n(1959), claim, the district court held that, even if post-\\nconviction DNA testing rendered a portion of Moore’s\\ntestimony false, the California Supreme Court could have\\nreasonably concluded that it did not render all of the\\ntestimony false and that his testimony was immaterial in\\nlight of the other evidence. In this discussion, the district\\ncourt also rejected Panah’s claim of ineffective assistance for\\nfailure to investigate the State’s forensic evidence because\\n\\f14 PANAH V. CHAPPELL\\n\\nthe California Supreme Court reasonably could have\\nconcluded that Panah was not prejudiced.\\n\\n The district court granted a certificate of appealability\\n(“COA”) for Panah’s Napue claim, discussed below in\\nsection III. On appeal, Panah has raised a number of\\nuncertified issues in his opening brief, which we treat as a\\nrequest to expand the COA. 9th Cir. R. 22-1(e). After\\nasking the State to respond to several of the uncertified\\nissues, we expand the COA to encompass Panah’s guilt-\\nphase ineffective assistance claim, addressed below in\\nsection IV, but deny Panah’s request to expand the COA as\\nto the other uncertified claims. We evaluate Panah’s two\\ncertified claims in turn.\\n\\n\\n\\n We review the district court’s denial of habeas relief de\\nnovo. Lewis v. Mayle, 391 F.3d 989, 995 (9th Cir. 2004).\\n\\n Because Panah filed his federal habeas petition after\\nApril 24, 1996, it is subject to the Antiterrorism and\\nEffective Death Penalty Act of 1996 (“AEDPA”). See\\nMurray v. Schriro, 745 F.3d 984, 996 (9th Cir. 2014). Under\\nAEDPA, we may grant relief only if the adjudication\\n“resulted in a decision that was contrary to, or involved an\\nunreasonable application of, clearly established Federal law,\\nas determined by the Supreme Court of the United States,”\\n28 U.S.C. § 2254(d)(1), or “was based on an unreasonable\\ndetermination of the facts in light of the evidence presented\\nin the State court proceeding,” id. § 2254(d)(2).\\n\\n Although we typically “look through” a summary\\ndisposition to the last reasoned state court decision, Ylst v.\\nNunnemaker, 501 U.S. 797, 806 (1991), here there is no\\nreasoned state court decision addressing either certified\\n\\f PANAH V. CHAPPELL 15\\n\\nclaim. Therefore, we independently review the record to\\ndetermine whether the California Supreme Court had any\\nreasonable basis to deny Panah relief. See Reis-Campos v.\\nBiter, 832 F.3d 968, 974 (9th Cir. 2016) (citing Harrington\\nv. Richter, 562 U.S. 86, 102 (2011)).\\n\\n\\n\\n Relying on the two post-conviction DNA reports, Panah\\ncontends that he was prejudiced by the State’s presentation\\nof Moore’s serology testimony, which he argues the State\\nknew was false or misleading.\\n\\n In Napue, the Supreme Court held “that a conviction\\nobtained through use of false evidence, known to be such by\\nrepresentatives of the State, must fall under the Fourteenth\\nAmendment.” 360 U.S. at 269. Nonetheless, a Napue claim\\nsucceeds only if three elements are satisfied. See United\\nStates v. Zuno-Arce, 339 F.3d 886, 889 (9th Cir. 2003).\\nFirst, the testimony or evidence in question must have been\\nfalse or misleading. See id.; see also Alcorta v. Texas,\\n355 U.S. 28, 31 (1957) (considering whether “testimony,\\ntaken as a whole, gave the jury [a] false impression”).\\nSecond, the State must have known or should have known\\nthat it was false or misleading. See Zuno-Arce, 339 F.3d\\nat 889; see also Maxwell v. Roe, 628 F.3d 486, 506 (9th Cir.\\n2010) (“[E]ven false evidence presented in good faith . . .\\nhardly comports with fundamental fairness.” (quoting\\nKillian v. Poole, 282 F.3d 1204, 1209 (9th Cir. 2002))). And\\nthird, because “Napue does not create a ‘per se rule of\\nreversal,’” the testimony or evidence in question must be\\nmaterial. Sivak v. Hardison, 658 F.3d 898, 912 (9th Cir.\\n2011) (quoting Jackson v. Brown, 513 F.3d 1057, 1076 (9th\\nCir. 2008)); see also Zuno-Arce, 339 F.3d at 889.\\n\\f16 PANAH V. CHAPPELL\\n\\n Materiality under Napue requires a “lesser showing of\\nharm . . . than under ordinary harmless error review.” Dow\\nv. Virga, 729 F.3d 1041, 1048 (9th Cir. 2013). But, after\\nweighing the effect of alleged Napue violations collectively,\\nsee Phillips v. Ornoski, 673 F.3d 1168, 1189 (9th Cir. 2012),\\nthere still needs to be a “reasonable likelihood that the false\\ntestimony could have affected the judgment of the jury.”\\nHayes v. Brown, 399 F.3d 972, 985 (9th Cir. 2005) (en banc)\\n(quoting Belmontes v. Woodford, 350 F.3d 861, 881 (9th Cir.\\n2003)). Thus, a Napue claim fails if, absent the false\\ntestimony or evidence, the petitioner still “received a fair\\ntrial, understood as a trial resulting in a verdict worthy of\\nconfidence.” Hayes, 399 F.3d at 984 (quoting Hall v. Dir.\\nof Corr., 343 F.3d 976, 984 (9th Cir. 2003) (per curiam)).\\n\\n Even if we assume there was no reasonable basis for the\\nstate court to deny Panah’s claim as to the first two Napue\\nrequirements, we cannot say that it would be unreasonable\\nto conclude that Moore’s testimony was immaterial. See\\nTowery v. Schriro, 641 F.3d 300, 308 (9th Cir. 2010) (not\\nreviewing all three requirements because petitioner’s\\n“argument fails at the second Napue prong”). The State\\npresented a powerful case of Panah’s guilt, with substantial\\nevidence linking him to Parker’s murder and sexual assault.\\nMoore’s testimony was just one – and not a crucial – piece\\nof that presentation. Because the “verdict” is still reasonably\\n“worthy of confidence,” Phillips, 673 F.3d at 1189 (quoting\\nSivak, 658 F.3d at 912), we hold that the California Supreme\\nCourt would not have erred in finding no “reasonable\\nlikelihood that [Moore’s testimony] could have affected the\\njudgment of the jury.” Hayes, 399 F.3d at 985 (quoting\\nBelmontes, 350 F.3d at 881); see also Phillips, 673 F.3d\\nat 1190 (“[T]he prosecution’s Napue violations, although\\n‘pernicious’ and ‘reprehensible,’ were not material to\\n\\f PANAH V. CHAPPELL 17\\n\\n[petitioner’s] conviction of first-degree murder.” (quoting\\nHayes, 399 F.3d at 981)).\\n\\n Even setting aside Moore’s testimony, the case against\\nPanah was devastating. Parker’s naked body was found in a\\nsuitcase in Panah’s bedroom closet. Blood stains matching\\nParker’s blood type – according to both Moore and the post-\\nconviction reports – were found on Panah’s robe. Moreover,\\nPanah’s behavior on the day of and after Parker went missing\\nwas highly suspicious. Hours after she disappeared, Panah\\ntried to divert her father away from where her body was later\\nfound. Then, after the police called Panah at work to ask if\\nhe had seen Parker, he left without explanation, later telling\\nhis manager and Campbell that he was in serious trouble.\\nThe next morning, Panah even confided in Campbell that he\\nwas involved in Parker’s death. See Arizona v. Fulminante,\\n499 U.S. 279, 296 (1991) (“A confession is like no other\\nevidence. Indeed, ‘the defendant’s own confession is\\nprobably the most probative and damaging evidence that can\\nbe admitted against him.’” (quoting Bruton v. United States,\\n391 U.S. 123, 139 (1968) (White, J., dissenting))). And then,\\nparticularly incriminating in light of his statements, Panah\\nattempted suicide the morning after Parker’s disappearance.\\n\\n The jury then heard Dr. Heuser’s impactful testimony\\nabout the crime itself. She described Parker’s extensive\\ninjuries from the sexual assault, including significant trauma\\nto Parker’s vaginal and rectal areas, indicative of digital and\\npenile penetration. In addition, Dr. Heuser described the\\nviolent nature of the assault: Parker was hit with “blunt\\nforce,” consistent with her “head striking . . . a wall or a\\nfloor” or being hit with a “fist.” As a result, Parker’s brain\\nwas swollen from this “significant impact,” and she had\\nbruises – some caused by “manual strangulation” – along her\\nface, neck, arms, buttocks, and legs, and scratches on her\\n\\f18 PANAH V. CHAPPELL\\n\\nlegs. Dr. Heuser cited strangulation and sodomy as “the\\nmost lethal injuries.”\\n\\n Although Dr. Heuser’s testimony did not directly\\nimplicate Panah, it was nonetheless critical. It established\\nthat a crime – and a brutal one – took place. When added to\\nthe evidence the jury heard about where Parker’s body was\\nfound and Panah’s statements, it was more than sufficient for\\nthe jury to render a guilty verdict. And, while Panah\\ncontends that Moore’s testimony was prejudicial because of\\nits at-times graphic descriptions, particularly of oral\\ncopulation, Dr. Heuser’s testimony offered an even more\\ngraphic and detailed description of the entire sexual assault\\nand murder. As the State itself said, Dr. Heuser’s testimony\\nis “probably the most telling evidence of what happened.”\\n\\n The state court also reasonably could have found\\nMoore’s testimony to be an immaterial part of the State’s\\ncase because it offered the jury, at most, hypotheticals and\\nwavering findings. Unlike Dr. Heuser’s straightforward\\ntestimony about Parker’s injuries, Moore acknowledged that\\nhis findings rested on a number of assumptions, such as that\\nthe A and B antigens came from two people rather than one\\nwith AB blood type. And even if he was correct that the A\\nand B antigens came from two people, Moore neither could\\ndefinitively say they came from Parker and Panah, nor could\\nhe even narrow the pool of possible matches to less than\\nhundreds of thousands of people. He said frankly on the\\nstand: “I cannot establish any certainty based on\\nconventional serology. I can only demonstrate consistency.”\\nFor this reason, Moore’s testimony was couched in\\ninconclusive terms, such as “could have originated from” or\\nwere “consistent” with. The State even acknowledged this\\nweakness in Moore’s findings in closing argument:\\n\\f PANAH V. CHAPPELL 19\\n\\n Now the question is, did a person with AB\\n blood leave . . . body fluids such as blood,\\n semen[,] and saliva, on the sheets, on the\\n toilet paper, on the robe. That is one\\n interpretation. The other interpretation, of\\n course, is that you have two separate people,\\n one of whom has type A, and one has type B.\\n\\n Therefore, at best, Moore’s testimony required the jury\\nto draw its own inferences. This was not the case of\\nimpactful expert testimony telling the jury that there was a\\none-in-a-million chance the evidence matched anyone but\\nPanah. Rather, any effect Moore’s findings may have had\\non the jury – which was reasonably none – was fully\\ndependent on the other weighty evidence presented by the\\nState. For instance, without having found Parker’s body in\\nPanah’s bedroom, no juror could have reasonably inferred\\nthat Parker was the A antigen contributor.\\n\\n The jury’s verdict removes any lingering doubt about the\\nmateriality of Moore’s testimony. If the State needed\\nMoore’s testimony at all, it was to prove the special\\ncircumstance allegation and substantive charge of oral\\ncopulation. In its closing, for instance, the State referred to\\nMoore’s findings of mixed bodily fluids on the tissue and\\nbedsheet to prove this sexual act. Yet, the jury did not find\\ntrue the special circumstance allegation that Parker’s murder\\nwas committed while in the commission of oral copulation.\\nAlthough the jury did find Panah guilty of the felony of oral\\ncopulation, the verdict is still telling. A reasonable\\ninterpretation of the jury’s rejection of this special\\ncircumstance is that the jury was not entirely persuaded by\\nMoore’s mixture theory. In contrast, this highlights the\\neffectiveness of Dr. Heuser’s testimony. The State relied on\\nher findings to prove the special circumstance allegations\\n\\f20 PANAH V. CHAPPELL\\n\\nand substantive charges of sodomy and lewd acts, which the\\njury found to be true.\\n\\n Still, Panah contends that Moore’s testimony was\\nmaterial because it was the only evidence identifying him as\\nthe perpetrator. Although creative, this argument makes\\nlittle headway. It ignores the substantial evidence tying\\nPanah to Parker’s murder and sexual assault. This was not a\\ncase where the police had no leads on a suspect. Nor was it\\na case where the prosecution needed Moore’s serology\\nevidence to place Panah at the crime scene. Rather, as the\\nState emphasized in its closing: “You have a body in his\\ncloset, in his suitcase. There isn’t a whole lot more you need\\nto do after that in terms of looking and investigating outside\\nof the obvious, which is that Mr. Panah is the person\\ninvolved.” And, as previously mentioned, Panah’s own\\nadmissions linked him to the assault. Thus, even without\\nMoore’s testimony, the State had no difficulty proving\\nidentity.\\n\\n For these reasons, Panah’s reliance on Miller v. Pate,\\n386 U.S. 1 (1967), is unavailing. In Miller, the petitioner\\nwas also charged with the murder and sexual assault of a\\nyoung girl without any eyewitnesses to the crime. Id. at 2–\\n3. The State’s only evidence linking petitioner to the victim\\nwas male underwear, which an expert said had blood stains\\non it matching the victim, found near the crime scene. Id.\\nat 3–4. Post-conviction testing proved the underwear stains\\nwere paint, not blood, and that the State had known this at\\ntrial. Id. at 5–6. The Supreme Court reversed petitioner’s\\nconviction because, while the State successfully used the\\nunderwear as “an important link in the chain of\\ncircumstantial evidence against the petitioner,” in reality it\\nwas “virtually valueless as evidence.” Id. at 4, 6.\\n\\f PANAH V. CHAPPELL 21\\n\\n Miller is clearly distinguishable. First, although the\\nparties agree that the tissue testimony was false, post-\\nconviction testing did not render the majority of Moore’s\\ntestimony false. Neither post-conviction report conclusively\\nrefuted his findings as to the bedsheet or robe stains.\\nTherefore, unlike in Miller, the evidence in question here did\\nnot become “virtually valueless” to convict. Id. at 6.\\nSecond, Moore’s testimony was not the “important link” in\\nproving Panah’s guilt. Id. at 4. The State did not even need\\nMoore’s testimony to convict Panah. This case differs\\nsignificantly from Miller because here the state court could\\nreasonably rely on an abundance of other evidence to still\\nhave confidence in the conviction.\\n\\n Panah’s reliance on Alcorta, 355 U.S. 28, also is\\nmisplaced. There, the Court overturned petitioner’s\\nconviction because the false “testimony was seriously\\nprejudicial to petitioner” and “tended squarely to refute his\\nclaim.” Id. at 31. Had the false testimony been corrected,\\npetitioner’s defense would have been corroborated. But,\\nhere, even if Moore’s testimony had been corrected or\\nentirely excluded, the jury would not have heard a\\nsignificantly different presentation of evidence. At most,\\nalthough we think unlikely, the State’s case may have\\nbecome marginally weaker.\\n\\n Rather, this case is quite similar to Sivak, 658 F.3d 898,\\nin which we rejected a guilt-phase Napue claim because the\\ncourt had “full confidence that the jury would still have\\nconvicted.” Id. at 913. There, we held: “[E]ven if the jury\\ndisbelieved [the false testimony] entirely . . . there still is no\\n‘reasonable likelihood that the false testimony could have\\naffected the judgment of the jury.’” Id. at 914 (quoting\\nJackson, 513 F.3d at 1076). As in Sivak, we think the\\nCalifornia Supreme Court reasonably still could have had\\n\\f22 PANAH V. CHAPPELL\\n\\n“full confidence” that the jury would have returned the same\\nverdict even in absence of Moore’s testimony. Id. at 913.\\n\\n In sum, we do not think Moore’s testimony was critical\\nin convicting Panah. Excluding Moore’s testimony, the\\nState’s case was still devastating and largely unchallenged.\\nMoore’s testimony was certainly not “the centerpiece of the\\nprosecution’s case.” Hayes, 399 F.3d at 985. Rather, in light\\nof the overwhelming evidence against Panah and the jury’s\\nrejection of the oral copulation special circumstance, it is\\nreasonable to conclude that his testimony had essentially no\\neffect on the jury’s decision making. Cf. Dow, 729 F.3d\\nat 1049–50 (concluding that false testimony was material\\nbecause “[t]he evidence against [petitioner] was weak”);\\nMaxwell, 628 F.3d at 508 (holding that false testimony was\\nmaterial because it came from a “‘make-or-break’ witness\\nfor the State” and there was a “paucity of other evidence”);\\nHall, 343 F.3d at 984 (reversing under Napue “[i]n light of\\nthe already scant evidence on which the conviction was\\nbased”). As such, we conclude that the California Supreme\\nCourt reasonably rejected Panah’s Napue claim.\\n\\n\\n\\n We next turn to Panah’s guilt-phase ineffective\\nassistance claim. To prevail, Panah must show that his\\ncounsel’s performance “fell below an objective standard of\\nreasonableness” and “a reasonable probability that, but for\\ncounsel’s unprofessional errors, the result of the proceeding\\nwould have been different.” Strickland v. Washington,\\n466 U.S. 668, 688, 694 (1984). “The likelihood of a\\ndifferent result,” however, “must be substantial, not just\\nconceivable.” Richter, 562 U.S. at 112. This already\\nimposing standard becomes doubly difficult to satisfy once\\nAEDPA deference is tacked on. See Cullen v. Pinholster,\\n563 U.S. 170, 189 (2011) (“[T]he benchmark for judging any\\n\\f PANAH V. CHAPPELL 23\\n\\nclaim of ineffectiveness [under Strickland] must be whether\\ncounsel’s conduct so undermined the proper functioning of\\nthe adversarial process that the trial cannot be relied on as\\nhaving produced a just result.” (quoting Strickland, 466 U.S.\\nat 686)).\\n\\n Panah argues that his trial counsel was ineffective for\\nfailing to conduct a reasonable investigation and therefore\\nnot rebutting the State’s serology and pathology evidence.\\nIn support of his claim, Panah notes that his counsel never\\nretained an expert to independently analyze the pathology\\nand serology evidence or to testify at the guilt phase.\\nInstead, defense counsel essentially seemed to accept the\\nState’s evidence as true. In post-conviction proceedings,\\ncounsel acknowledged that his inordinate focus on\\nsettlement resulted in too little, if even any, pre-trial\\ninvestigation.\\n\\n While we are instructed to “avoid the temptation to\\nsecond-guess [counsel’s] performance,” Mayfield v.\\nWoodford, 270 F.3d 915, 927 (9th Cir. 2001) (en banc), we\\nare concerned with defense counsel’s lack of pre-trial\\ninvestigation. See Duncan v. Ornoski, 528 F.3d 1222, 1235\\n(9th Cir. 2008) (“[L]awyers [have] considerable discretion\\nto make strategic decisions about what to investigate, but\\nonly after those lawyers ‘have gathered sufficient evidence\\nupon which to base their tactical choices.’” (quoting\\nJennings v. Woodford, 290 F.3d 1006, 1014 (9th Cir.\\n2002))). But we “need not determine whether counsel’s\\nperformance was deficient before examining the prejudice\\nsuffered by the defendant as a result of the alleged\\ndeficiencies.” Strickland, 466 U.S. at 697. Here, even\\nassuming counsel’s performance was deficient, we cannot\\nsay – in light of the overwhelming evidence of Panah’s guilt\\nand the deference we owe the state court judgment – that the\\n\\f24 PANAH V. CHAPPELL\\n\\nCalifornia Supreme Court would have erred in finding no\\n“reasonable probability that, but for counsel’s\\nunprofessional errors, the result of the proceeding would\\nhave been different.” Id. at 694.\\n\\n After weighing counsel’s deficiencies cumulatively with\\n“the strength of the government’s case,” Rios v. Rocha,\\n299 F.3d 796, 808–09 (9th Cir. 2002) (quoting Eggleston v.\\nUnited States, 798 F.2d 374, 376 (9th Cir. 1986)), we believe\\nthe state court reasonably rejected Panah’s assertion that the\\ntrial’s outcome “would have been dramatically different”\\nhad counsel’s performance not been deficient. Our\\nreasoning on prejudice bears significant similarities to why\\nwe reject Panah’s Napue claim. We do not wish to harp on\\nwhat was detailed in the preceding section, but the State had\\na uniquely strong case against Panah. Parker’s body was\\nfound in his bedroom; Panah’s behavior the day of and after\\nher disappearance was incriminating; Panah admitted his\\nown involvement in her death; and Parker’s serious physical\\ninjuries, including to her genitalia, were well-documented.\\nDefense counsel even acknowledged that this evidence was\\nthe foundation of the State’s case: “The critical pieces of\\nevidence are obviously that the child’s body is found in Mr.\\nPanah’s closet, her naked body with a considerable amount\\nof blood. There is evidence . . . that the child was beaten.”\\n\\n It is, therefore, inconceivable, even had defense counsel\\nindependently investigated the serology and pathology\\nevidence, that the jury would have reached a different\\nverdict. See Williams v. Filson, 908 F.3d 546, 570 (9th Cir.\\n2018) (“We have long recognized . . . that ‘prejudice\\nresulting from ineffective assistance of counsel must be\\n“considered collectively, not item by item.”’” (quoting Doe\\nv. Ayers, 782 F.3d 425, 460 n.62 (9th Cir. 2015))). Even if\\nthe weaknesses in Moore’s and Dr. Heuser’s findings that\\n\\f PANAH V. CHAPPELL 25\\n\\ncame to light post-conviction were raised at trial, that would\\nhave done nothing to reasonably change the outcome. Not\\nonly was there no “strong, unequivocal, exculpatory\\nevidence available,” Rios, 299 F.3d at 813, but there was\\nnothing to substantively challenge the serology or pathology\\nevidence. It is true that counsel could have told the jury that\\nMoore’s findings as to the tissue were wrong. But counsel\\ncould not have refuted Moore’s findings as to the bedsheet\\nand robe stains, and even a different outcome on the felony\\nof oral copulation would not affect Panah’s guilty verdict\\nand death sentence. Also, Panah’s contention that he was\\nprejudiced by counsel’s failure to rebut Moore’s mixture\\ntheory because it put him at the crime scene fails to\\nacknowledge that finding Parker’s body in his bedroom\\nalone was sufficient to do that. Therefore, whatever rebuttal\\nof the State’s expert witnesses that Panah believes he was\\ndeprived of and thus prejudiced by would not have overcome\\nthe other significant evidence of guilt.\\n\\n Moreover, for however deficient defense counsel’s\\ninvestigation was, the state court also could have reasonably\\nfound no prejudice because counsel adequately challenged\\nthe State’s expert witnesses on the stand. See Richter,\\n562 U.S. at 111 (“In many instances cross-examination will\\nbe sufficient to expose defects in an expert’s presentation.”).\\nDuring Moore’s cross-examination, defense counsel\\nattempted to cast doubt on his findings. His questions\\npushed Moore to acknowledge the high statistical\\nprobability of persons other than Panah and Parker\\ncontributing to the stains, and that Moore could not even\\n“determine when th[e] stains were deposited.” Counsel also\\nremarked that Moore offered nothing more than\\n“inconclusive blood typing,” hypothesized that Moore\\nsimply “construct[ed] some sort of theory whereby you can\\nlink that blood to Mr. Panah or to the deceased,” and\\n\\f26 PANAH V. CHAPPELL\\n\\nquestioned why the jury was “not offered DNA evidence.”\\nNotably, even one of the post-conviction reports described\\nthe cross-examination of Moore as “reasonably successful.”\\n\\n Similarly, we do not think there was anything in the\\ndefense’s questioning of Dr. Heuser that would have\\nchanged the outcome. There was nothing to rebut her\\ndetailed testimony about Parker’s extensive injuries. Panah\\ninstead alleges that he was prejudiced by counsel’s failure to\\nrebut Dr. Heuser’s assessment of the cause and time of death.\\nBut at trial, counsel did flag concerns with this part of her\\ntestimony. Thus, we agree with the district court’s\\nconclusion that, “in light of the ‘setting of a sexual assault,’”\\nfurther challenging Dr. Heuser’s testimony on the cause of\\ndeath “would have been no more palatable to the jury.”\\n\\n As a result, the state court would not have unreasonably\\ndetermined that counsel’s casting-doubt strategy was\\nappropriate, even effective, and thus found a lack of\\nprejudice. With little to do about the State’s formidable\\nevidence against Panah, counsel still sought to inform the\\njury of weaknesses in the experts’ testimony. See id. (“When\\ndefense counsel does not have a solid case, the best strategy\\ncan be to say that there is too much doubt about the State’s\\ntheory for a jury to convict.”). And here the strategy even\\nseems to have been somewhat successful, as the jury rejected\\nthe special circumstance allegation of oral copulation. Like\\nin the Napue analysis, this is probative evidence that the jury\\ndid not give persuasive weight to the serology testimony,\\npresumably because of defense counsel’s strategy and cross-\\nexamination.\\n\\n Our analysis is further guided by Richter, in which the\\nSupreme Court rejected a similar ineffective assistance\\nclaim based on counsel’s failure to present an expert to rebut\\nthe State’s forensic evidence. Id. at 111–13. The Court’s\\n\\f PANAH V. CHAPPELL 27\\n\\nreasoning is almost entirely applicable here. Holding that\\nthe petitioner was not prejudiced, the Court concluded that\\nthere was little chance of a different outcome because the\\npost-conviction evidence did not exonerate petitioner and\\nbecause some of petitioner’s post-conviction evidentiary\\nconcerns were already raised by counsel at trial before the\\njury. The Court went on to determine – as is also applicable\\nhere – that the petitioner also did not prove any likelihood of\\na different outcome because he had done nothing to rebut the\\nother “sufficient conventional circumstantial evidence\\npointing to [his] guilt.” 3 Id. at 113.\\n\\n We do not hold that counsel should not have done more.\\nBut, based on the particular facts before us, we recognize\\nthere was “nothing more than a theoretical possibility” of a\\ndifferent verdict. Id. at 112. The evidence of Panah’s guilt\\nwas so strong that there remained an “ample basis for the\\nCalifornia Supreme Court to think any real possibility of\\n[Panah’s] being acquitted was eclipsed by the remaining\\nevidence pointing to guilt.” Id. at 113. Therefore, we affirm\\nthe district court’s denial of Panah’s ineffective assistance\\nclaim.\\n\\n AFFIRMED.\\n\\n\\n\\n\\n 3\\n Panah’s attempt to analogize his case to several out-of-circuit cases\\nfalls short. For instance, in Elmore v. Ozmint, 661 F.3d 783, 871–72 (4th\\nCir. 2011), the Fourth Circuit held that the petitioner was prejudiced by\\ncounsel’s failure to investigate any of the prosecution’s forensic\\nevidence. But, there, “[t]he case was a real ‘who-done-it,’” and, with a\\nproper “investigation, the jury undeniably would have seen a drastically\\ndifferent – and significantly weaker – prosecution case.” Id. at 861, 870.\\nNone of that is true here.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4090,"cells":{"id":{"kind":"number","value":4654479,"string":"4,654,479"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Geraldine Nicholson v. Miguel Gutierrez"},"decision_date":{"kind":"string","value":"2019-08-21"},"docket_number":{"kind":"string","value":"17-56648"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Andrew J. Kleinfeld, R. Guy Cole, Jr., * and Jacqueline H. Nguyen, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"NGUYEN, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nGERALDINE NICHOLSON; JOSE No. 17-56648\\nFERNANDO HUERTA; J. H., a minor\\nby and through his guardian ad litem, D.C. No.\\nJose Fernando Huerta; J. N. G., a 2:15-cv-07594-\\nminor by and through his guardian DDP-RAO\\nad litem, Geraldine Nicholson,\\n Plaintiffs-Appellees,\\n OPINION\\n v.\\n\\nCITY OF LOS ANGELES; EVERARDO\\nAMARAL, individually and in his\\nofficial capacity as a Police Officer\\nof the Los Angeles Police\\nDepartment,\\n Defendants,\\n\\n and\\n\\nMIGUEL GUTIERREZ, individually\\nand in his official capacity as a\\nPolice Officer of the Los Angeles\\nPolice Department,\\n Defendant-Appellant.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Dean D. Pregerson, District Judge, Presiding\\n\\f2 NICHOLSON V. GUTIERREZ\\n\\n Argued and Submitted March 4, 2019\\n Pasadena, California\\n\\n Filed August 21, 2019\\n\\n Before: Andrew J. Kleinfeld, R. Guy Cole, Jr., *\\n and Jacqueline H. Nguyen, Circuit Judges.\\n\\n Opinion by Judge Nguyen\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel reversed in part and affirmed in part the\\ndistrict court’s denial of qualified immunity to a Los Angeles\\nPolice Department officer in an action brought pursuant to\\n42 U.S.C. § 1983 for violations of plaintiffs’ Fourth\\nAmendment rights to be free from excessive force and\\nunreasonable seizure and violations of their Fourteenth\\nAmendment substantive due process rights.\\n\\n Plaintiffs were among a group of teenagers who had met\\nin an alleyway near their school to listen to and sing rap\\nmusic. One of the teenagers, plaintiff J.N.G., was shot by\\ndefendant Gutierrez after Gutierrez mistook a plastic Airsoft\\nreplica gun held by one of the other teenagers for an actual\\n\\n *\\n The Honorable R. Guy Cole, Jr., United States Circuit Judge for\\nthe U.S. Court of Appeals for the Sixth Circuit, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f NICHOLSON V. GUTIERREZ 3\\n\\ngun. After the shooting, officers detained the group for over\\nfive hours while they investigated. J.N.G. and J.H. filed a\\nlawsuit and the district court denied qualified immunity on\\nplaintiffs’ Fourth and Fourteenth Amendment claims.\\n\\n Addressing the Fourth Amendment claim, the panel\\nagreed with the district court that under the circumstances,\\nplaintiffs’ continued detention for five hours after the\\nshooting—well after any probable cause would have\\ndissipated—and the use of handcuffs throughout the\\nduration of the detention violated plaintiffs’ clearly\\nestablished Fourth Amendment rights to be free from\\nunlawful arrest and excessive force. The panel rejected\\nGutierrez’s argument that while he participated in the initial\\nhandcuffing and detention, he was not responsible for any\\nsubsequent constitutional violation because he played no\\nrole in that conduct. The panel held that an officer can be\\nheld liable where he is just one participant in a sequence of\\nevents that gives rise to a constitutional violation. Here,\\nviewing the evidence in the light most favorable to plaintiffs,\\nGutierrez was more than a “mere bystander” in the alleged\\nconstitutional violations. The panel affirmed the district\\ncourt’s denial of qualified immunity on the Fourth\\nAmendment violations because, ultimately, a reasonable\\njury could conclude that Gutierrez played an integral role in\\nthe unlawfully prolonged detention and sustained\\nhandcuffing of plaintiffs.\\n\\n Addressing the Fourteenth Amendment substantive due\\nprocess claim, the panel held that, viewing the totality of the\\nevidence in the light most favorable to the plaintiffs, the\\nshooting violated plaintiffs’ due process rights. Under the\\ncircumstances, a rational finder of fact could find that\\nGutierrez’s use of deadly force shocked the conscience and\\nwas unconstitutional under the Fourteenth Amendment.\\n\\f4 NICHOLSON V. GUTIERREZ\\n\\nNevertheless, the panel held that because no analogous case\\nexisted at the time of the shooting, the district court erred by\\ndenying Gutierrez qualified immunity for this claim. The\\npanel accordingly reversed the district court and remanded\\nfor an entry of qualified immunity on the Fourteenth\\nAmendment claim.\\n\\n\\n COUNSEL\\n\\nDenise L. Rocawich (argued) and James R. Touchstone,\\nJones & Mayer, Fullerton, California, for Defendant-\\nAppellant.\\n\\nHerbert-John S. Hayden (argued) and John W. Harris, Harris\\n& Associates, Los Angeles, California, for Plaintiffs-\\nAppellees.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n On the morning of February 10, 2015, four teenagers met\\nin an alleyway near their school to listen to and sing rap\\nmusic. As the teenagers—Michael Sanders, Abdul Wooten,\\nJ.N.G., and J.H. 1—stood in a tight circle dancing and\\nrapping, Sanders was holding a plastic Airsoft replica gun\\nwith a bright orange tip as a prop. Just as they turned off the\\nmusic and were getting ready to head to school, J.N.G. was\\n\\n 1\\n Throughout the record, J.N.G. (Jamar Nicholson Green) and J.H.\\n(Jason Huerta) were referred to by their initials because they were minors\\nat the time of the incident and for some time after filing this lawsuit. For\\nconsistency, we also refer to them by their initials.\\n\\f NICHOLSON V. GUTIERREZ 5\\n\\nshot by Officer Michael Gutierrez of the Los Angeles Police\\nDepartment (“LAPD”). Officer Gutierrez fired his weapon\\nbecause he mistook Sanders’s replica gun for an actual gun.\\nGutierrez fired multiple shots, one of which hit J.N.G. in the\\nback. After the shooting, officers detained the group for over\\nfive hours while they investigated.\\n\\n J.N.G. and J.H. (collectively, “Plaintiffs”) filed a lawsuit\\nagainst the officers, the LAPD, and the City of Los Angeles,\\nalleging violations of the Fourth and Fourteenth\\nAmendments and various state laws. The district court\\ndenied qualified immunity on two of Plaintiffs’\\nconstitutional claims. Gutierrez appeals. We affirm in part\\nand reverse in part.\\n\\n BACKGROUND 2\\n\\n At around 7:15 a.m. on February 10, 2015, J.N.G., J.H.,\\nMichael Sanders, and Abdul Wooten met in an alley at the\\ncorner of 10th Avenue and Florence Avenue in Los Angeles,\\nCA, a few blocks from their high school. They regularly\\ngathered in that alleyway before and after school to listen to\\nmusic and freestyle rap. That morning, as they were rapping\\nand dancing in a circle, Sanders was holding a plastic toy\\ngun with a bright orange tip. J.N.G., J.H., and Wooten\\nmaintain that Sanders kept the gun pointed downward\\naround waist-level and did not fire the gun that morning. At\\napproximately 7:40 a.m., the teenagers turned off the music\\nand began preparing to head to school.\\n\\n\\n\\n\\n 2\\n At this stage of the proceedings, we view the facts in the light most\\nfavorable to Plaintiffs. George v. Morris, 736 F.3d 829, 836 (9th Cir.\\n2013) (9th Cir. 2013).\\n\\f6 NICHOLSON V. GUTIERREZ\\n\\n Around this time, Officer Everardo Amaral was driving\\ndown 10th Avenue in an unmarked car with his partner,\\nOfficer Gutierrez. From the passenger seat, Gutierrez “saw\\na person (later identified as Michael Sanders) pointing . . . a\\nblue steel handgun at another person (later identified as\\nPlaintiff J.H.).” Gutierrez, believing that J.H. was “being\\nrobbed at gun point or was about to be murdered,” yelled\\n“Gun, gun, gun!” Amaral stopped the vehicle south of the\\nalley on 10th Avenue. Without conferring with Amaral,\\nGutierrez immediately jumped out of the car and ran into the\\nalley. Amaral parked the car and followed Gutierrez.\\nNeither officer was in uniform.\\n\\n Gutierrez claims he identified himself as an LAPD\\nofficer and commanded Sanders to drop the gun. However,\\nJ.H., J.N.G., and Wooten all contend that Gutierrez did not\\nidentify himself or make any verbal commands prior to\\nshooting his weapon. A few seconds after he entered the\\nalley, Officer Gutierrez fired at least three shots, one of\\nwhich hit J.N.G. in the back. J.N.G. and J.H. contend that\\nGutierrez fired his gun with one hand while running toward\\nthem, while Gutierrez stated that he fired only after stopping\\na few feet away from the group. When the shots were fired,\\nJ.H. was about to put on his school uniform, and J.N.G. was\\nspraying cologne on his face. The four of them had been\\nstanding in a tight circle, “within a foot or so of each other.”\\nSanders soon turned and dropped the toy gun, though the\\nparties dispute whether this occurred before or after\\nGutierrez fired.\\n\\n Shortly after Officer Gutierrez fired, Officer Amaral\\narrived and requested three additional units. Amaral also\\nrequested an ambulance when he realized that J.N.G. had\\nbeen shot. The officers held the group at gunpoint, face\\ndown on the ground. The parties dispute how far the\\n\\f NICHOLSON V. GUTIERREZ 7\\n\\ndropped “gun” was from the teenagers when they were on\\nthe ground, but neither officer picked it up or moved it away\\nfrom them. While on the ground, J.H. shouted that the gun\\nwas “not even a real gun” and repeatedly asked why the\\nofficers shot at them and “What did we do wrong?” The\\nofficers remained silent in response to his questions, with\\ndumbfounded expressions on their faces. Responding\\nofficers soon arrived, and they searched and handcuffed the\\ngroup. Gutierrez was “involved in the decision to handcuff\\nthem.” Officer Amaral later explained in his deposition that\\nthe detention of the boys was “[f]or [a] weapons violation.”\\nOfficers Gutierrez and Amaral were separated and\\nmonitored soon after additional units arrived on the scene.\\nJ.H. remained in handcuffs throughout the investigation,\\nwhich lasted until around 1:00 p.m., over five hours after the\\nshooting. J.N.G. also remained in handcuffs for over five\\nhours—through the duration of his hospital examination—\\nuntil detectives interrogated him.\\n\\n J.N.G., by and through his mother and guardian ad litem\\nGeraldine Nicholson, and J.H., by and through his father and\\nguardian ad litem Jose Fernando Huerta, sued Officer\\nGutierrez, Officer Amaral, the City of Los Angeles, the\\nLAPD, Chief of Police Charles Beck, and Commander\\nAndrew Smith. In addition to various state law claims,\\nPlaintiffs J.N.G. and J.H. alleged claims under 42 U.S.C.\\n§ 1983 for violations of their Fourth Amendment rights to be\\nfree from excessive force and unreasonable seizure and\\nviolations of their Fourteenth Amendment substantive due\\nprocess rights. Defendants Gutierrez, Amaral, and the City\\nof Los Angeles jointly moved for summary judgment,\\narguing in part that no constitutional violation occurred and\\nthat qualified immunity applied.\\n\\f8 NICHOLSON V. GUTIERREZ\\n\\n The district court granted in part and denied in part\\nDefendants’ joint motion for summary judgment. Relevant\\nto this appeal, the district court denied Gutierrez qualified\\nimmunity in part on Plaintiffs’ Fourth Amendment claim and\\non Plaintiffs’ Fourteenth Amendment claim. Gutierrez\\ntimely appealed.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291 to review\\nthe denial of qualified immunity at the summary judgment\\nstage. Plumhoff v. Rickard, 572 U.S. 765, 771–73 (2014).\\nBut “the scope of our review over the appeal is\\ncircumscribed” because we may not “consider questions of\\neviden[tiary] sufficiency, i.e., which facts a party may, or\\nmay not, be able to prove at trial.” Morris, 736 F.3d at 834\\n(quoting CarePartners, LLC v. Lashway, 545 F.3d 867, 875\\n(9th Cir. 2008)). “Thus, in this appeal, we are confined to\\nthe question of ‘whether the defendant[s] would be entitled\\nto qualified immunity as a matter of law, assuming all factual\\ndisputes are resolved, and all reasonable inferences are\\ndrawn, in plaintiff’s favor.’” Id. at 836 (quoting Karl v. City\\nof Mountlake Terrace, 678 F.3d 1062, 1068 (9th Cir. 2012)).\\n\\n DISCUSSION\\n\\n “Qualified immunity attaches when an official’s conduct\\ndoes not violate clearly established statutory or\\nconstitutional rights of which a reasonable person would\\nhave known.” Kisela v. Hughes, 138 S. Ct. 1148, 1152\\n(2018) (per curiam). Once a defendant has raised qualified\\nimmunity as a defense to a claim, a plaintiff must show\\n“(1) that the right was violated; and (2) that the right was\\nclearly established at the time of the alleged misconduct.”\\nIsayeva v. Sacramento Sheriff’s Dep’t, 872 F.3d 938, 946\\n(9th Cir. 2017). Although a right is not clearly established\\n\\f NICHOLSON V. GUTIERREZ 9\\n\\nwhere merely defined “at a high level of generality,”\\nqualified immunity does not “require a case directly on\\npoint.” Kisela, 138 S. Ct. at 1152. Instead, the “focus is on\\nwhether the officer had fair notice that her conduct was\\nunlawful,” id., for example, through “any cases of\\ncontrolling authority in their jurisdiction at the time of the\\nincident,” Wilson v. Layne, 526 U.S. 603, 617 (1999).\\n\\n We now turn to each of Plaintiffs’ constitutional claims.\\n\\n I. Fourth Amendment Claim\\n\\n Plaintiffs allege that, after the shooting, Defendants\\nviolated their Fourth Amendment rights by unlawfully\\narresting them, using excessive force, and prolonging their\\ndetention. 3 In denying Officer Gutierrez qualified\\nimmunity, the district court concluded that immediately after\\nthe shooting, an investigatory stop was reasonable while the\\nofficers assessed the situation. But “[a]t some point, the\\ndetention evolved into a full-fledged arrest that required\\nprobable cause that J.H. and J.N.G. had been engaged in\\ncriminal activity.” The district court further concluded that\\n“a reasonable jury could determine that the sustained\\nhandcuffing of J.H. and J.N.G. . . . constituted excessive\\nforce.”\\n\\n\\n 3\\n Plaintiffs also alleged that the shooting itself also violated their\\nFourth Amendment rights. The district court granted summary judgment\\nin favor of Defendants on this claim, finding that the teenagers were not\\n“seized” by Gutierrez’s gunfire under the Fourth Amendment because\\nGutierrez intended to use deadly force against Sanders, not against\\nJ.N.G. or J.H. See United States v. Al Nasser, 555 F.3d 722, 728 (9th\\nCir. 2009). Plaintiffs do not challenge this ruling and thus the question\\nof whether the shooting violated their Fourth Amendment rights is not\\nbefore us in this appeal.\\n\\f10 NICHOLSON V. GUTIERREZ\\n\\n It is well-established that a “person may not be arrested,\\nor must be released from arrest, if previously established\\nprobable cause has dissipated.” United States v. Ortiz-\\nHernandez, 427 F.3d 567, 574 (9th Cir. 2005) (per curiam).\\n“As a corollary . . . of the rule that the police may rely on the\\ntotality of facts available to them in establishing probable\\ncause, they also may not disregard facts tending to dissipate\\nprobable cause.” Id. (quoting Bigford v. Taylor, 834 F.2d\\n1213, 1218 (5th Cir. 1988)). A reasonable officer would\\nknow that participation in an ongoing seizure after any\\nprobable cause had dissipated violates the Fourth\\nAmendment.\\n\\n Here, it was soon apparent to the officers that the\\nteenagers were unarmed, posed no threat to anyone, and\\nwere not engaged in any criminal activity. The incident\\noccurred in the morning right before the start of school\\nhours, and Plaintiffs had their school uniforms and\\nbackpacks. In fact, as Officer Gutierrez approached the\\nscene, J.N.G. was spraying on cologne and J.H. was donning\\nhis school uniform. Moreover, Officer Gutierrez admitted\\nthat he perceived at least J.H. to be a possible victim, not a\\nsuspect, further undermining any justification to detain him.\\nWe agree with the district court that under these\\ncircumstances, Plaintiffs’ continued detention for five\\nhours—well after any probable cause would have\\ndissipated—and the use of handcuffs throughout the\\nduration of the detention violated Plaintiffs’ clearly\\nestablished Fourth Amendment rights to be free from\\nunlawful arrest and excessive force.\\n\\n On appeal, Officer Gutierrez does not appear to dispute\\nthat the prolonged detention and handcuffing violated\\nPlaintiffs’ clearly established Fourth Amendment rights.\\nInstead, Gutierrez focuses his argument on the fact that he\\n\\f NICHOLSON V. GUTIERREZ 11\\n\\nwas separated and monitored after the shooting. Thus, his\\nargument goes, while he admittedly “was an integral\\nparticipant in the initial handcuffing and detention,” he is not\\nresponsible for any subsequent constitutional violation\\nbecause he played no role in that conduct.\\n\\n A police officer need not have been the sole party\\nresponsible for a constitutional violation before liability may\\nattach. “An officer’s liability under section 1983 is\\npredicated on his ‘integral participation’ in the alleged\\nviolation.” Blankenhorn v. City of Orange, 485 F.3d 463,\\n481 n.12 (9th Cir. 2007) (quoting Chuman v. Wright, 76 F.3d\\n292, 294–95 (9th Cir. 1996)). This theory of liability “does\\nnot require that each officer’s actions themselves rise to the\\nlevel of a constitutional violation.” Id. (quoting Boyd v.\\nBenton County, 374 F.3d 773, 780 (9th Cir. 2004)). Instead,\\nliability may attach if the officer has “some fundamental\\ninvolvement in the conduct that allegedly caused the\\nviolation.” Id. A theory of integral participation thus\\ncomports with general tort principles of causation applicable\\nto a § 1983 action: “[G]overnment officials, like other\\ndefendants, are generally responsible for the ‘natural’ or\\n‘reasonably foreseeable’ consequences of their actions.”\\nStoot v. City of Everett, 582 F.3d 910, 926 (9th Cir. 2009).\\nEven though “an intervening decision of an informed,\\nneutral decision-maker ‘breaks’ the chain of causation,” the\\nchain of causation is not broken where the intervening\\ndecision was foreseeably influenced by the defendant. Id.\\n(quoting Murray v. Earle, 405 F.3d 278, 292 (5th Cir.\\n2005)). Thus, under our case law, an officer could be held\\nliable where he is just one participant in a sequence of events\\nthat gives rise to a constitutional violation.\\n\\n For example, in Boyd v. Benton County, we held that\\neach of the officers in a search operation were liable for\\n\\f12 NICHOLSON V. GUTIERREZ\\n\\nexcessive force for the use of a flash-bang grenade, even\\nthough only one officer reached into the home and deployed\\nthe device. 374 F.3d at 780. We reasoned that the use of the\\ndevice was “part of the search operation in which every\\nofficer participated in some meaningful way.” Id.\\nMoreover, each officer “was aware of the decision . . . , did\\nnot object to it, and participated in the search operation\\nknowing the flash-bang was to be deployed.” Id. By\\ncontrast, in Torres v. City of Los Angeles, we found that a\\ndetective was not an integral participant in an allegedly\\nunlawful arrest, in part because she “was not present [at the\\narrest], and there is no evidence that [she] instructed the\\nother detectives to arrest [the plaintiff] or that any of those\\ndetectives consulted with her before making the arrest.”\\n548 F.3d 1197, 1206 (9th Cir. 2008).\\n\\n Here, viewing the evidence in the light most favorable to\\nPlaintiffs, Gutierrez was more than a “mere bystander” in the\\nalleged constitutional violations. See Chuman, 76 F.3d at\\n294. Plaintiffs alleged that Officers Amaral and Gutierrez\\n“directed the other officers to handcuff, search and arrest all\\nof us for reasons unknown to any of us.” Gutierrez himself\\nacknowledged that he was “involved in the decision to\\nhandcuff [Plaintiffs].” In contrast to the absent officer who\\nwas not consulted prior to the arrest in Torres, Gutierrez was\\nthe initial officer who set these events into motion, and either\\ninstructed the other officers to arrest Plaintiffs or consulted\\nwith them in that decision. See 548 F.3d at 1206. The\\ndistrict court did not decide exactly when during the\\nprolonged detention probable cause dissipated and,\\nsimilarly, how proximate Gutierrez’s conduct was to that\\nviolation. But these issues present questions for the jury\\nbecause their resolution depends on disputed material facts.\\nUltimately, because a reasonable jury could conclude that\\nGutierrez played an integral role in the unlawfully prolonged\\n\\f NICHOLSON V. GUTIERREZ 13\\n\\ndetention and sustained handcuffing of Plaintiffs, we affirm\\nthe district court’s denial of qualified immunity on the\\nFourth Amendment violation.\\n\\n II. Fourteenth Amendment Claim\\n\\n Plaintiffs contend that the unlawful shooting violated\\ntheir substantive due process rights under the Fourteenth\\nAmendment. The district court denied Gutierrez qualified\\nimmunity because a jury could reasonably conclude that his\\nconduct amounted to deliberate indifference. The district\\ncourt also found that the substantive due process right at\\nissue was clearly established. We first address whether\\nGutierrez’s conduct violated Plaintiffs’ substantive due\\nprocess rights under the Fourteenth Amendment, and then\\nwhether the right was clearly established at the time of the\\nincident.\\n\\n A.\\n\\n To prevail on a substantive due process claim under the\\nFourteenth Amendment, Plaintiffs must show that an\\nofficer’s conduct “shocks the conscience.” See Wilkinson v.\\nTorres, 610 F.3d 546, 554 (9th Cir. 2010). The “critical\\nconsideration [is] whether the circumstances are such that\\nactual deliberation is practical.” Porter v. Osborn, 546 F.3d\\n1131, 1137 (9th Cir. 2008) (quoting Moreland v. Las Vegas\\nMetro. Police Dep’t, 159 F.3d 365, 372 (9th Cir. 1998)). If\\nso, “an officer’s ‘deliberate indifference’ may suffice to\\nshock the conscience,” Wilkinson, 610 F.3d at 554, and the\\nplaintiff may prevail by showing that the officer\\n“disregarded a known or obvious consequence of his\\naction,” Patel v. Kent Sch. Dist., 648 F.3d 965, 974 (9th Cir.\\n2011). The “deliberate-indifference inquiry should go to the\\njury if any rational factfinder could find this requisite mental\\nstate.” Id. Here, having found a triable issue on whether\\n\\f14 NICHOLSON V. GUTIERREZ\\n\\n“deliberation was practical under the circumstances,” the\\ndistrict court held that “[a] finder of fact could conclude . . .\\nthat Gutierrez disregarded the known or obvious risks of\\ninjury to J.H. and J.N.G. when he fired at Sanders without\\ntaking time to assess the situation.”\\n\\n We agree with the district court and hold that, viewing\\nthe totality of the evidence in the light most favorable to the\\nPlaintiffs, the shooting violated Plaintiffs’ due process\\nrights. We do not discount the seriousness of the situation\\nthat Officer Gutierrez thought he observed: a person holding\\nwhat appeared to be a gun standing near others who may\\nhave been in danger. But Sanders “was not engaged in any\\nthreatening or menacing behavior, and he kept the airsoft\\ngun securely pointed toward the ground.” The alleyway was\\nnear a school, and Plaintiffs were “equipped with school\\nuniforms and backpacks . . . [appearing] to be minors on\\ntheir way to school and not gang members.” Yet within\\nseconds of observing the “gun,” without consulting with his\\npartner, Gutierrez rushed down the alleyway. As he ran, he\\nfired his gun toward both Sanders, the perceived perpetrator\\nof a possible crime, and innocent bystanders, with one bullet\\nultimately striking J.N.G. in the back. Under these\\ncircumstances, a rational finder of fact could find that\\nGutierrez’s use of deadly force shocks the conscience and\\nwas unconstitutional under the Fourteenth Amendment.\\n\\n Gutierrez’s arguments to the contrary are unpersuasive.\\nAs a preliminary matter, Gutierrez challenges the district\\ncourt’s finding that a reasonable jury could find that\\ndeliberation was practical under the circumstances. But\\nbecause this is an interlocutory appeal, we are not free to\\nrevisit the district court’s conclusions as to “which facts a\\nparty may, or may not, be able to prove at trial.” See Morris,\\n736 F.3d at 834. Our review is limited to whether “the denial\\n\\f NICHOLSON V. GUTIERREZ 15\\n\\nof qualified immunity was appropriate by assuming that the\\nversion of the material facts asserted by the non-moving\\nparty is correct.” Bingue v. Prunchak, 512 F.3d 1169, 1172–\\n73 (9th Cir. 2008). Assuming the facts in Plaintiffs’ favor,\\nno attendant circumstances weighed in favor of the\\nimmediate use of deadly force, other than Gutierrez’s belief\\nthat Sanders was holding a gun. As the district court noted,\\nunder these circumstances, Gutierrez’s immediate use of\\nforce without communicating with his partner, his failure to\\nseek cover, and his failure to formulate a plan before acting\\nwere all contrary to LAPD’s training and policy. We agree\\nwith the district court that the evidence is sufficient to create\\na genuine dispute of fact on whether deliberation was\\npractical under the circumstances, and, in any event, we\\nwould be without power to reverse on this ground.\\n\\n Gutierrez next argues that, as a matter of law, the district\\ncourt erred in failing to apply the “intent to harm” standard,\\nunder which Plaintiffs must show that he acted “with a\\npurpose to harm unrelated to legitimate law enforcement\\nobjectives.” Wilkinson, 610 F.3d at 554. By that standard,\\nGutierrez argues, he is entitled to qualified immunity even if\\nhe acted with deliberate indifference, because he did not\\nintend to shoot J.N.G.\\n\\n We have previously carved out a narrow situation—\\nhigh-speed police car chases—in which we have found,\\ncategorically, that an officer does not have time to deliberate.\\nBingue, 512 F.3d at 1177. In Bingue v. Prunchak, we\\nconsidered whether high-speed pursuits should categorically\\ngive rise to the application of the “intent to harm” standard,\\nor whether the “deliberate indifference” standard may apply\\ndepending on the circumstances. In holding that “the intent\\nto harm” standard “applies to all high-speed police chases,”\\nid., we reasoned that such a rule best accounts for an officer’s\\n\\f16 NICHOLSON V. GUTIERREZ\\n\\n“repeated split-second decisions about how best to\\napprehend the fleeing suspect in a manner that will minimize\\nrisk to [the officer’s] own safety and the safety of the general\\npublic.” Id. at 1176. A suspect fleeing in a car at high speed\\ngives an officer “no time for reflection and precious little\\ntime for deliberation concerning either the decision to join\\nthe chase in the first place or the serial decisions about how\\nbest to pursue the suspect.” Id.\\n\\n Here, in contrast, under Plaintiffs’ version of the facts,\\nGutierrez may have seen what he believed to be a gun\\n(although it had the orange tip signifying a toy), but he did\\nnot see Sanders point it at anyone. Sanders was standing\\namong a group of school-aged youths with their backpacks\\nand school uniforms. Without more, we cannot say that this\\nsituation gives an officer “no time for reflection,” as in a\\nhigh-speed chase involving a fleeing suspect in a fast-\\nmoving vehicle. Cf. Estate of Lopez v. Gelhaus, 871 F.3d\\n998, 1011 (9th Cir. 2017), cert. denied, 138 S. Ct. 2680\\n(2018) (holding that mistaking a child’s toy gun for an\\nassault rifle did not as a matter of law justify an officer’s use\\nof deadly force).\\n\\n Gutierrez also relies on cases involving serious and\\nimmediate threats to public safety, but these cases too are\\neasily distinguishable. In Porter v. Osborn, the suspect\\nengaged in “evasive actions” in response to questions and\\ndirect orders from police officers, including refusing to exit\\nhis vehicle and then driving his vehicle in the direction of\\none of the officers in a perceived attempt to run over the\\nofficer. 546 F.3d at 1134–35, 1137. There, we held that the\\n“rapidly escalating nature” of the confrontation between the\\nofficer and the suspect behind the wheel left too little time\\nfor adequate deliberation and thus necessitated application\\nof the “intent to harm” standard. Id. In Moreland v. Las\\n\\f NICHOLSON V. GUTIERREZ 17\\n\\nVegas Metropolitan Police Department, the officers were\\nresponding to “the extreme emergency” of an active\\n“gunfight in progress threaten[ing] the lives of the 50 to 100\\npeople who were trapped in the parking lot.” 159 F.3d\\nat 372–73. The suspect was indisputably firing a\\nsemiautomatic handgun and refusing to comply with the\\nofficers’ orders to stop. Id. at 372. Given the ongoing crisis,\\nwe held that the officers had no opportunity to deliberate in\\nlight of the “immediate risk of serious harm or death to the\\nmany innocent individuals trapped in the parking lot.” Id.\\nThus, the “intent to harm” standard was appropriate. By\\ncontrast, assuming Plaintiffs’ contentions that Sanders was\\nnot pointing the toy gun at anyone and had not given any\\nindication that he was likely to harm anyone, there was no\\n“rapidly escalating” confrontation or “extreme emergency”\\nhere that would have deprived Gutierrez of the opportunity\\nto confer with his partner and formulate a plan to ascertain\\nwhat was happening before charging in with gunfire.\\nContrary to Gutierrez’s argument, we have not previously\\napplied the intent to harm standard to Fourteenth\\nAmendment claims involving facts similar to this case, and\\nwe decline to do so now.\\n\\n We thus agree that application of the deliberate\\nindifference standard is warranted under these\\ncircumstances. As the district court explained, in “minimal\\ninformation” situations, an officer must take some time to\\nassess what is happening before employing deadly force.\\nHolding otherwise would result in an “intolerably high risk\\nof a tragic shooting that may otherwise have been avoided\\nby proper deliberation whenever practical.” As such,\\napplying the deliberate indifference standard to Plaintiffs’\\nversion of the facts, we hold that Gutierrez’s shooting\\nviolated their substantive due process rights under the\\nFourteenth Amendment.\\n\\f18 NICHOLSON V. GUTIERREZ\\n\\n B.\\n\\n Even if a constitutional violation occurred, qualified\\nimmunity nevertheless applies unless the violation was\\nclearly established. Because no analogous case existed at\\nthe time of the shooting, we hold that the district court erred\\nin denying Gutierrez qualified immunity for this claim.\\n\\n Kisela v. Hughes is instructive. 138 S. Ct. at 1151. In\\nKisela, the Supreme Court reversed our decision denying\\nqualified immunity to an officer for an excessive force claim\\nafter he shot an individual armed with a knife as she\\napproached a bystander. Id. The Court reiterated its\\nrepeated admonition to courts “and the Ninth Circuit in\\nparticular—not to define clearly established law at a high\\nlevel of generality.” Id. at 1152 (quoting City and County of\\nSan Francisco v. Sheehan, 135 S. Ct. 1765, 1775–76\\n(2015)). The Court explained that, where “the result\\ndepends very much on the facts of each case . . . officers are\\nentitled to qualified immunity unless existing precedent\\n‘squarely governs’ the specific facts at issue.” Id. at 1153\\n(quoting Mullenix v. Luna, 136 S. Ct. 305, 309 (2015) (per\\ncuriam)).\\n\\n Here, Plaintiffs failed to identify any authority that\\nrendered the contours of the substantive due process right at\\nissue “sufficiently definite that any reasonable official in the\\ndefendant’s shoes would have understood he was violating\\nit.” See Kisela, 138 S. Ct. at 1153. In their briefing,\\nPlaintiffs cited cases establishing broadly that “the\\nConstitution protects a citizen’s liberty interest in her own\\nbodily security,” which define the right at much too high a\\nlevel of generality to clearly establish a rule of conduct. See,\\ne.g., Ingraham v. Wright, 430 U.S. 651, 673–74 (1977).\\nThey also discuss our “state-created danger exception”\\ncases, but these involve failures to act that lead to injuries\\n\\f NICHOLSON V. GUTIERREZ 19\\n\\nfrom third parties, rather than affirmative actions by officers\\nthat directly cause injury to the plaintiff. See, e.g., Kennedy\\nv. City of Ridgefield, 439 F.3d 1055 (9th Cir. 2006) (failure\\nto warn allegedly caused shooting by neighbor); Patel,\\n648 F.3d 965 (failure to adequately supervise a disabled high\\nschool student allegedly caused sexual abuse by another\\nstudent); Wood v. Ostrander, 879 F.2d 583 (9th Cir. 1989)\\n(failure to leave plaintiff in safe location allegedly caused\\nrape in a high-crime area). These cases are too factually\\ndissimilar to clearly establish a constitutional violation by an\\nofficer’s accidental shooting of a bystander.\\n\\n At oral argument, Plaintiffs conceded that it was “it was\\ndifficult to find a case that was squarely on point,” where a\\ncourt found a constitutional violation in the context of a\\nbystander shooting. Instead, the gravamen of Plaintiffs’\\nanalysis is that the use of deadly force against Sanders was\\nlikely unreasonable, relying principally on our cases\\nanalyzing Fourth Amendment claims of excessive force.\\nE.g., Hughes v. Kisela, 862 F.3d 775, 789 (9th Cir. 2016),\\nrev’d sub nom. Kisela v. Hughes, 138 S. Ct. 1148 (2018);\\nEmmons v. City of Escondido, No. 16-55771, 2018 WL\\n1531064 (9th Cir. Mar. 29, 2018), rev’d sub nom. City of\\nEscondido v. Emmons, 139 S. Ct. 500 (2019). 4 While these\\ncases may help to identify whether the use of force against\\nSanders amounted to a Fourth Amendment violation, they\\ndo not clearly establish that a shooting in these\\n\\n 4\\n Similarly, the district court also relied principally on Fourth\\nAmendment cases in denying qualified immunity on this claim.\\nHowever, as with Plaintiffs’ citations, some of these cases were decided\\nafter the incident, e.g., Gelhaus, 871 F.3d 998, and one has since been\\nreversed by the Supreme Court on the “clearly established” question.\\nSee Kisela, 138 S. Ct. 1148. These problems present additional reasons\\nwhy we find that the law was not clearly established on Plaintiffs’\\nFourteenth Amendment claim.\\n\\f20 NICHOLSON V. GUTIERREZ\\n\\ncircumstances constitutes deliberate indifference to\\nPlaintiffs. Sanders is not a plaintiff in this lawsuit, and\\nPlaintiffs would not have standing to raise a Fourth\\nAmendment claim on his behalf. See Plumhoff, 572 U.S. at\\n778 (“Our cases make it clear that ‘Fourth Amendment\\nrights are personal rights which . . . may not be vicariously\\nasserted.’” (quoting Alderman v. United States, 394 U.S.\\n165, 174 (1969)). The Fourth Amendment cases therefore\\ndo not clearly establish the contours of the Fourteenth\\nAmendment substantive due process rights at hand. 5\\n\\n Because no binding circuit or Supreme Court precedent\\nhas established a substantive due process violation under\\ncomparable circumstances, the Fourteenth Amendment right\\nat issue lacked “contours . . . sufficiently definite” to place\\nthe issue “beyond debate.” See Kisela, 138 S. Ct. at 1152–\\n53 (quoting Plumhoff, 572 U.S. at 779). We accordingly\\nreverse the district court and remand for an entry of qualified\\nimmunity on this claim.\\n\\n\\n\\n 5\\n Certainly, considerations of reasonableness germane to a Fourth\\nAmendment analysis are relevant to the substantive due process inquiry.\\nHistorically, our cases have recognized some overlap in these two\\nconstitutional protections. See, e.g., P.B. v. Koch, 96 F.3d 1298, 1303\\nn.4 (9th Cir. 1996) (holding that, “[r]egardless of the appropriate ‘home’\\nfor plaintiffs’ right to be free from excessive force, there was a clearly\\nestablished right to be free such force” because “[u]nder any standard,\\n[the defendant’s] alleged actions were clearly unlawful.”). But we have\\nheld that a Fourteenth Amendment claim of excessive force “must be\\ngoverned by a different standard than” a Fourth Amendment claim of\\nexcessive force. Byrd v. Guess, 137 F.3d 1126, 1133–34 (9th Cir. 1998),\\nsuperseded by statute on other grounds as recognized in Little v. City of\\nManhattan Beach, 21 Fed. App’x 651 (9th Cir. 2001). Thus, our Fourth\\nAmendment cases cannot clearly establish the contours of the Fourteenth\\nAmendment right, despite similarities between the standards.\\n\\f NICHOLSON V. GUTIERREZ 21\\n\\n III.\\n\\n CONCLUSION\\n\\n We reverse the district court’s denial of qualified\\nimmunity to Gutierrez as to the Fourteenth Amendment\\nclaim and affirm the court’s ruling as to the Fourth\\nAmendment claim. We remand for proceedings consistent\\nwith this opinion.\\n\\n REVERSED in part, AFFIRMED in part, and\\nREMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4431732/\", \"author_raw\": \"NGUYEN, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"ANDREW J KLEINFELD"},"judge2":{"kind":"string","value":"R GUY COLE JR"},"judge3":{"kind":"string","value":"JACQUELINE H NGUYEN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4654479/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Geraldine Nicholson v. Miguel Gutierrez\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-21\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nGERALDINE NICHOLSON; JOSE No. 17-56648\nFERNANDO HUERTA; J. H., a minor\nby and through his guardian ad litem, D.C. No.\nJose Fernando Huerta; J. N. G., a 2:15-cv-07594-\nminor by and through his guardian DDP-RAO\nad litem, Geraldine Nicholson,\n Plaintiffs-Appellees,\n OPINION\n v.\n\nCITY OF LOS ANGELES; EVERARDO\nAMARAL, individually and in his\nofficial capacity as a Police Officer\nof the Los Angeles Police\nDepartment,\n Defendants,\n\n and\n\nMIGUEL GUTIERREZ, individually\nand in his official capacity as a\nPolice Officer of the Los Angeles\nPolice Department,\n Defendant-Appellant.\n\n Appeal from the United States District Court\n for the Central District of California\n Dean D. Pregerson, District Judge, Presiding\n\f2 NICHOLSON V. GUTIERREZ\n\n Argued and Submitted March 4, 2019\n Pasadena, California\n\n Filed August 21, 2019\n\n Before: Andrew J. Kleinfeld, R. Guy Cole, Jr., *\n and Jacqueline H. Nguyen, Circuit Judges.\n\n Opinion by Judge Nguyen\n\n\n SUMMARY **\n\n\n Civil Rights\n\n The panel reversed in part and affirmed in part the\ndistrict court’s denial of qualified immunity to a Los Angeles\nPolice Department officer in an action brought pursuant to\n42 U.S.C. § 1983 for violations of plaintiffs’ Fourth\nAmendment rights to be free from excessive force and\nunreasonable seizure and violations of their Fourteenth\nAmendment substantive due process rights.\n\n Plaintiffs were among a group of teenagers who had met\nin an alleyway near their school to listen to and sing rap\nmusic. One of the teenagers, plaintiff J.N.G., was shot by\ndefendant Gutierrez after Gutierrez mistook a plastic Airsoft\nreplica gun held by one of the other teenagers for an actual\n\n *\n The Honorable R. Guy Cole, Jr., United States Circuit Judge for\nthe U.S. Court of Appeals for the Sixth Circuit, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f NICHOLSON V. GUTIERREZ 3\n\ngun. After the shooting, officers detained the group for over\nfive hours while they investigated. J.N.G. and J.H. filed a\nlawsuit and the district court denied qualified immunity on\nplaintiffs’ Fourth and Fourteenth Amendment claims.\n\n Addressing the Fourth Amendment claim, the panel\nagreed with the district court that under the circumstances,\nplaintiffs’ continued detention for five hours after the\nshooting—well after any probable cause would have\ndissipated—and the use of handcuffs throughout the\nduration of the detention violated plaintiffs’ clearly\nestablished Fourth Amendment rights to be free from\nunlawful arrest and excessive force. The panel rejected\nGutierrez’s argument that while he participated in the initial\nhandcuffing and detention, he was not responsible for any\nsubsequent constitutional violation because he played no\nrole in that conduct. The panel held that an officer can be\nheld liable where he is just one participant in a sequence of\nevents that gives rise to a constitutional violation. Here,\nviewing the evidence in the light most favorable to plaintiffs,\nGutierrez was more than a “mere bystander” in the alleged\nconstitutional violations. The panel affirmed the district\ncourt’s denial of qualified immunity on the Fourth\nAmendment violations because, ultimately, a reasonable\njury could conclude that Gutierrez played an integral role in\nthe unlawfully prolonged detention and sustained\nhandcuffing of plaintiffs.\n\n Addressing the Fourteenth Amendment substantive due\nprocess claim, the panel held that, viewing the totality of the\nevidence in the light most favorable to the plaintiffs, the\nshooting violated plaintiffs’ due process rights. Under the\ncircumstances, a rational finder of fact could find that\nGutierrez’s use of deadly force shocked the conscience and\nwas unconstitutional under the Fourteenth Amendment.\n\f4 NICHOLSON V. GUTIERREZ\n\nNevertheless, the panel held that because no analogous case\nexisted at the time of the shooting, the district court erred by\ndenying Gutierrez qualified immunity for this claim. The\npanel accordingly reversed the district court and remanded\nfor an entry of qualified immunity on the Fourteenth\nAmendment claim.\n\n\n COUNSEL\n\nDenise L. Rocawich (argued) and James R. Touchstone,\nJones & Mayer, Fullerton, California, for Defendant-\nAppellant.\n\nHerbert-John S. Hayden (argued) and John W. Harris, Harris\n& Associates, Los Angeles, California, for Plaintiffs-\nAppellees.\n\n\n OPINION\n\nNGUYEN, Circuit Judge:\n\n On the morning of February 10, 2015, four teenagers met\nin an alleyway near their school to listen to and sing rap\nmusic. As the teenagers—Michael Sanders, Abdul Wooten,\nJ.N.G., and J.H. 1—stood in a tight circle dancing and\nrapping, Sanders was holding a plastic Airsoft replica gun\nwith a bright orange tip as a prop. Just as they turned off the\nmusic and were getting ready to head to school, J.N.G. was\n\n 1\n Throughout the record, J.N.G. (Jamar Nicholson Green) and J.H.\n(Jason Huerta) were referred to by their initials because they were minors\nat the time of the incident and for some time after filing this lawsuit. For\nconsistency, we also refer to them by their initials.\n\f NICHOLSON V. GUTIERREZ 5\n\nshot by Officer Michael Gutierrez of the Los Angeles Police\nDepartment (“LAPD”). Officer Gutierrez fired his weapon\nbecause he mistook Sanders’s replica gun for an actual gun.\nGutierrez fired multiple shots, one of which hit J.N.G. in the\nback. After the shooting, officers detained the group for over\nfive hours while they investigated.\n\n J.N.G. and J.H. (collectively, “Plaintiffs”) filed a lawsuit\nagainst the officers, the LAPD, and the City of Los Angeles,\nalleging violations of the Fourth and Fourteenth\nAmendments and various state laws. The district court\ndenied qualified immunity on two of Plaintiffs’\nconstitutional claims. Gutierrez appeals. We affirm in part\nand reverse in part.\n\n BACKGROUND 2\n\n At around 7:15 a.m. on February 10, 2015, J.N.G., J.H.,\nMichael Sanders, and Abdul Wooten met in an alley at the\ncorner of 10th Avenue and Florence Avenue in Los Angeles,\nCA, a few blocks from their high school. They regularly\ngathered in that alleyway before and after school to listen to\nmusic and freestyle rap. That morning, as they were rapping\nand dancing in a circle, Sanders was holding a plastic toy\ngun with a bright orange tip. J.N.G., J.H., and Wooten\nmaintain that Sanders kept the gun pointed downward\naround waist-level and did not fire the gun that morning. At\napproximately 7:40 a.m., the teenagers turned off the music\nand began preparing to head to school.\n\n\n\n\n 2\n At this stage of the proceedings, we view the facts in the light most\nfavorable to Plaintiffs. George v. Morris, 736 F.3d 829, 836 (9th Cir.\n2013) (9th Cir. 2013).\n\f6 NICHOLSON V. GUTIERREZ\n\n Around this time, Officer Everardo Amaral was driving\ndown 10th Avenue in an unmarked car with his partner,\nOfficer Gutierrez. From the passenger seat, Gutierrez “saw\na person (later identified as Michael Sanders) pointing . . . a\nblue steel handgun at another person (later identified as\nPlaintiff J.H.).” Gutierrez, believing that J.H. was “being\nrobbed at gun point or was about to be murdered,” yelled\n“Gun, gun, gun!” Amaral stopped the vehicle south of the\nalley on 10th Avenue. Without conferring with Amaral,\nGutierrez immediately jumped out of the car and ran into the\nalley. Amaral parked the car and followed Gutierrez.\nNeither officer was in uniform.\n\n Gutierrez claims he identified himself as an LAPD\nofficer and commanded Sanders to drop the gun. However,\nJ.H., J.N.G., and Wooten all contend that Gutierrez did not\nidentify himself or make any verbal commands prior to\nshooting his weapon. A few seconds after he entered the\nalley, Officer Gutierrez fired at least three shots, one of\nwhich hit J.N.G. in the back. J.N.G. and J.H. contend that\nGutierrez fired his gun with one hand while running toward\nthem, while Gutierrez stated that he fired only after stopping\na few feet away from the group. When the shots were fired,\nJ.H. was about to put on his school uniform, and J.N.G. was\nspraying cologne on his face. The four of them had been\nstanding in a tight circle, “within a foot or so of each other.”\nSanders soon turned and dropped the toy gun, though the\nparties dispute whether this occurred before or after\nGutierrez fired.\n\n Shortly after Officer Gutierrez fired, Officer Amaral\narrived and requested three additional units. Amaral also\nrequested an ambulance when he realized that J.N.G. had\nbeen shot. The officers held the group at gunpoint, face\ndown on the ground. The parties dispute how far the\n\f NICHOLSON V. GUTIERREZ 7\n\ndropped “gun” was from the teenagers when they were on\nthe ground, but neither officer picked it up or moved it away\nfrom them. While on the ground, J.H. shouted that the gun\nwas “not even a real gun” and repeatedly asked why the\nofficers shot at them and “What did we do wrong?” The\nofficers remained silent in response to his questions, with\ndumbfounded expressions on their faces. Responding\nofficers soon arrived, and they searched and handcuffed the\ngroup. Gutierrez was “involved in the decision to handcuff\nthem.” Officer Amaral later explained in his deposition that\nthe detention of the boys was “[f]or [a] weapons violation.”\nOfficers Gutierrez and Amaral were separated and\nmonitored soon after additional units arrived on the scene.\nJ.H. remained in handcuffs throughout the investigation,\nwhich lasted until around 1:00 p.m., over five hours after the\nshooting. J.N.G. also remained in handcuffs for over five\nhours—through the duration of his hospital examination—\nuntil detectives interrogated him.\n\n J.N.G., by and through his mother and guardian ad litem\nGeraldine Nicholson, and J.H., by and through his father and\nguardian ad litem Jose Fernando Huerta, sued Officer\nGutierrez, Officer Amaral, the City of Los Angeles, the\nLAPD, Chief of Police Charles Beck, and Commander\nAndrew Smith. In addition to various state law claims,\nPlaintiffs J.N.G. and J.H. alleged claims under 42 U.S.C.\n§ 1983 for violations of their Fourth Amendment rights to be\nfree from excessive force and unreasonable seizure and\nviolations of their Fourteenth Amendment substantive due\nprocess rights. Defendants Gutierrez, Amaral, and the City\nof Los Angeles jointly moved for summary judgment,\narguing in part that no constitutional violation occurred and\nthat qualified immunity applied.\n\f8 NICHOLSON V. GUTIERREZ\n\n The district court granted in part and denied in part\nDefendants’ joint motion for summary judgment. Relevant\nto this appeal, the district court denied Gutierrez qualified\nimmunity in part on Plaintiffs’ Fourth Amendment claim and\non Plaintiffs’ Fourteenth Amendment claim. Gutierrez\ntimely appealed.\n\n JURISDICTION AND STANDARD OF REVIEW\n\n We have jurisdiction under 28 U.S.C. § 1291 to review\nthe denial of qualified immunity at the summary judgment\nstage. Plumhoff v. Rickard, 572 U.S. 765, 771–73 (2014).\nBut “the scope of our review over the appeal is\ncircumscribed” because we may not “consider questions of\neviden[tiary] sufficiency, i.e., which facts a party may, or\nmay not, be able to prove at trial.” Morris, 736 F.3d at 834\n(quoting CarePartners, LLC v. Lashway, 545 F.3d 867, 875\n(9th Cir. 2008)). “Thus, in this appeal, we are confined to\nthe question of ‘whether the defendant[s] would be entitled\nto qualified immunity as a matter of law, assuming all factual\ndisputes are resolved, and all reasonable inferences are\ndrawn, in plaintiff’s favor.’” Id. at 836 (quoting Karl v. City\nof Mountlake Terrace, 678 F.3d 1062, 1068 (9th Cir. 2012)).\n\n DISCUSSION\n\n “Qualified immunity attaches when an official’s conduct\ndoes not violate clearly established statutory or\nconstitutional rights of which a reasonable person would\nhave known.” Kisela v. Hughes, 138 S. Ct. 1148, 1152\n(2018) (per curiam). Once a defendant has raised qualified\nimmunity as a defense to a claim, a plaintiff must show\n“(1) that the right was violated; and (2) that the right was\nclearly established at the time of the alleged misconduct.”\nIsayeva v. Sacramento Sheriff’s Dep’t, 872 F.3d 938, 946\n(9th Cir. 2017). Although a right is not clearly established\n\f NICHOLSON V. GUTIERREZ 9\n\nwhere merely defined “at a high level of generality,”\nqualified immunity does not “require a case directly on\npoint.” Kisela, 138 S. Ct. at 1152. Instead, the “focus is on\nwhether the officer had fair notice that her conduct was\nunlawful,” id., for example, through “any cases of\ncontrolling authority in their jurisdiction at the time of the\nincident,” Wilson v. Layne, 526 U.S. 603, 617 (1999).\n\n We now turn to each of Plaintiffs’ constitutional claims.\n\n I. Fourth Amendment Claim\n\n Plaintiffs allege that, after the shooting, Defendants\nviolated their Fourth Amendment rights by unlawfully\narresting them, using excessive force, and prolonging their\ndetention. 3 In denying Officer Gutierrez qualified\nimmunity, the district court concluded that immediately after\nthe shooting, an investigatory stop was reasonable while the\nofficers assessed the situation. But “[a]t some point, the\ndetention evolved into a full-fledged arrest that required\nprobable cause that J.H. and J.N.G. had been engaged in\ncriminal activity.” The district court further concluded that\n“a reasonable jury could determine that the sustained\nhandcuffing of J.H. and J.N.G. . . . constituted excessive\nforce.”\n\n\n 3\n Plaintiffs also alleged that the shooting itself also violated their\nFourth Amendment rights. The district court granted summary judgment\nin favor of Defendants on this claim, finding that the teenagers were not\n“seized” by Gutierrez’s gunfire under the Fourth Amendment because\nGutierrez intended to use deadly force against Sanders, not against\nJ.N.G. or J.H. See United States v. Al Nasser, 555 F.3d 722, 728 (9th\nCir. 2009). Plaintiffs do not challenge this ruling and thus the question\nof whether the shooting violated their Fourth Amendment rights is not\nbefore us in this appeal.\n\f10 NICHOLSON V. GUTIERREZ\n\n It is well-established that a “person may not be arrested,\nor must be released from arrest, if previously established\nprobable cause has dissipated.” United States v. Ortiz-\nHernandez, 427 F.3d 567, 574 (9th Cir. 2005) (per curiam).\n“As a corollary . . . of the rule that the police may rely on the\ntotality of facts available to them in establishing probable\ncause, they also may not disregard facts tending to dissipate\nprobable cause.” Id. (quoting Bigford v. Taylor, 834 F.2d\n1213, 1218 (5th Cir. 1988)). A reasonable officer would\nknow that participation in an ongoing seizure after any\nprobable cause had dissipated violates the Fourth\nAmendment.\n\n Here, it was soon apparent to the officers that the\nteenagers were unarmed, posed no threat to anyone, and\nwere not engaged in any criminal activity. The incident\noccurred in the morning right before the start of school\nhours, and Plaintiffs had their school uniforms and\nbackpacks. In fact, as Officer Gutierrez approached the\nscene, J.N.G. was spraying on cologne and J.H. was donning\nhis school uniform. Moreover, Officer Gutierrez admitted\nthat he perceived at least J.H. to be a possible victim, not a\nsuspect, further undermining any justification to detain him.\nWe agree with the district court that under these\ncircumstances, Plaintiffs’ continued detention for five\nhours—well after any probable cause would have\ndissipated—and the use of handcuffs throughout the\nduration of the detention violated Plaintiffs’ clearly\nestablished Fourth Amendment rights to be free from\nunlawful arrest and excessive force.\n\n On appeal, Officer Gutierrez does not appear to dispute\nthat the prolonged detention and handcuffing violated\nPlaintiffs’ clearly established Fourth Amendment rights.\nInstead, Gutierrez focuses his argument on the fact that he\n\f NICHOLSON V. GUTIERREZ 11\n\nwas separated and monitored after the shooting. Thus, his\nargument goes, while he admittedly “was an integral\nparticipant in the initial handcuffing and detention,” he is not\nresponsible for any subsequent constitutional violation\nbecause he played no role in that conduct.\n\n A police officer need not have been the sole party\nresponsible for a constitutional violation before liability may\nattach. “An officer’s liability under section 1983 is\npredicated on his ‘integral participation’ in the alleged\nviolation.” Blankenhorn v. City of Orange, 485 F.3d 463,\n481 n.12 (9th Cir. 2007) (quoting Chuman v. Wright, 76 F.3d\n292, 294–95 (9th Cir. 1996)). This theory of liability “does\nnot require that each officer’s actions themselves rise to the\nlevel of a constitutional violation.” Id. (quoting Boyd v.\nBenton County, 374 F.3d 773, 780 (9th Cir. 2004)). Instead,\nliability may attach if the officer has “some fundamental\ninvolvement in the conduct that allegedly caused the\nviolation.” Id. A theory of integral participation thus\ncomports with general tort principles of causation applicable\nto a § 1983 action: “[G]overnment officials, like other\ndefendants, are generally responsible for the ‘natural’ or\n‘reasonably foreseeable’ consequences of their actions.”\nStoot v. City of Everett, 582 F.3d 910, 926 (9th Cir. 2009).\nEven though “an intervening decision of an informed,\nneutral decision-maker ‘breaks’ the chain of causation,” the\nchain of causation is not broken where the intervening\ndecision was foreseeably influenced by the defendant. Id.\n(quoting Murray v. Earle, 405 F.3d 278, 292 (5th Cir.\n2005)). Thus, under our case law, an officer could be held\nliable where he is just one participant in a sequence of events\nthat gives rise to a constitutional violation.\n\n For example, in Boyd v. Benton County, we held that\neach of the officers in a search operation were liable for\n\f12 NICHOLSON V. GUTIERREZ\n\nexcessive force for the use of a flash-bang grenade, even\nthough only one officer reached into the home and deployed\nthe device. 374 F.3d at 780. We reasoned that the use of the\ndevice was “part of the search operation in which every\nofficer participated in some meaningful way.” Id.\nMoreover, each officer “was aware of the decision . . . , did\nnot object to it, and participated in the search operation\nknowing the flash-bang was to be deployed.” Id. By\ncontrast, in Torres v. City of Los Angeles, we found that a\ndetective was not an integral participant in an allegedly\nunlawful arrest, in part because she “was not present [at the\narrest], and there is no evidence that [she] instructed the\nother detectives to arrest [the plaintiff] or that any of those\ndetectives consulted with her before making the arrest.”\n548 F.3d 1197, 1206 (9th Cir. 2008).\n\n Here, viewing the evidence in the light most favorable to\nPlaintiffs, Gutierrez was more than a “mere bystander” in the\nalleged constitutional violations. See Chuman, 76 F.3d at\n294. Plaintiffs alleged that Officers Amaral and Gutierrez\n“directed the other officers to handcuff, search and arrest all\nof us for reasons unknown to any of us.” Gutierrez himself\nacknowledged that he was “involved in the decision to\nhandcuff [Plaintiffs].” In contrast to the absent officer who\nwas not consulted prior to the arrest in Torres, Gutierrez was\nthe initial officer who set these events into motion, and either\ninstructed the other officers to arrest Plaintiffs or consulted\nwith them in that decision. See 548 F.3d at 1206. The\ndistrict court did not decide exactly when during the\nprolonged detention probable cause dissipated and,\nsimilarly, how proximate Gutierrez’s conduct was to that\nviolation. But these issues present questions for the jury\nbecause their resolution depends on disputed material facts.\nUltimately, because a reasonable jury could conclude that\nGutierrez played an integral role in the unlawfully prolonged\n\f NICHOLSON V. GUTIERREZ 13\n\ndetention and sustained handcuffing of Plaintiffs, we affirm\nthe district court’s denial of qualified immunity on the\nFourth Amendment violation.\n\n II. Fourteenth Amendment Claim\n\n Plaintiffs contend that the unlawful shooting violated\ntheir substantive due process rights under the Fourteenth\nAmendment. The district court denied Gutierrez qualified\nimmunity because a jury could reasonably conclude that his\nconduct amounted to deliberate indifference. The district\ncourt also found that the substantive due process right at\nissue was clearly established. We first address whether\nGutierrez’s conduct violated Plaintiffs’ substantive due\nprocess rights under the Fourteenth Amendment, and then\nwhether the right was clearly established at the time of the\nincident.\n\n A.\n\n To prevail on a substantive due process claim under the\nFourteenth Amendment, Plaintiffs must show that an\nofficer’s conduct “shocks the conscience.” See Wilkinson v.\nTorres, 610 F.3d 546, 554 (9th Cir. 2010). The “critical\nconsideration [is] whether the circumstances are such that\nactual deliberation is practical.” Porter v. Osborn, 546 F.3d\n1131, 1137 (9th Cir. 2008) (quoting Moreland v. Las Vegas\nMetro. Police Dep’t, 159 F.3d 365, 372 (9th Cir. 1998)). If\nso, “an officer’s ‘deliberate indifference’ may suffice to\nshock the conscience,” Wilkinson, 610 F.3d at 554, and the\nplaintiff may prevail by showing that the officer\n“disregarded a known or obvious consequence of his\naction,” Patel v. Kent Sch. Dist., 648 F.3d 965, 974 (9th Cir.\n2011). The “deliberate-indifference inquiry should go to the\njury if any rational factfinder could find this requisite mental\nstate.” Id. Here, having found a triable issue on whether\n\f14 NICHOLSON V. GUTIERREZ\n\n“deliberation was practical under the circumstances,” the\ndistrict court held that “[a] finder of fact could conclude . . .\nthat Gutierrez disregarded the known or obvious risks of\ninjury to J.H. and J.N.G. when he fired at Sanders without\ntaking time to assess the situation.”\n\n We agree with the district court and hold that, viewing\nthe totality of the evidence in the light most favorable to the\nPlaintiffs, the shooting violated Plaintiffs’ due process\nrights. We do not discount the seriousness of the situation\nthat Officer Gutierrez thought he observed: a person holding\nwhat appeared to be a gun standing near others who may\nhave been in danger. But Sanders “was not engaged in any\nthreatening or menacing behavior, and he kept the airsoft\ngun securely pointed toward the ground.” The alleyway was\nnear a school, and Plaintiffs were “equipped with school\nuniforms and backpacks . . . [appearing] to be minors on\ntheir way to school and not gang members.” Yet within\nseconds of observing the “gun,” without consulting with his\npartner, Gutierrez rushed down the alleyway. As he ran, he\nfired his gun toward both Sanders, the perceived perpetrator\nof a possible crime, and innocent bystanders, with one bullet\nultimately striking J.N.G. in the back. Under these\ncircumstances, a rational finder of fact could find that\nGutierrez’s use of deadly force shocks the conscience and\nwas unconstitutional under the Fourteenth Amendment.\n\n Gutierrez’s arguments to the contrary are unpersuasive.\nAs a preliminary matter, Gutierrez challenges the district\ncourt’s finding that a reasonable jury could find that\ndeliberation was practical under the circumstances. But\nbecause this is an interlocutory appeal, we are not free to\nrevisit the district court’s conclusions as to “which facts a\nparty may, or may not, be able to prove at trial.” See Morris,\n736 F.3d at 834. Our review is limited to whether “the denial\n\f NICHOLSON V. GUTIERREZ 15\n\nof qualified immunity was appropriate by assuming that the\nversion of the material facts asserted by the non-moving\nparty is correct.” Bingue v. Prunchak, 512 F.3d 1169, 1172–\n73 (9th Cir. 2008). Assuming the facts in Plaintiffs’ favor,\nno attendant circumstances weighed in favor of the\nimmediate use of deadly force, other than Gutierrez’s belief\nthat Sanders was holding a gun. As the district court noted,\nunder these circumstances, Gutierrez’s immediate use of\nforce without communicating with his partner, his failure to\nseek cover, and his failure to formulate a plan before acting\nwere all contrary to LAPD’s training and policy. We agree\nwith the district court that the evidence is sufficient to create\na genuine dispute of fact on whether deliberation was\npractical under the circumstances, and, in any event, we\nwould be without power to reverse on this ground.\n\n Gutierrez next argues that, as a matter of law, the district\ncourt erred in failing to apply the “intent to harm” standard,\nunder which Plaintiffs must show that he acted “with a\npurpose to harm unrelated to legitimate law enforcement\nobjectives.” Wilkinson, 610 F.3d at 554. By that standard,\nGutierrez argues, he is entitled to qualified immunity even if\nhe acted with deliberate indifference, because he did not\nintend to shoot J.N.G.\n\n We have previously carved out a narrow situation—\nhigh-speed police car chases—in which we have found,\ncategorically, that an officer does not have time to deliberate.\nBingue, 512 F.3d at 1177. In Bingue v. Prunchak, we\nconsidered whether high-speed pursuits should categorically\ngive rise to the application of the “intent to harm” standard,\nor whether the “deliberate indifference” standard may apply\ndepending on the circumstances. In holding that “the intent\nto harm” standard “applies to all high-speed police chases,”\nid., we reasoned that such a rule best accounts for an officer’s\n\f16 NICHOLSON V. GUTIERREZ\n\n“repeated split-second decisions about how best to\napprehend the fleeing suspect in a manner that will minimize\nrisk to [the officer’s] own safety and the safety of the general\npublic.” Id. at 1176. A suspect fleeing in a car at high speed\ngives an officer “no time for reflection and precious little\ntime for deliberation concerning either the decision to join\nthe chase in the first place or the serial decisions about how\nbest to pursue the suspect.” Id.\n\n Here, in contrast, under Plaintiffs’ version of the facts,\nGutierrez may have seen what he believed to be a gun\n(although it had the orange tip signifying a toy), but he did\nnot see Sanders point it at anyone. Sanders was standing\namong a group of school-aged youths with their backpacks\nand school uniforms. Without more, we cannot say that this\nsituation gives an officer “no time for reflection,” as in a\nhigh-speed chase involving a fleeing suspect in a fast-\nmoving vehicle. Cf. Estate of Lopez v. Gelhaus, 871 F.3d\n998, 1011 (9th Cir. 2017), cert. denied, 138 S. Ct. 2680\n(2018) (holding that mistaking a child’s toy gun for an\nassault rifle did not as a matter of law justify an officer’s use\nof deadly force).\n\n Gutierrez also relies on cases involving serious and\nimmediate threats to public safety, but these cases too are\neasily distinguishable. In Porter v. Osborn, the suspect\nengaged in “evasive actions” in response to questions and\ndirect orders from police officers, including refusing to exit\nhis vehicle and then driving his vehicle in the direction of\none of the officers in a perceived attempt to run over the\nofficer. 546 F.3d at 1134–35, 1137. There, we held that the\n“rapidly escalating nature” of the confrontation between the\nofficer and the suspect behind the wheel left too little time\nfor adequate deliberation and thus necessitated application\nof the “intent to harm” standard. Id. In Moreland v. Las\n\f NICHOLSON V. GUTIERREZ 17\n\nVegas Metropolitan Police Department, the officers were\nresponding to “the extreme emergency” of an active\n“gunfight in progress threaten[ing] the lives of the 50 to 100\npeople who were trapped in the parking lot.” 159 F.3d\nat 372–73. The suspect was indisputably firing a\nsemiautomatic handgun and refusing to comply with the\nofficers’ orders to stop. Id. at 372. Given the ongoing crisis,\nwe held that the officers had no opportunity to deliberate in\nlight of the “immediate risk of serious harm or death to the\nmany innocent individuals trapped in the parking lot.” Id.\nThus, the “intent to harm” standard was appropriate. By\ncontrast, assuming Plaintiffs’ contentions that Sanders was\nnot pointing the toy gun at anyone and had not given any\nindication that he was likely to harm anyone, there was no\n“rapidly escalating” confrontation or “extreme emergency”\nhere that would have deprived Gutierrez of the opportunity\nto confer with his partner and formulate a plan to ascertain\nwhat was happening before charging in with gunfire.\nContrary to Gutierrez’s argument, we have not previously\napplied the intent to harm standard to Fourteenth\nAmendment claims involving facts similar to this case, and\nwe decline to do so now.\n\n We thus agree that application of the deliberate\nindifference standard is warranted under these\ncircumstances. As the district court explained, in “minimal\ninformation” situations, an officer must take some time to\nassess what is happening before employing deadly force.\nHolding otherwise would result in an “intolerably high risk\nof a tragic shooting that may otherwise have been avoided\nby proper deliberation whenever practical.” As such,\napplying the deliberate indifference standard to Plaintiffs’\nversion of the facts, we hold that Gutierrez’s shooting\nviolated their substantive due process rights under the\nFourteenth Amendment.\n\f18 NICHOLSON V. GUTIERREZ\n\n B.\n\n Even if a constitutional violation occurred, qualified\nimmunity nevertheless applies unless the violation was\nclearly established. Because no analogous case existed at\nthe time of the shooting, we hold that the district court erred\nin denying Gutierrez qualified immunity for this claim.\n\n Kisela v. Hughes is instructive. 138 S. Ct. at 1151. In\nKisela, the Supreme Court reversed our decision denying\nqualified immunity to an officer for an excessive force claim\nafter he shot an individual armed with a knife as she\napproached a bystander. Id. The Court reiterated its\nrepeated admonition to courts “and the Ninth Circuit in\nparticular—not to define clearly established law at a high\nlevel of generality.” Id. at 1152 (quoting City and County of\nSan Francisco v. Sheehan, 135 S. Ct. 1765, 1775–76\n(2015)). The Court explained that, where “the result\ndepends very much on the facts of each case . . . officers are\nentitled to qualified immunity unless existing precedent\n‘squarely governs’ the specific facts at issue.” Id. at 1153\n(quoting Mullenix v. Luna, 136 S. Ct. 305, 309 (2015) (per\ncuriam)).\n\n Here, Plaintiffs failed to identify any authority that\nrendered the contours of the substantive due process right at\nissue “sufficiently definite that any reasonable official in the\ndefendant’s shoes would have understood he was violating\nit.” See Kisela, 138 S. Ct. at 1153. In their briefing,\nPlaintiffs cited cases establishing broadly that “the\nConstitution protects a citizen’s liberty interest in her own\nbodily security,” which define the right at much too high a\nlevel of generality to clearly establish a rule of conduct. See,\ne.g., Ingraham v. Wright, 430 U.S. 651, 673–74 (1977).\nThey also discuss our “state-created danger exception”\ncases, but these involve failures to act that lead to injuries\n\f NICHOLSON V. GUTIERREZ 19\n\nfrom third parties, rather than affirmative actions by officers\nthat directly cause injury to the plaintiff. See, e.g., Kennedy\nv. City of Ridgefield, 439 F.3d 1055 (9th Cir. 2006) (failure\nto warn allegedly caused shooting by neighbor); Patel,\n648 F.3d 965 (failure to adequately supervise a disabled high\nschool student allegedly caused sexual abuse by another\nstudent); Wood v. Ostrander, 879 F.2d 583 (9th Cir. 1989)\n(failure to leave plaintiff in safe location allegedly caused\nrape in a high-crime area). These cases are too factually\ndissimilar to clearly establish a constitutional violation by an\nofficer’s accidental shooting of a bystander.\n\n At oral argument, Plaintiffs conceded that it was “it was\ndifficult to find a case that was squarely on point,” where a\ncourt found a constitutional violation in the context of a\nbystander shooting. Instead, the gravamen of Plaintiffs’\nanalysis is that the use of deadly force against Sanders was\nlikely unreasonable, relying principally on our cases\nanalyzing Fourth Amendment claims of excessive force.\nE.g., Hughes v. Kisela, 862 F.3d 775, 789 (9th Cir. 2016),\nrev’d sub nom. Kisela v. Hughes, 138 S. Ct. 1148 (2018);\nEmmons v. City of Escondido, No. 16-55771, 2018 WL\n1531064 (9th Cir. Mar. 29, 2018), rev’d sub nom. City of\nEscondido v. Emmons, 139 S. Ct. 500 (2019). 4 While these\ncases may help to identify whether the use of force against\nSanders amounted to a Fourth Amendment violation, they\ndo not clearly establish that a shooting in these\n\n 4\n Similarly, the district court also relied principally on Fourth\nAmendment cases in denying qualified immunity on this claim.\nHowever, as with Plaintiffs’ citations, some of these cases were decided\nafter the incident, e.g., Gelhaus, 871 F.3d 998, and one has since been\nreversed by the Supreme Court on the “clearly established” question.\nSee Kisela, 138 S. Ct. 1148. These problems present additional reasons\nwhy we find that the law was not clearly established on Plaintiffs’\nFourteenth Amendment claim.\n\f20 NICHOLSON V. GUTIERREZ\n\ncircumstances constitutes deliberate indifference to\nPlaintiffs. Sanders is not a plaintiff in this lawsuit, and\nPlaintiffs would not have standing to raise a Fourth\nAmendment claim on his behalf. See Plumhoff, 572 U.S. at\n778 (“Our cases make it clear that ‘Fourth Amendment\nrights are personal rights which . . . may not be vicariously\nasserted.’” (quoting Alderman v. United States, 394 U.S.\n165, 174 (1969)). The Fourth Amendment cases therefore\ndo not clearly establish the contours of the Fourteenth\nAmendment substantive due process rights at hand. 5\n\n Because no binding circuit or Supreme Court precedent\nhas established a substantive due process violation under\ncomparable circumstances, the Fourteenth Amendment right\nat issue lacked “contours . . . sufficiently definite” to place\nthe issue “beyond debate.” See Kisela, 138 S. Ct. at 1152–\n53 (quoting Plumhoff, 572 U.S. at 779). We accordingly\nreverse the district court and remand for an entry of qualified\nimmunity on this claim.\n\n\n\n 5\n Certainly, considerations of reasonableness germane to a Fourth\nAmendment analysis are relevant to the substantive due process inquiry.\nHistorically, our cases have recognized some overlap in these two\nconstitutional protections. See, e.g., P.B. v. Koch, 96 F.3d 1298, 1303\nn.4 (9th Cir. 1996) (holding that, “[r]egardless of the appropriate ‘home’\nfor plaintiffs’ right to be free from excessive force, there was a clearly\nestablished right to be free such force” because “[u]nder any standard,\n[the defendant’s] alleged actions were clearly unlawful.”). But we have\nheld that a Fourteenth Amendment claim of excessive force “must be\ngoverned by a different standard than” a Fourth Amendment claim of\nexcessive force. Byrd v. Guess, 137 F.3d 1126, 1133–34 (9th Cir. 1998),\nsuperseded by statute on other grounds as recognized in Little v. City of\nManhattan Beach, 21 Fed. App’x 651 (9th Cir. 2001). Thus, our Fourth\nAmendment cases cannot clearly establish the contours of the Fourteenth\nAmendment right, despite similarities between the standards.\n\f NICHOLSON V. GUTIERREZ 21\n\n III.\n\n CONCLUSION\n\n We reverse the district court’s denial of qualified\nimmunity to Gutierrez as to the Fourteenth Amendment\nclaim and affirm the court’s ruling as to the Fourth\nAmendment claim. We remand for proceedings consistent\nwith this opinion.\n\n REVERSED in part, AFFIRMED in part, and\nREMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Geraldine Nicholson v. Miguel Gutierrez\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-21\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nGERALDINE NICHOLSON; JOSE No. 17-56648\\nFERNANDO HUERTA; J. H., a minor\\nby and through his guardian ad litem, D.C. No.\\nJose Fernando Huerta; J. N. G., a 2:15-cv-07594-\\nminor by and through his guardian DDP-RAO\\nad litem, Geraldine Nicholson,\\n Plaintiffs-Appellees,\\n OPINION\\n v.\\n\\nCITY OF LOS ANGELES; EVERARDO\\nAMARAL, individually and in his\\nofficial capacity as a Police Officer\\nof the Los Angeles Police\\nDepartment,\\n Defendants,\\n\\n and\\n\\nMIGUEL GUTIERREZ, individually\\nand in his official capacity as a\\nPolice Officer of the Los Angeles\\nPolice Department,\\n Defendant-Appellant.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Dean D. Pregerson, District Judge, Presiding\\n\\f2 NICHOLSON V. GUTIERREZ\\n\\n Argued and Submitted March 4, 2019\\n Pasadena, California\\n\\n Filed August 21, 2019\\n\\n Before: Andrew J. Kleinfeld, R. Guy Cole, Jr., *\\n and Jacqueline H. Nguyen, Circuit Judges.\\n\\n Opinion by Judge Nguyen\\n\\n\\n SUMMARY **\\n\\n\\n Civil Rights\\n\\n The panel reversed in part and affirmed in part the\\ndistrict court’s denial of qualified immunity to a Los Angeles\\nPolice Department officer in an action brought pursuant to\\n42 U.S.C. § 1983 for violations of plaintiffs’ Fourth\\nAmendment rights to be free from excessive force and\\nunreasonable seizure and violations of their Fourteenth\\nAmendment substantive due process rights.\\n\\n Plaintiffs were among a group of teenagers who had met\\nin an alleyway near their school to listen to and sing rap\\nmusic. One of the teenagers, plaintiff J.N.G., was shot by\\ndefendant Gutierrez after Gutierrez mistook a plastic Airsoft\\nreplica gun held by one of the other teenagers for an actual\\n\\n *\\n The Honorable R. Guy Cole, Jr., United States Circuit Judge for\\nthe U.S. Court of Appeals for the Sixth Circuit, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f NICHOLSON V. GUTIERREZ 3\\n\\ngun. After the shooting, officers detained the group for over\\nfive hours while they investigated. J.N.G. and J.H. filed a\\nlawsuit and the district court denied qualified immunity on\\nplaintiffs’ Fourth and Fourteenth Amendment claims.\\n\\n Addressing the Fourth Amendment claim, the panel\\nagreed with the district court that under the circumstances,\\nplaintiffs’ continued detention for five hours after the\\nshooting—well after any probable cause would have\\ndissipated—and the use of handcuffs throughout the\\nduration of the detention violated plaintiffs’ clearly\\nestablished Fourth Amendment rights to be free from\\nunlawful arrest and excessive force. The panel rejected\\nGutierrez’s argument that while he participated in the initial\\nhandcuffing and detention, he was not responsible for any\\nsubsequent constitutional violation because he played no\\nrole in that conduct. The panel held that an officer can be\\nheld liable where he is just one participant in a sequence of\\nevents that gives rise to a constitutional violation. Here,\\nviewing the evidence in the light most favorable to plaintiffs,\\nGutierrez was more than a “mere bystander” in the alleged\\nconstitutional violations. The panel affirmed the district\\ncourt’s denial of qualified immunity on the Fourth\\nAmendment violations because, ultimately, a reasonable\\njury could conclude that Gutierrez played an integral role in\\nthe unlawfully prolonged detention and sustained\\nhandcuffing of plaintiffs.\\n\\n Addressing the Fourteenth Amendment substantive due\\nprocess claim, the panel held that, viewing the totality of the\\nevidence in the light most favorable to the plaintiffs, the\\nshooting violated plaintiffs’ due process rights. Under the\\ncircumstances, a rational finder of fact could find that\\nGutierrez’s use of deadly force shocked the conscience and\\nwas unconstitutional under the Fourteenth Amendment.\\n\\f4 NICHOLSON V. GUTIERREZ\\n\\nNevertheless, the panel held that because no analogous case\\nexisted at the time of the shooting, the district court erred by\\ndenying Gutierrez qualified immunity for this claim. The\\npanel accordingly reversed the district court and remanded\\nfor an entry of qualified immunity on the Fourteenth\\nAmendment claim.\\n\\n\\n COUNSEL\\n\\nDenise L. Rocawich (argued) and James R. Touchstone,\\nJones & Mayer, Fullerton, California, for Defendant-\\nAppellant.\\n\\nHerbert-John S. Hayden (argued) and John W. Harris, Harris\\n& Associates, Los Angeles, California, for Plaintiffs-\\nAppellees.\\n\\n\\n OPINION\\n\\nNGUYEN, Circuit Judge:\\n\\n On the morning of February 10, 2015, four teenagers met\\nin an alleyway near their school to listen to and sing rap\\nmusic. As the teenagers—Michael Sanders, Abdul Wooten,\\nJ.N.G., and J.H. 1—stood in a tight circle dancing and\\nrapping, Sanders was holding a plastic Airsoft replica gun\\nwith a bright orange tip as a prop. Just as they turned off the\\nmusic and were getting ready to head to school, J.N.G. was\\n\\n 1\\n Throughout the record, J.N.G. (Jamar Nicholson Green) and J.H.\\n(Jason Huerta) were referred to by their initials because they were minors\\nat the time of the incident and for some time after filing this lawsuit. For\\nconsistency, we also refer to them by their initials.\\n\\f NICHOLSON V. GUTIERREZ 5\\n\\nshot by Officer Michael Gutierrez of the Los Angeles Police\\nDepartment (“LAPD”). Officer Gutierrez fired his weapon\\nbecause he mistook Sanders’s replica gun for an actual gun.\\nGutierrez fired multiple shots, one of which hit J.N.G. in the\\nback. After the shooting, officers detained the group for over\\nfive hours while they investigated.\\n\\n J.N.G. and J.H. (collectively, “Plaintiffs”) filed a lawsuit\\nagainst the officers, the LAPD, and the City of Los Angeles,\\nalleging violations of the Fourth and Fourteenth\\nAmendments and various state laws. The district court\\ndenied qualified immunity on two of Plaintiffs’\\nconstitutional claims. Gutierrez appeals. We affirm in part\\nand reverse in part.\\n\\n BACKGROUND 2\\n\\n At around 7:15 a.m. on February 10, 2015, J.N.G., J.H.,\\nMichael Sanders, and Abdul Wooten met in an alley at the\\ncorner of 10th Avenue and Florence Avenue in Los Angeles,\\nCA, a few blocks from their high school. They regularly\\ngathered in that alleyway before and after school to listen to\\nmusic and freestyle rap. That morning, as they were rapping\\nand dancing in a circle, Sanders was holding a plastic toy\\ngun with a bright orange tip. J.N.G., J.H., and Wooten\\nmaintain that Sanders kept the gun pointed downward\\naround waist-level and did not fire the gun that morning. At\\napproximately 7:40 a.m., the teenagers turned off the music\\nand began preparing to head to school.\\n\\n\\n\\n\\n 2\\n At this stage of the proceedings, we view the facts in the light most\\nfavorable to Plaintiffs. George v. Morris, 736 F.3d 829, 836 (9th Cir.\\n2013) (9th Cir. 2013).\\n\\f6 NICHOLSON V. GUTIERREZ\\n\\n Around this time, Officer Everardo Amaral was driving\\ndown 10th Avenue in an unmarked car with his partner,\\nOfficer Gutierrez. From the passenger seat, Gutierrez “saw\\na person (later identified as Michael Sanders) pointing . . . a\\nblue steel handgun at another person (later identified as\\nPlaintiff J.H.).” Gutierrez, believing that J.H. was “being\\nrobbed at gun point or was about to be murdered,” yelled\\n“Gun, gun, gun!” Amaral stopped the vehicle south of the\\nalley on 10th Avenue. Without conferring with Amaral,\\nGutierrez immediately jumped out of the car and ran into the\\nalley. Amaral parked the car and followed Gutierrez.\\nNeither officer was in uniform.\\n\\n Gutierrez claims he identified himself as an LAPD\\nofficer and commanded Sanders to drop the gun. However,\\nJ.H., J.N.G., and Wooten all contend that Gutierrez did not\\nidentify himself or make any verbal commands prior to\\nshooting his weapon. A few seconds after he entered the\\nalley, Officer Gutierrez fired at least three shots, one of\\nwhich hit J.N.G. in the back. J.N.G. and J.H. contend that\\nGutierrez fired his gun with one hand while running toward\\nthem, while Gutierrez stated that he fired only after stopping\\na few feet away from the group. When the shots were fired,\\nJ.H. was about to put on his school uniform, and J.N.G. was\\nspraying cologne on his face. The four of them had been\\nstanding in a tight circle, “within a foot or so of each other.”\\nSanders soon turned and dropped the toy gun, though the\\nparties dispute whether this occurred before or after\\nGutierrez fired.\\n\\n Shortly after Officer Gutierrez fired, Officer Amaral\\narrived and requested three additional units. Amaral also\\nrequested an ambulance when he realized that J.N.G. had\\nbeen shot. The officers held the group at gunpoint, face\\ndown on the ground. The parties dispute how far the\\n\\f NICHOLSON V. GUTIERREZ 7\\n\\ndropped “gun” was from the teenagers when they were on\\nthe ground, but neither officer picked it up or moved it away\\nfrom them. While on the ground, J.H. shouted that the gun\\nwas “not even a real gun” and repeatedly asked why the\\nofficers shot at them and “What did we do wrong?” The\\nofficers remained silent in response to his questions, with\\ndumbfounded expressions on their faces. Responding\\nofficers soon arrived, and they searched and handcuffed the\\ngroup. Gutierrez was “involved in the decision to handcuff\\nthem.” Officer Amaral later explained in his deposition that\\nthe detention of the boys was “[f]or [a] weapons violation.”\\nOfficers Gutierrez and Amaral were separated and\\nmonitored soon after additional units arrived on the scene.\\nJ.H. remained in handcuffs throughout the investigation,\\nwhich lasted until around 1:00 p.m., over five hours after the\\nshooting. J.N.G. also remained in handcuffs for over five\\nhours—through the duration of his hospital examination—\\nuntil detectives interrogated him.\\n\\n J.N.G., by and through his mother and guardian ad litem\\nGeraldine Nicholson, and J.H., by and through his father and\\nguardian ad litem Jose Fernando Huerta, sued Officer\\nGutierrez, Officer Amaral, the City of Los Angeles, the\\nLAPD, Chief of Police Charles Beck, and Commander\\nAndrew Smith. In addition to various state law claims,\\nPlaintiffs J.N.G. and J.H. alleged claims under 42 U.S.C.\\n§ 1983 for violations of their Fourth Amendment rights to be\\nfree from excessive force and unreasonable seizure and\\nviolations of their Fourteenth Amendment substantive due\\nprocess rights. Defendants Gutierrez, Amaral, and the City\\nof Los Angeles jointly moved for summary judgment,\\narguing in part that no constitutional violation occurred and\\nthat qualified immunity applied.\\n\\f8 NICHOLSON V. GUTIERREZ\\n\\n The district court granted in part and denied in part\\nDefendants’ joint motion for summary judgment. Relevant\\nto this appeal, the district court denied Gutierrez qualified\\nimmunity in part on Plaintiffs’ Fourth Amendment claim and\\non Plaintiffs’ Fourteenth Amendment claim. Gutierrez\\ntimely appealed.\\n\\n JURISDICTION AND STANDARD OF REVIEW\\n\\n We have jurisdiction under 28 U.S.C. § 1291 to review\\nthe denial of qualified immunity at the summary judgment\\nstage. Plumhoff v. Rickard, 572 U.S. 765, 771–73 (2014).\\nBut “the scope of our review over the appeal is\\ncircumscribed” because we may not “consider questions of\\neviden[tiary] sufficiency, i.e., which facts a party may, or\\nmay not, be able to prove at trial.” Morris, 736 F.3d at 834\\n(quoting CarePartners, LLC v. Lashway, 545 F.3d 867, 875\\n(9th Cir. 2008)). “Thus, in this appeal, we are confined to\\nthe question of ‘whether the defendant[s] would be entitled\\nto qualified immunity as a matter of law, assuming all factual\\ndisputes are resolved, and all reasonable inferences are\\ndrawn, in plaintiff’s favor.’” Id. at 836 (quoting Karl v. City\\nof Mountlake Terrace, 678 F.3d 1062, 1068 (9th Cir. 2012)).\\n\\n DISCUSSION\\n\\n “Qualified immunity attaches when an official’s conduct\\ndoes not violate clearly established statutory or\\nconstitutional rights of which a reasonable person would\\nhave known.” Kisela v. Hughes, 138 S. Ct. 1148, 1152\\n(2018) (per curiam). Once a defendant has raised qualified\\nimmunity as a defense to a claim, a plaintiff must show\\n“(1) that the right was violated; and (2) that the right was\\nclearly established at the time of the alleged misconduct.”\\nIsayeva v. Sacramento Sheriff’s Dep’t, 872 F.3d 938, 946\\n(9th Cir. 2017). Although a right is not clearly established\\n\\f NICHOLSON V. GUTIERREZ 9\\n\\nwhere merely defined “at a high level of generality,”\\nqualified immunity does not “require a case directly on\\npoint.” Kisela, 138 S. Ct. at 1152. Instead, the “focus is on\\nwhether the officer had fair notice that her conduct was\\nunlawful,” id., for example, through “any cases of\\ncontrolling authority in their jurisdiction at the time of the\\nincident,” Wilson v. Layne, 526 U.S. 603, 617 (1999).\\n\\n We now turn to each of Plaintiffs’ constitutional claims.\\n\\n I. Fourth Amendment Claim\\n\\n Plaintiffs allege that, after the shooting, Defendants\\nviolated their Fourth Amendment rights by unlawfully\\narresting them, using excessive force, and prolonging their\\ndetention. 3 In denying Officer Gutierrez qualified\\nimmunity, the district court concluded that immediately after\\nthe shooting, an investigatory stop was reasonable while the\\nofficers assessed the situation. But “[a]t some point, the\\ndetention evolved into a full-fledged arrest that required\\nprobable cause that J.H. and J.N.G. had been engaged in\\ncriminal activity.” The district court further concluded that\\n“a reasonable jury could determine that the sustained\\nhandcuffing of J.H. and J.N.G. . . . constituted excessive\\nforce.”\\n\\n\\n 3\\n Plaintiffs also alleged that the shooting itself also violated their\\nFourth Amendment rights. The district court granted summary judgment\\nin favor of Defendants on this claim, finding that the teenagers were not\\n“seized” by Gutierrez’s gunfire under the Fourth Amendment because\\nGutierrez intended to use deadly force against Sanders, not against\\nJ.N.G. or J.H. See United States v. Al Nasser, 555 F.3d 722, 728 (9th\\nCir. 2009). Plaintiffs do not challenge this ruling and thus the question\\nof whether the shooting violated their Fourth Amendment rights is not\\nbefore us in this appeal.\\n\\f10 NICHOLSON V. GUTIERREZ\\n\\n It is well-established that a “person may not be arrested,\\nor must be released from arrest, if previously established\\nprobable cause has dissipated.” United States v. Ortiz-\\nHernandez, 427 F.3d 567, 574 (9th Cir. 2005) (per curiam).\\n“As a corollary . . . of the rule that the police may rely on the\\ntotality of facts available to them in establishing probable\\ncause, they also may not disregard facts tending to dissipate\\nprobable cause.” Id. (quoting Bigford v. Taylor, 834 F.2d\\n1213, 1218 (5th Cir. 1988)). A reasonable officer would\\nknow that participation in an ongoing seizure after any\\nprobable cause had dissipated violates the Fourth\\nAmendment.\\n\\n Here, it was soon apparent to the officers that the\\nteenagers were unarmed, posed no threat to anyone, and\\nwere not engaged in any criminal activity. The incident\\noccurred in the morning right before the start of school\\nhours, and Plaintiffs had their school uniforms and\\nbackpacks. In fact, as Officer Gutierrez approached the\\nscene, J.N.G. was spraying on cologne and J.H. was donning\\nhis school uniform. Moreover, Officer Gutierrez admitted\\nthat he perceived at least J.H. to be a possible victim, not a\\nsuspect, further undermining any justification to detain him.\\nWe agree with the district court that under these\\ncircumstances, Plaintiffs’ continued detention for five\\nhours—well after any probable cause would have\\ndissipated—and the use of handcuffs throughout the\\nduration of the detention violated Plaintiffs’ clearly\\nestablished Fourth Amendment rights to be free from\\nunlawful arrest and excessive force.\\n\\n On appeal, Officer Gutierrez does not appear to dispute\\nthat the prolonged detention and handcuffing violated\\nPlaintiffs’ clearly established Fourth Amendment rights.\\nInstead, Gutierrez focuses his argument on the fact that he\\n\\f NICHOLSON V. GUTIERREZ 11\\n\\nwas separated and monitored after the shooting. Thus, his\\nargument goes, while he admittedly “was an integral\\nparticipant in the initial handcuffing and detention,” he is not\\nresponsible for any subsequent constitutional violation\\nbecause he played no role in that conduct.\\n\\n A police officer need not have been the sole party\\nresponsible for a constitutional violation before liability may\\nattach. “An officer’s liability under section 1983 is\\npredicated on his ‘integral participation’ in the alleged\\nviolation.” Blankenhorn v. City of Orange, 485 F.3d 463,\\n481 n.12 (9th Cir. 2007) (quoting Chuman v. Wright, 76 F.3d\\n292, 294–95 (9th Cir. 1996)). This theory of liability “does\\nnot require that each officer’s actions themselves rise to the\\nlevel of a constitutional violation.” Id. (quoting Boyd v.\\nBenton County, 374 F.3d 773, 780 (9th Cir. 2004)). Instead,\\nliability may attach if the officer has “some fundamental\\ninvolvement in the conduct that allegedly caused the\\nviolation.” Id. A theory of integral participation thus\\ncomports with general tort principles of causation applicable\\nto a § 1983 action: “[G]overnment officials, like other\\ndefendants, are generally responsible for the ‘natural’ or\\n‘reasonably foreseeable’ consequences of their actions.”\\nStoot v. City of Everett, 582 F.3d 910, 926 (9th Cir. 2009).\\nEven though “an intervening decision of an informed,\\nneutral decision-maker ‘breaks’ the chain of causation,” the\\nchain of causation is not broken where the intervening\\ndecision was foreseeably influenced by the defendant. Id.\\n(quoting Murray v. Earle, 405 F.3d 278, 292 (5th Cir.\\n2005)). Thus, under our case law, an officer could be held\\nliable where he is just one participant in a sequence of events\\nthat gives rise to a constitutional violation.\\n\\n For example, in Boyd v. Benton County, we held that\\neach of the officers in a search operation were liable for\\n\\f12 NICHOLSON V. GUTIERREZ\\n\\nexcessive force for the use of a flash-bang grenade, even\\nthough only one officer reached into the home and deployed\\nthe device. 374 F.3d at 780. We reasoned that the use of the\\ndevice was “part of the search operation in which every\\nofficer participated in some meaningful way.” Id.\\nMoreover, each officer “was aware of the decision . . . , did\\nnot object to it, and participated in the search operation\\nknowing the flash-bang was to be deployed.” Id. By\\ncontrast, in Torres v. City of Los Angeles, we found that a\\ndetective was not an integral participant in an allegedly\\nunlawful arrest, in part because she “was not present [at the\\narrest], and there is no evidence that [she] instructed the\\nother detectives to arrest [the plaintiff] or that any of those\\ndetectives consulted with her before making the arrest.”\\n548 F.3d 1197, 1206 (9th Cir. 2008).\\n\\n Here, viewing the evidence in the light most favorable to\\nPlaintiffs, Gutierrez was more than a “mere bystander” in the\\nalleged constitutional violations. See Chuman, 76 F.3d at\\n294. Plaintiffs alleged that Officers Amaral and Gutierrez\\n“directed the other officers to handcuff, search and arrest all\\nof us for reasons unknown to any of us.” Gutierrez himself\\nacknowledged that he was “involved in the decision to\\nhandcuff [Plaintiffs].” In contrast to the absent officer who\\nwas not consulted prior to the arrest in Torres, Gutierrez was\\nthe initial officer who set these events into motion, and either\\ninstructed the other officers to arrest Plaintiffs or consulted\\nwith them in that decision. See 548 F.3d at 1206. The\\ndistrict court did not decide exactly when during the\\nprolonged detention probable cause dissipated and,\\nsimilarly, how proximate Gutierrez’s conduct was to that\\nviolation. But these issues present questions for the jury\\nbecause their resolution depends on disputed material facts.\\nUltimately, because a reasonable jury could conclude that\\nGutierrez played an integral role in the unlawfully prolonged\\n\\f NICHOLSON V. GUTIERREZ 13\\n\\ndetention and sustained handcuffing of Plaintiffs, we affirm\\nthe district court’s denial of qualified immunity on the\\nFourth Amendment violation.\\n\\n II. Fourteenth Amendment Claim\\n\\n Plaintiffs contend that the unlawful shooting violated\\ntheir substantive due process rights under the Fourteenth\\nAmendment. The district court denied Gutierrez qualified\\nimmunity because a jury could reasonably conclude that his\\nconduct amounted to deliberate indifference. The district\\ncourt also found that the substantive due process right at\\nissue was clearly established. We first address whether\\nGutierrez’s conduct violated Plaintiffs’ substantive due\\nprocess rights under the Fourteenth Amendment, and then\\nwhether the right was clearly established at the time of the\\nincident.\\n\\n A.\\n\\n To prevail on a substantive due process claim under the\\nFourteenth Amendment, Plaintiffs must show that an\\nofficer’s conduct “shocks the conscience.” See Wilkinson v.\\nTorres, 610 F.3d 546, 554 (9th Cir. 2010). The “critical\\nconsideration [is] whether the circumstances are such that\\nactual deliberation is practical.” Porter v. Osborn, 546 F.3d\\n1131, 1137 (9th Cir. 2008) (quoting Moreland v. Las Vegas\\nMetro. Police Dep’t, 159 F.3d 365, 372 (9th Cir. 1998)). If\\nso, “an officer’s ‘deliberate indifference’ may suffice to\\nshock the conscience,” Wilkinson, 610 F.3d at 554, and the\\nplaintiff may prevail by showing that the officer\\n“disregarded a known or obvious consequence of his\\naction,” Patel v. Kent Sch. Dist., 648 F.3d 965, 974 (9th Cir.\\n2011). The “deliberate-indifference inquiry should go to the\\njury if any rational factfinder could find this requisite mental\\nstate.” Id. Here, having found a triable issue on whether\\n\\f14 NICHOLSON V. GUTIERREZ\\n\\n“deliberation was practical under the circumstances,” the\\ndistrict court held that “[a] finder of fact could conclude . . .\\nthat Gutierrez disregarded the known or obvious risks of\\ninjury to J.H. and J.N.G. when he fired at Sanders without\\ntaking time to assess the situation.”\\n\\n We agree with the district court and hold that, viewing\\nthe totality of the evidence in the light most favorable to the\\nPlaintiffs, the shooting violated Plaintiffs’ due process\\nrights. We do not discount the seriousness of the situation\\nthat Officer Gutierrez thought he observed: a person holding\\nwhat appeared to be a gun standing near others who may\\nhave been in danger. But Sanders “was not engaged in any\\nthreatening or menacing behavior, and he kept the airsoft\\ngun securely pointed toward the ground.” The alleyway was\\nnear a school, and Plaintiffs were “equipped with school\\nuniforms and backpacks . . . [appearing] to be minors on\\ntheir way to school and not gang members.” Yet within\\nseconds of observing the “gun,” without consulting with his\\npartner, Gutierrez rushed down the alleyway. As he ran, he\\nfired his gun toward both Sanders, the perceived perpetrator\\nof a possible crime, and innocent bystanders, with one bullet\\nultimately striking J.N.G. in the back. Under these\\ncircumstances, a rational finder of fact could find that\\nGutierrez’s use of deadly force shocks the conscience and\\nwas unconstitutional under the Fourteenth Amendment.\\n\\n Gutierrez’s arguments to the contrary are unpersuasive.\\nAs a preliminary matter, Gutierrez challenges the district\\ncourt’s finding that a reasonable jury could find that\\ndeliberation was practical under the circumstances. But\\nbecause this is an interlocutory appeal, we are not free to\\nrevisit the district court’s conclusions as to “which facts a\\nparty may, or may not, be able to prove at trial.” See Morris,\\n736 F.3d at 834. Our review is limited to whether “the denial\\n\\f NICHOLSON V. GUTIERREZ 15\\n\\nof qualified immunity was appropriate by assuming that the\\nversion of the material facts asserted by the non-moving\\nparty is correct.” Bingue v. Prunchak, 512 F.3d 1169, 1172–\\n73 (9th Cir. 2008). Assuming the facts in Plaintiffs’ favor,\\nno attendant circumstances weighed in favor of the\\nimmediate use of deadly force, other than Gutierrez’s belief\\nthat Sanders was holding a gun. As the district court noted,\\nunder these circumstances, Gutierrez’s immediate use of\\nforce without communicating with his partner, his failure to\\nseek cover, and his failure to formulate a plan before acting\\nwere all contrary to LAPD’s training and policy. We agree\\nwith the district court that the evidence is sufficient to create\\na genuine dispute of fact on whether deliberation was\\npractical under the circumstances, and, in any event, we\\nwould be without power to reverse on this ground.\\n\\n Gutierrez next argues that, as a matter of law, the district\\ncourt erred in failing to apply the “intent to harm” standard,\\nunder which Plaintiffs must show that he acted “with a\\npurpose to harm unrelated to legitimate law enforcement\\nobjectives.” Wilkinson, 610 F.3d at 554. By that standard,\\nGutierrez argues, he is entitled to qualified immunity even if\\nhe acted with deliberate indifference, because he did not\\nintend to shoot J.N.G.\\n\\n We have previously carved out a narrow situation—\\nhigh-speed police car chases—in which we have found,\\ncategorically, that an officer does not have time to deliberate.\\nBingue, 512 F.3d at 1177. In Bingue v. Prunchak, we\\nconsidered whether high-speed pursuits should categorically\\ngive rise to the application of the “intent to harm” standard,\\nor whether the “deliberate indifference” standard may apply\\ndepending on the circumstances. In holding that “the intent\\nto harm” standard “applies to all high-speed police chases,”\\nid., we reasoned that such a rule best accounts for an officer’s\\n\\f16 NICHOLSON V. GUTIERREZ\\n\\n“repeated split-second decisions about how best to\\napprehend the fleeing suspect in a manner that will minimize\\nrisk to [the officer’s] own safety and the safety of the general\\npublic.” Id. at 1176. A suspect fleeing in a car at high speed\\ngives an officer “no time for reflection and precious little\\ntime for deliberation concerning either the decision to join\\nthe chase in the first place or the serial decisions about how\\nbest to pursue the suspect.” Id.\\n\\n Here, in contrast, under Plaintiffs’ version of the facts,\\nGutierrez may have seen what he believed to be a gun\\n(although it had the orange tip signifying a toy), but he did\\nnot see Sanders point it at anyone. Sanders was standing\\namong a group of school-aged youths with their backpacks\\nand school uniforms. Without more, we cannot say that this\\nsituation gives an officer “no time for reflection,” as in a\\nhigh-speed chase involving a fleeing suspect in a fast-\\nmoving vehicle. Cf. Estate of Lopez v. Gelhaus, 871 F.3d\\n998, 1011 (9th Cir. 2017), cert. denied, 138 S. Ct. 2680\\n(2018) (holding that mistaking a child’s toy gun for an\\nassault rifle did not as a matter of law justify an officer’s use\\nof deadly force).\\n\\n Gutierrez also relies on cases involving serious and\\nimmediate threats to public safety, but these cases too are\\neasily distinguishable. In Porter v. Osborn, the suspect\\nengaged in “evasive actions” in response to questions and\\ndirect orders from police officers, including refusing to exit\\nhis vehicle and then driving his vehicle in the direction of\\none of the officers in a perceived attempt to run over the\\nofficer. 546 F.3d at 1134–35, 1137. There, we held that the\\n“rapidly escalating nature” of the confrontation between the\\nofficer and the suspect behind the wheel left too little time\\nfor adequate deliberation and thus necessitated application\\nof the “intent to harm” standard. Id. In Moreland v. Las\\n\\f NICHOLSON V. GUTIERREZ 17\\n\\nVegas Metropolitan Police Department, the officers were\\nresponding to “the extreme emergency” of an active\\n“gunfight in progress threaten[ing] the lives of the 50 to 100\\npeople who were trapped in the parking lot.” 159 F.3d\\nat 372–73. The suspect was indisputably firing a\\nsemiautomatic handgun and refusing to comply with the\\nofficers’ orders to stop. Id. at 372. Given the ongoing crisis,\\nwe held that the officers had no opportunity to deliberate in\\nlight of the “immediate risk of serious harm or death to the\\nmany innocent individuals trapped in the parking lot.” Id.\\nThus, the “intent to harm” standard was appropriate. By\\ncontrast, assuming Plaintiffs’ contentions that Sanders was\\nnot pointing the toy gun at anyone and had not given any\\nindication that he was likely to harm anyone, there was no\\n“rapidly escalating” confrontation or “extreme emergency”\\nhere that would have deprived Gutierrez of the opportunity\\nto confer with his partner and formulate a plan to ascertain\\nwhat was happening before charging in with gunfire.\\nContrary to Gutierrez’s argument, we have not previously\\napplied the intent to harm standard to Fourteenth\\nAmendment claims involving facts similar to this case, and\\nwe decline to do so now.\\n\\n We thus agree that application of the deliberate\\nindifference standard is warranted under these\\ncircumstances. As the district court explained, in “minimal\\ninformation” situations, an officer must take some time to\\nassess what is happening before employing deadly force.\\nHolding otherwise would result in an “intolerably high risk\\nof a tragic shooting that may otherwise have been avoided\\nby proper deliberation whenever practical.” As such,\\napplying the deliberate indifference standard to Plaintiffs’\\nversion of the facts, we hold that Gutierrez’s shooting\\nviolated their substantive due process rights under the\\nFourteenth Amendment.\\n\\f18 NICHOLSON V. GUTIERREZ\\n\\n B.\\n\\n Even if a constitutional violation occurred, qualified\\nimmunity nevertheless applies unless the violation was\\nclearly established. Because no analogous case existed at\\nthe time of the shooting, we hold that the district court erred\\nin denying Gutierrez qualified immunity for this claim.\\n\\n Kisela v. Hughes is instructive. 138 S. Ct. at 1151. In\\nKisela, the Supreme Court reversed our decision denying\\nqualified immunity to an officer for an excessive force claim\\nafter he shot an individual armed with a knife as she\\napproached a bystander. Id. The Court reiterated its\\nrepeated admonition to courts “and the Ninth Circuit in\\nparticular—not to define clearly established law at a high\\nlevel of generality.” Id. at 1152 (quoting City and County of\\nSan Francisco v. Sheehan, 135 S. Ct. 1765, 1775–76\\n(2015)). The Court explained that, where “the result\\ndepends very much on the facts of each case . . . officers are\\nentitled to qualified immunity unless existing precedent\\n‘squarely governs’ the specific facts at issue.” Id. at 1153\\n(quoting Mullenix v. Luna, 136 S. Ct. 305, 309 (2015) (per\\ncuriam)).\\n\\n Here, Plaintiffs failed to identify any authority that\\nrendered the contours of the substantive due process right at\\nissue “sufficiently definite that any reasonable official in the\\ndefendant’s shoes would have understood he was violating\\nit.” See Kisela, 138 S. Ct. at 1153. In their briefing,\\nPlaintiffs cited cases establishing broadly that “the\\nConstitution protects a citizen’s liberty interest in her own\\nbodily security,” which define the right at much too high a\\nlevel of generality to clearly establish a rule of conduct. See,\\ne.g., Ingraham v. Wright, 430 U.S. 651, 673–74 (1977).\\nThey also discuss our “state-created danger exception”\\ncases, but these involve failures to act that lead to injuries\\n\\f NICHOLSON V. GUTIERREZ 19\\n\\nfrom third parties, rather than affirmative actions by officers\\nthat directly cause injury to the plaintiff. See, e.g., Kennedy\\nv. City of Ridgefield, 439 F.3d 1055 (9th Cir. 2006) (failure\\nto warn allegedly caused shooting by neighbor); Patel,\\n648 F.3d 965 (failure to adequately supervise a disabled high\\nschool student allegedly caused sexual abuse by another\\nstudent); Wood v. Ostrander, 879 F.2d 583 (9th Cir. 1989)\\n(failure to leave plaintiff in safe location allegedly caused\\nrape in a high-crime area). These cases are too factually\\ndissimilar to clearly establish a constitutional violation by an\\nofficer’s accidental shooting of a bystander.\\n\\n At oral argument, Plaintiffs conceded that it was “it was\\ndifficult to find a case that was squarely on point,” where a\\ncourt found a constitutional violation in the context of a\\nbystander shooting. Instead, the gravamen of Plaintiffs’\\nanalysis is that the use of deadly force against Sanders was\\nlikely unreasonable, relying principally on our cases\\nanalyzing Fourth Amendment claims of excessive force.\\nE.g., Hughes v. Kisela, 862 F.3d 775, 789 (9th Cir. 2016),\\nrev’d sub nom. Kisela v. Hughes, 138 S. Ct. 1148 (2018);\\nEmmons v. City of Escondido, No. 16-55771, 2018 WL\\n1531064 (9th Cir. Mar. 29, 2018), rev’d sub nom. City of\\nEscondido v. Emmons, 139 S. Ct. 500 (2019). 4 While these\\ncases may help to identify whether the use of force against\\nSanders amounted to a Fourth Amendment violation, they\\ndo not clearly establish that a shooting in these\\n\\n 4\\n Similarly, the district court also relied principally on Fourth\\nAmendment cases in denying qualified immunity on this claim.\\nHowever, as with Plaintiffs’ citations, some of these cases were decided\\nafter the incident, e.g., Gelhaus, 871 F.3d 998, and one has since been\\nreversed by the Supreme Court on the “clearly established” question.\\nSee Kisela, 138 S. Ct. 1148. These problems present additional reasons\\nwhy we find that the law was not clearly established on Plaintiffs’\\nFourteenth Amendment claim.\\n\\f20 NICHOLSON V. GUTIERREZ\\n\\ncircumstances constitutes deliberate indifference to\\nPlaintiffs. Sanders is not a plaintiff in this lawsuit, and\\nPlaintiffs would not have standing to raise a Fourth\\nAmendment claim on his behalf. See Plumhoff, 572 U.S. at\\n778 (“Our cases make it clear that ‘Fourth Amendment\\nrights are personal rights which . . . may not be vicariously\\nasserted.’” (quoting Alderman v. United States, 394 U.S.\\n165, 174 (1969)). The Fourth Amendment cases therefore\\ndo not clearly establish the contours of the Fourteenth\\nAmendment substantive due process rights at hand. 5\\n\\n Because no binding circuit or Supreme Court precedent\\nhas established a substantive due process violation under\\ncomparable circumstances, the Fourteenth Amendment right\\nat issue lacked “contours . . . sufficiently definite” to place\\nthe issue “beyond debate.” See Kisela, 138 S. Ct. at 1152–\\n53 (quoting Plumhoff, 572 U.S. at 779). We accordingly\\nreverse the district court and remand for an entry of qualified\\nimmunity on this claim.\\n\\n\\n\\n 5\\n Certainly, considerations of reasonableness germane to a Fourth\\nAmendment analysis are relevant to the substantive due process inquiry.\\nHistorically, our cases have recognized some overlap in these two\\nconstitutional protections. See, e.g., P.B. v. Koch, 96 F.3d 1298, 1303\\nn.4 (9th Cir. 1996) (holding that, “[r]egardless of the appropriate ‘home’\\nfor plaintiffs’ right to be free from excessive force, there was a clearly\\nestablished right to be free such force” because “[u]nder any standard,\\n[the defendant’s] alleged actions were clearly unlawful.”). But we have\\nheld that a Fourteenth Amendment claim of excessive force “must be\\ngoverned by a different standard than” a Fourth Amendment claim of\\nexcessive force. Byrd v. Guess, 137 F.3d 1126, 1133–34 (9th Cir. 1998),\\nsuperseded by statute on other grounds as recognized in Little v. City of\\nManhattan Beach, 21 Fed. App’x 651 (9th Cir. 2001). Thus, our Fourth\\nAmendment cases cannot clearly establish the contours of the Fourteenth\\nAmendment right, despite similarities between the standards.\\n\\f NICHOLSON V. GUTIERREZ 21\\n\\n III.\\n\\n CONCLUSION\\n\\n We reverse the district court’s denial of qualified\\nimmunity to Gutierrez as to the Fourteenth Amendment\\nclaim and affirm the court’s ruling as to the Fourth\\nAmendment claim. We remand for proceedings consistent\\nwith this opinion.\\n\\n REVERSED in part, AFFIRMED in part, and\\nREMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4091,"cells":{"id":{"kind":"number","value":4654959,"string":"4,654,959"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Trina Ray v. County of Los Angeles"},"decision_date":{"kind":"string","value":"2019-08-22"},"docket_number":{"kind":"string","value":"17-56581"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Kim McLane Wardlaw and Mark J. Bennett, Circuit Judges, and Kathleen Cardone, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"BENNETT, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nTRINA RAY, individually, and on No. 17-56581\\nbehalf of others similarly situated,\\n Plaintiff-Appellee, D.C. No.\\n 2:17-cv-04239-\\n v. PA-SK\\n\\nCOUNTY OF LOS ANGELES,\\n Defendant-Appellant.\\n\\n\\nTRINA RAY; SASHA WALKER, No. 18-55276\\nindividually, and on behalf of all\\nothers similarly situated, D.C. No.\\n Plaintiffs-Appellants, 2:17-cv-04239-\\n PA-SK\\n v.\\n\\nLOS ANGELES COUNTY DEPARTMENT OPINION\\nOF PUBLIC SOCIAL SERVICES,\\nErroneously Sued As County of Los\\nAngeles,\\n Defendant-Appellee.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Percy Anderson, District Judge, Presiding\\n\\f2 RAY V. COUNTY OF LOS ANGELES\\n\\n Argued and Submitted March 7, 2019\\n Pasadena, California\\n\\n Filed August 22, 2019\\n\\n Before: Kim McLane Wardlaw and Mark J. Bennett,\\n Circuit Judges, and Kathleen Cardone, * District Judge.\\n\\n Opinion by Judge Bennett\\n\\n\\n SUMMARY **\\n\\n\\n Labor Law / Eleventh Amendment Immunity\\n\\n The panel affirmed the district court’s order denying a\\ndefendant county’s motion to dismiss, on Eleventh\\nAmendment immunity grounds, a putative collective action\\nunder the Fair Labor Standards Act; reversed the district\\ncourt’s order regarding the putative collective period; and\\nremanded.\\n\\n Plaintiff homecare providers were employed through\\nCalifornia’s In-Home Supportive Services program, which\\nis implemented and run by the State and its counties. In\\nOctober 2013, the Department of Labor promulgated a new\\nrule providing that homecare providers would be entitled to\\novertime pay under the FLSA. The final rule had an\\n\\n *\\n The Honorable Kathleen Cardone, United States District Judge for\\nthe Western District of Texas, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f RAY V. COUNTY OF LOS ANGELES 3\\n\\neffective date of January 1, 2015. In 2014, the District Court\\nfor the District of Columbia vacated the rule. On August 21,\\n2015, the D.C. Circuit reversed and ordered the district court\\nto enter summary judgment for the Department of Labor. On\\nSeptember 14, 2015, the Department of Labor announced\\nthat it would not bring enforcement actions against any\\nemployer for violations of the new rule for 30 days after\\nissuance of the mandate of the D.C. Circuit. On October 27,\\n2015, the Department of Labor said it would not begin\\nenforcing the new rule until November 12, 2015. The State\\nbegan paying overtime wages on February 1, 2016.\\n\\n Affirming in part, the panel held that the County of Los\\nAngeles was not entitled to Eleventh Amendment immunity.\\nThe panel assumed without deciding that a county might be\\nentitled to immunity if acting as an arm of the state. The\\npanel held that, under the five-part Mitchell test, the County\\nwas not an arm of the State when it administered the IHSS\\nprogram because the state-treasury factor, which is the most\\nimportant, and all but one of the other Mitchell factors\\nweighed against immunity. The panel held that a later\\nSupreme Court case, Hess v. Port Auth. Trans-Hudson\\nCorp., 513 U.S. 30 (1994), did not undermine Mitchell such\\nthat it should be overruled.\\n\\n Reversing in part, the panel held that the effective date\\nof the Department of Labor’s rule was January 1, 2015,\\nbecause the legal effect of the D.C. Circuit’s vacatur was to\\nreinstate the original effective date. The panel held that the\\nDepartment of Labor’s choice against enforcing the rule\\nuntil November 12, 2015, did not eliminate the availability\\nof private rights of action until that date. Accordingly, the\\nbeginning of the putative collective period was January 1,\\n2015.\\n\\f4 RAY V. COUNTY OF LOS ANGELES\\n\\n COUNSEL\\n\\nJennifer Mira Hashmall (argued) and Jeffrey B. White,\\nMiller Barondess LLP, Los Angeles, California, for\\nDefendant-Appellant/Cross-Appellee.\\n\\nMatthew C. Helland (argued) and Daniel S. Brome, Nichols\\nKaster LLP, San Francisco, California; Philip Bohrer,\\nBohrer Brady LLC, Baton Rouge, Louisiana; for Plaintiff-\\nAppellee/Cross-Appellants.\\n\\n\\n OPINION\\n\\nBENNETT, Circuit Judge:\\n\\n This case concerns whether a county is an arm of the\\nstate and thus entitled to Eleventh Amendment immunity\\nwhen it shares responsibility with the state for implementing\\na state-wide homecare program. We also consider the\\neffective date of regulations that (1) a district court vacated\\nbefore their original effective date; (2) an appellate court\\nupheld, reversing the district court; and (3) the agency then\\ndecided not to enforce until a date after the original effective\\ndate. We agree with the district court that the County of Los\\nAngeles is not entitled to Eleventh Amendment immunity\\nbut disagree as to the effective date of the regulations, which\\nwe hold is the original effective date of January 1, 2015. We\\nthus affirm in part, reverse in part, and remand.\\n\\n FACTS\\n\\n California’s In-Home Supportive Services program\\n(“IHSS program” or “the program”) provides in-home\\nsupportive services to eligible low-income elderly, blind, or\\ndisabled persons. Homecare providers help recipients with\\n\\f RAY V. COUNTY OF LOS ANGELES 5\\n\\ndaily activities like housework, meal preparation, and\\npersonal care. The program serves hundreds of thousands of\\nrecipients. In the County of Los Angeles alone there are\\nabout 170,000 homecare providers and more than 200,000\\nrecipients. California implements the program through\\nregulations promulgated by the California Department of\\nSocial Services (CDSS), and the program is administered in\\npart by California counties. Plaintiffs are current or former\\nLos Angeles IHSS homecare providers.\\n\\n The State and its counties share responsibility for\\nimplementing and running the IHSS program. The CDSS\\nensures that “in-home supportive services [are] provided in\\na uniform manner in every county,” Cal. Welf. & Inst. Code\\n§ 12301(a), and it must “adopt regulations establishing a\\nuniform range of services available to all eligible recipients\\nbased upon individual needs,” id. § 12301.1(a). The State\\nalso procures and implements a “Case Management\\nInformation and Payroll System.” Id. § 12317(b).\\n\\n But counties have some oversight of the IHSS program\\nas well. They, like the State, may terminate homecare\\nproviders. See id. § 12300.4(b)(5). And counties evaluate\\nrecipients and ensure quality compliance. See id. § 12301.1.\\nCounties also “ensure that services are provided to all\\neligible recipients.” Id. § 12302. Plaintiffs claim that\\nalthough they receive paychecks from the State, the County\\nis responsible for a “share” of their wages. For example, if\\na county imposes “any increase in provider wages or benefits\\n[that] is locally negotiated,” then “the county shall use\\ncounty-only funds” to fund that increase. Id. § 12306.1(a).\\nEach county also determines whether its providers may\\nexceed the maximum number of hours set by the CDSS. See\\nid. § 12300.4(d)(3).\\n\\f6 RAY V. COUNTY OF LOS ANGELES\\n\\n As employers of the homecare providers, the State and\\nCounty must comply with the Fair Labor Standards Act’s\\n(FLSA) overtime wage requirements. See 29 U.S.C.\\n§ 207(a)(1). But that wasn’t always the case.\\n\\n In 1974, Congress created a “companionship exemption”\\nto the FLSA for employees “employed in domestic service\\nemployment to provide companionship services for\\nindividuals who (because of age or infirmity) are unable to\\ncare for themselves.” See id. § 213 (a)(15); Fair Labor\\nStandards Amendments of 1974, Pub. L. No. 93-259,\\n88 Stat. 55. This exemption applied to homecare providers\\nlike Plaintiffs.\\n\\n In October 2013, however, the Department of Labor\\n(DOL) promulgated a new rule that changed the definition\\nof “companionship services” so that homecare providers like\\nPlaintiffs would be entitled to overtime pay under the FLSA.\\nSee Application of the Fair Labor Standards Act to Domestic\\nService, 78 Fed. Reg. 60,454, 60,454 (Oct. 1, 2013)\\n(codified at 29 C.F.R. pt. 552). The final rule had an\\neffective date of January 1, 2015. See id.\\n\\n Before the rule’s effective date, a group of “trade\\nassociations that represent businesses employing workers”\\nsubject to the FLSA exemption filed a lawsuit in the District\\nCourt for the District of Columbia. See Home Care Ass’n of\\nAm. v. Weil, 76 F. Supp. 3d 138, 142 (D.D.C. 2014) (Weil I).\\nThe plaintiffs claimed that the rule was arbitrary and\\ncapricious and thus sought to enjoin its implementation. Id.\\nat 139. At step one of its Chevron analysis, the district court\\nfound that Congress had “clearly spoken” on the issue. Id.\\nat 146. The district court then vacated the rule, id. at 148,\\nand the DOL appealed.\\n\\f RAY V. COUNTY OF LOS ANGELES 7\\n\\n On August 21, 2015, the D.C. Circuit reversed and\\nordered the district court to enter summary judgment for the\\nDOL. Home Care Ass’n of Am. v. Weil, 799 F.3d 1084, 1087\\n(D.C. Cir. 2015) (Weil II). Although the DOL prevailed, on\\nSeptember 14, 2015 it announced that it would “not bring\\nenforcement actions against any employer for violations of\\nFLSA obligations resulting from the amended domestic\\nservice regulations for 30 days after the date the mandate\\nissues.” 1 Application of the Fair Labor Standards Act to\\nDomestic Service; Announcement of 30-Day Period of Non-\\nEnforcement, 80 Fed. Reg. 55,029, 55,029 (Sept. 14, 2015)\\n(codified at 29 C.F.R. pt. 552). The Weil II mandate issued\\non October 13, 2015.\\n\\n On October 27, 2015, the DOL said that it would not\\nbegin enforcing the final rule until November 12, 2015.\\n\\n\\n 1\\n The DOL also stated:\\n\\n This 30-day non-enforcement policy does not replace\\n or affect the timeline of the Department’s existing\\n time-limited non-enforcement policy announced in\\n October 2014. 79 FR 60974. Under that policy,\\n through December 31, 2015, the Department will\\n exercise prosecutorial discretion in determining\\n whether to bring enforcement actions, with particular\\n consideration given to the extent to which States and\\n other entities have made good faith efforts to bring\\n their home care programs into compliance with the\\n FLSA since the promulgation of the Final Rule. The\\n Department will also continue to provide intensive\\n technical assistance to the regulated community, as it\\n has since promulgation of the Final Rule.\\n\\nApplication of the Fair Labor Standards Act to Domestic Service;\\nAnnouncement of 30-Day Period of Non-Enforcement, 80 Fed. Reg. at\\n55,029.\\n\\f8 RAY V. COUNTY OF LOS ANGELES\\n\\nAnd, echoing its September 14, 2015 statement, the DOL\\nagain said that\\n\\n from November 12, 2015 through December\\n 31, 2015, [it would] exercise prosecutorial\\n discretion in determining whether to bring\\n enforcement actions, with particular\\n consideration given to the extent to which\\n States and other entities have made good faith\\n efforts to bring their home care programs into\\n compliance with the FLSA since the\\n promulgation of the Final Rule.\\n\\nApplication of the Fair Labor Standards Act to Domestic\\nService; Dates of Previously Announced 30-Day Period of\\nNon-Enforcement, 80 Fed. Reg. 65,646, 65,646 (Oct. 27,\\n2015) (codified at 29 C.F.R. pt. 552).\\n\\n Before the Weil I decision, California (through the\\nCDSS) began taking steps to “meet the January 1, 2015,\\nimplementation date,” including modifying its systems to\\n“process and calculate overtime compensation.” But after\\nthe Weil I decision, the CDSS decided that it would not\\nimplement overtime payments “until further notice.” After\\nWeil II, the CDSS again said that it would comply with the\\novertime requirements—but not until February 1, 2016.\\n\\n In June 2017, Ray filed a putative collective action, 2\\nunder Section 216(b) of the FLSA, against the State of\\nCalifornia and the County of Los Angeles. Ray’s complaint\\nsought relief for herself and the putative collective for\\n\\n 2\\n Collective actions are provided for in the FLSA and are different\\nfrom class actions, see Campbell v. City of L.A., 903 F.3d 1090, 1101\\n(9th Cir. 2018), but the differences are not relevant to this appeal.\\n\\f RAY V. COUNTY OF LOS ANGELES 9\\n\\nunpaid overtime wages between January 1, 2015—the rule’s\\noriginal effective date—and February 1, 2016, the date on\\nwhich the State began paying overtime wages.\\n\\n As relevant here, the County moved to dismiss the\\ncomplaint on Eleventh Amendment immunity grounds. 3 In\\nthe alternative, the County moved to strike all references in\\nthe complaint to overtime wages allegedly earned before\\nOctober 13, 2015—the date on which the mandate issued in\\nWeil II.\\n\\n The district court first held that the County had no\\nEleventh Amendment immunity. The district court noted\\nthat the Supreme Court has long refused to grant Eleventh\\nAmendment immunity to counties and that the Court has\\nalready held that California counties are not arms of the\\nState. The district court then assumed arguendo that a\\ncounty could be an arm of the State under the five-factor test\\nthat we set out in Mitchell v. Los Angeles Community\\nCollege District, 861 F.2d 198 (9th Cir. 1988) for\\ndetermining whether an entity is an arm of the state for\\npurposes of Eleventh Amendment immunity. The district\\ncourt found that only one of the five factors favored the\\nCounty, and thus it held that the County enjoyed no Eleventh\\nAmendment immunity.\\n\\n The district court then “reject[ed] Plaintiffs’ efforts to\\nenforce the FLSA companionship exemption regulations\\nretroactively to January 1, 2015.” Instead, it held “that the\\nputative collective period extends from November 12, 2015,\\nthrough January 31, 2016,” and not before. The court said\\n\\n 3\\n Early on, Ray voluntarily dismissed the CDSS as a defendant, and\\nPlaintiffs did not name the State as a defendant in the now-operative\\ncomplaint.\\n\\f10 RAY V. COUNTY OF LOS ANGELES\\n\\nthat although the Weil II decision applied retroactively, that\\ndecision was merely that the DOL could amend the FLSA\\nand that those amendments were not arbitrary and\\ncapricious. This, the district court held, differed from “the\\nretroactive application of the amended regulations\\nthemselves.” The district court reasoned:\\n\\n The rule of law announced by the D.C.\\n Circuit is given retroactive effect by allowing\\n DOL to reinstate those regulations without\\n having to begin a new rule-making process.\\n That is not the same thing as reinstating an\\n earlier and judicially vacated effective date\\n and retroactively creating liability for\\n violations of the reinstated regulations as if\\n the District Court’s vacation of the\\n regulations had never occurred.\\n\\n The district court also found it “compelling” that both the\\nD.C. Circuit and the DOL “intended” that the regulation\\nbecome effective “no earlier than November 12, 2015.” As\\nevidence of this intent, the district court pointed to the\\nDOL’s decision not to enforce the new regulations before\\nthat date.\\n\\n Finally, the district court found that its holding was\\nconsistent “with the general rule that a private right of action\\nshould ordinarily not exist when the applicable rule could\\nnot be enforced by the relevant enforcement agency.”\\n\\n The County filed an interlocutory appeal as to the denial\\nof Eleventh Amendment immunity. The district court\\ngranted Plaintiffs’ motion to certify for interlocutory appeal\\nthe district court’s holding that the putative collective period\\nbegan on November 12, 2015, and we granted Plaintiffs’\\nrequest to appeal that holding.\\n\\f RAY V. COUNTY OF LOS ANGELES 11\\n\\n DISCUSSION\\n\\n We review de novo the denial of Eleventh Amendment\\nimmunity. Cal. ex rel. Lockyer v. Dynegy, Inc., 375 F.3d\\n831, 843 n.12 (9th Cir. 2004). We construe the motion to\\nstrike as a motion to dismiss in part, and thus we review the\\neffective date holding de novo because it essentially\\ndismissed Plaintiffs’ overtime claims for the period between\\nJanuary 1, 2015 and November 12, 2015. See Yamaguchi v.\\nU.S. Dep’t of the Air Force, 109 F.3d 1475, 1482 (9th Cir.\\n1997).\\n\\n The County is not entitled to Eleventh Amendment\\n immunity.\\n\\n Plaintiffs first argue that Eleventh Amendment\\nimmunity is never available to counties. The County argues\\nthat it enjoys Eleventh Amendment immunity when acting\\nas an “arm of the State.”\\n\\n Federal courts have long declined to extend Eleventh\\nAmendment immunity to counties. 4 Indeed, the Supreme\\n\\n 4\\n See, e.g., Lake Country Estates, Inc. v. Tahoe Reg’l Planning\\nAgency, 440 U.S. 391, 401 (1979) (“[T]he Court has consistently refused\\nto construe the Amendment to afford protection to political subdivisions\\nsuch as counties and municipalities, even though such entities exercise a\\nslice of state power.” (internal quotation marks omitted)); Lincoln Cty.\\nv. Luning, 133 U.S. 529, 530 (1890) (holding that the Eleventh\\nAmendment does not bar a suit against a county, though the principle\\nadvanced has changed over time); Del Campo v. Kennedy, 517 F.3d\\n1070, 1075–76 (9th Cir. 2008) (“State sovereign immunity . . . does not\\nextend to counties and similar municipal corporations, even though they\\nshare some portion of state power.” (internal quotation marks omitted)\\n(quoting Mt. Healthy City Sch. Dist. Bd. of Educ. v. Doyle, 429 U.S. 274,\\n280 (1977))).\\n\\f12 RAY V. COUNTY OF LOS ANGELES\\n\\nCourt once said that Eleventh Amendment immunity does\\nnot extend to municipal corporations. Mt. Healthy, 429 U.S.\\nat 280. But thirty years later, the Supreme Court suggested\\nthat it was at least possible for a county to receive Eleventh\\nAmendment immunity. In Northern Insurance Company of\\nNew York v. Chatham County, 547 U.S. 189, 190 (2006),\\nwhich involved a county-operated drawbridge, the Court\\nstated that a county might be entitled to Eleventh\\nAmendment immunity if it were “acting as an arm of the\\nState, as delineated by this Court’s precedents, in operating\\nthe drawbridge.” 5\\n\\n The Court cited several cases for this proposition. First,\\nAlden v. Maine: “The second important limit to the principle\\nof sovereign immunity is that it bars suits against States but\\nnot lesser entities. The immunity does not extend to suits\\nprosecuted against a municipal corporation or other\\ngovernmental entity which is not an arm of the State.”\\n527 U.S. 706, 756 (1999). This sentence means one of two\\nthings: either (1) that Eleventh Amendment immunity does\\nnot extend to municipal corporations because they are not\\narms of the state or (2) that Eleventh Amendment immunity\\ndoes not extend to a municipal corporation unless it is acting,\\nin a particular circumstance, as an arm of the state. Alden in\\nturn cites Mt. Healthy, in which the Court considered\\nwhether “the Mt. Healthy Board of Education is to be treated\\nas an arm of the State partaking of the State’s Eleventh\\n\\n 5\\n At least one circuit has relied on this language and held that\\ncounties might be entitled to Eleventh Amendment immunity. See\\nFuesting v. Lafayette Par. Bayou Vermilion Dist., 470 F.3d 576, 579 (5th\\nCir. 2006) (“[A] municipality can be immune from suit if it was ‘acting\\nas an arm of the State, as delineated by [the Supreme] Court’s\\nprecedent’” (alteration in original) (quoting Chatham, 547 U.S. at 194)).\\nBut, to our knowledge, no court has ever actually extended Eleventh\\nAmendment immunity to a county.\\n\\f RAY V. COUNTY OF LOS ANGELES 13\\n\\nAmendment immunity, or is instead to be treated as a\\nmunicipal corporation or other political subdivision to which\\nthe Eleventh Amendment does not extend.” Mt. Healthy,\\n429 U.S. at 280. That citation suggests the former reading.\\n\\n The Chatham Court also cited Lake Country Estates, but\\nwhile that case noted that “some agencies exercising state\\npower have been permitted to invoke the [Eleventh]\\nAmendment in order to protect the state treasury from\\nliability that would have had essentially the same practical\\nconsequences as a judgment against the State itself,” it also\\nstated that “the Court has consistently refused to construe the\\nAmendment to afford protection to political subdivisions\\nsuch as counties and municipalities, even though such\\nentities exercise a ‘slice of state power.’” Lake Country\\nEstates, 440 U.S. at 400–01. Although these passages seem\\nto support Plaintiffs’ argument that counties never enjoy\\nEleventh Amendment immunity, it is not for us to clarify\\nChatham’s apparently contrary statement.\\n\\n The Chatham Court ultimately found it dispositive that\\nthe County there had conceded below that it had no Eleventh\\nAmendment immunity and that the question on which\\ncertiorari was granted assumed that conclusion. Given that\\nthe Supreme Court appears to have left open the possibility\\nthat a county could be entitled to Eleventh Amendment\\nimmunity in some cases, we decline to hold to the contrary.\\nWe therefore assume without deciding that, consistent with\\nthe Court’s language in Chatham, a county might be entitled\\nto Eleventh Amendment immunity if acting as an arm of the\\nstate.\\n\\n The County is not an arm of the State here.\\n\\n In Mitchell, we set out five factors for determining\\nwhether a government entity is an arm of its state for\\n\\f14 RAY V. COUNTY OF LOS ANGELES\\n\\nEleventh Amendment immunity purposes: (1) “whether a\\nmoney judgment would be satisfied out of state funds”;\\n(2) “whether the entity performs central governmental\\nfunctions”; (3) “whether the entity may sue or be sued”;\\n(4) “whether the entity has the power to take property in its\\nown name or only the name of the state”; and (5) “the\\ncorporate status of the entity.” 861 F.2d at 201. “To\\ndetermine these factors, the court looks to the way state law\\ntreats the entity.” Id.\\n\\n a. First Mitchell factor\\n\\n “The first Mitchell factor—whether a money judgment\\n. . . would be satisfied out of state funds—is the most\\nimportant.” Sato v. Orange Cty. Dep’t of Educ., 861 F.3d\\n923, 929 (9th Cir. 2017); see also Beentjes v. Placer Cty. Air\\nPollution Control Dist., 397 F.3d 775, 785 (9th Cir. 2005)\\n(noting that the first Mitchell factor is “the one given the\\nmost weight”). The County conceded, both below and on\\nappeal, that it cannot show that a money judgment would be\\npaid directly with State funds. 6 Thus, this factor weighs\\nagainst Eleventh Amendment immunity.\\n\\n b. Second Mitchell factor\\n\\n As to the second Mitchell factor—whether the County\\nperforms central governmental functions—we must\\ndetermine whether the County addresses “a matter of\\nstatewide rather than local or municipal concern, and the\\nextent to which the state exercises centralized governmental\\ncontrol over the entity.” Beentjes, 397 F.3d at 782 (internal\\n 6\\n The parties discuss at length how the County and the State allocate\\nthe costs of the program, but that is not relevant—what matters is who\\nwould be responsible for satisfying a money judgment against the\\nCounty, not who pays for the program.\\n\\f RAY V. COUNTY OF LOS ANGELES 15\\n\\nquotation marks omitted) (first quoting Belanger v. Madera\\nUnified Sch. Dist., 963 F.2d 248, 253 (9th Cir. 1992); then\\nquoting Savage v. Glendale Union High Sch., Dist. No. 205,\\nMaricopa Cty., 343 F.3d 1036, 1044 (9th Cir. 2003)).\\n\\n To begin, it is unclear whether the second Mitchell factor\\nconcerns whether the County performs central government\\nfunctions in general or whether the County performs central\\ngovernment functions in carrying out the particular function\\nat issue—here implementing the IHSS program.\\n\\n As the district court correctly noted, the closest analogue\\nin our case law is Streit v. County of Los Angeles, 236 F.3d\\n552 (9th Cir. 2001). There, the Los Angeles County\\nSheriff’s Department (LASD) would check its systems,\\nbefore releasing a prisoner, to see if the prisoner was wanted\\nby another law enforcement agency. Id. at 556. This\\nextended the period of incarceration one or two days past the\\nprisoners’ release dates. Id. The plaintiffs alleged that the\\nCounty delayed their release during these checks, in\\nviolation of their civil rights. Id. The LASD argued that\\nbecause it was an arm of the state, it was not a “person” that\\ncould be liable for damages under § 1983. Id. at 557.\\n\\n We looked at the LASD’s performance of the particular\\nfunction at issue—implementing the pre-release policy—not\\nthe LASD’s general function as a sheriff’s department. See\\nid. at 567. We held that “conducting the AJIS checks is not\\na central government function.” Id. (emphasis added).\\nThus, it appears from Streit that we look to whether the\\nCounty, in performing the particular function at issue,\\nperforms a central government function. This fits with the\\nCourt’s statement in Chatham that the county there might\\nhave been entitled to Eleventh Amendment immunity if it\\nwere “acting as an arm of the State, as delineated by this\\n\\f16 RAY V. COUNTY OF LOS ANGELES\\n\\nCourt’s precedents, in operating [a] drawbridge.”\\nChatham, 547 U.S. at 194 (emphasis added).\\n\\n i. A matter of statewide rather than local or\\n municipal concern\\n\\n The in-home care of the elderly and disabled is a matter\\nof both statewide and local concern. Plaintiffs are residents\\nof California, and the IHSS program is a statewide program\\nimplemented through State legislation that provides care to\\nhundreds of thousands of California residents. But Plaintiffs\\nare also, of course, residents of Los Angeles County, and the\\nCounty has an interest in the program and the care provided\\nin Los Angeles.\\n\\n ii. The extent to which the state exercises\\n centralized governmental control over the\\n entity\\n\\n Here we consider the extent to which the County, in\\nimplementing the program, has “discretionary powers” and\\n“substantial autonomy in carrying out [its] duties.” Beentjes,\\n397 F.3d at 783.\\n\\n The County may negotiate, implement, and pay for pay\\nraises. See Cal. Welf. & Inst. Code § 12306.1. The County\\nmay also allow its providers to exceed the maximum number\\nof hours that the CDSS has set. See id. § 12300.4(d)(3).\\nThus, the County has discretion to make some important\\nchoices on its own.\\n\\n But the County contends—and Plaintiffs do not\\ndispute—that it has no discretion over the action (or\\ninaction) that subjected it to potential liability here: payment\\nof overtime wages under the FLSA. In taking the actions\\nthat have subjected it to potential liability, the County had\\n\\f RAY V. COUNTY OF LOS ANGELES 17\\n\\nneither “discretionary powers” nor “substantial autonomy”\\nin carrying out its duties.\\n\\n We think this clearly tips the scales in the County’s favor\\nas to this factor. The County had no choice in the matter of\\nthe overtime wages, as the State mandated the payment start\\ndate. We therefore hold that the second Mitchell factor\\nfavors Eleventh Amendment immunity.\\n\\n c. Third, fourth, and fifth Mitchell factors\\n\\n The County does not dispute that it can sue and be sued\\n(third Mitchell factor), that it has the power to take property\\nin its own name (fourth Mitchell factor), or that it has an\\nindependent corporate status 7 separate from the State (fifth\\nMitchell factor). Thus, these three Mitchell factors weigh\\nagainst Eleventh Amendment immunity.\\n\\n * * *\\n\\n In sum, the first Mitchell factor is the most important,\\nand it weighs against Eleventh Amendment immunity. So\\ndo the third, fourth, and fifth Mitchell factors. Only the\\nsecond factor favors immunity. We therefore hold that,\\nunder Mitchell, the County is not an arm of the State when it\\n\\n\\n\\n\\n 7\\n The fifth Mitchell factor asks whether the entity has “independent\\ncorporate status,” Holz v. Nenana City Pub. Sch. Dist., 347 F.3d 1176,\\n1188 (9th Cir. 2003), or is, instead, merely an agency of the state without\\nan identity that is separate from the state, Beentjes, 397 F.3d at 785. Here\\nthe County does not dispute its independent corporate status, as the\\nSupreme Court has already held that California counties have\\nindependent corporate status and are not agents of the State of California.\\nSee Moor v. Alameda Cty., 411 U.S. 693, 719 (1973).\\n\\f18 RAY V. COUNTY OF LOS ANGELES\\n\\nadministers the IHSS program, and thus it has no Eleventh\\nAmendment immunity barring this action.\\n\\n The Supreme Court has not overruled or\\n undermined Mitchell.\\n\\n The County argues that we should overrule Mitchell\\nbecause a later Supreme Court case, Hess v. Port Authority\\nTrans-Hudson Corporation, 513 U.S. 30 (1994),\\nundermined it. As a three-judge panel, if we find that\\nintervening Supreme Court authority is clearly\\nirreconcilable with our own precedent, we must consider\\nourselves bound by the intervening higher authority and\\nconsider our precedent effectively overruled. See Miller v.\\nGammie, 335 F.3d 889, 900 (9th Cir. 2003). Because Hess\\nis not clearly irreconcilable with Mitchell, we reject the\\nCounty’s argument.\\n\\n In Hess, the Court held that a Congressionally approved\\nbistate entity—the Port Authority Trans-Hudson\\nCorporation (PATH), created to improve coordination of the\\n“terminal, transportation and other facilities of commerce in,\\nabout and through the port of New York”—did not have\\nEleventh Amendment immunity. 513 U.S at 35, 52–53\\n(citation omitted). The County argues that Hess established\\n“indicators of immunity” that undermine the Mitchell test.\\nWe disagree.\\n\\n The Hess Court noted that “current Eleventh\\nAmendment jurisprudence emphasizes the integrity retained\\nby each State in our federal system.” Id. at 39. The Court\\nthen emphasized the difference between PATH and the\\nStates of the Union: “The States, as separate sovereigns, are\\nthe constituent elements of the Union. Bistate entities, in\\ncontrast, typically are creations of three discrete sovereigns:\\ntwo States and the Federal Government.” Id. at 40.\\n\\f RAY V. COUNTY OF LOS ANGELES 19\\n\\n The Court stated that “[p]ointing away from Eleventh\\nAmendment immunity, the States lack financial\\nresponsibility” for the bistate entity. Id. at 45. Here,\\nCalifornia similarly lacks financial responsibility for the\\nCounty generally, but Plaintiffs allege that although\\nCalifornia writes their checks, the County pays a share of\\ntheir wages and sets their hours of work.\\n\\n In Hess, “indicators of immunity point[ed] in different\\ndirections.” Id. at 47. Perhaps they do here as well. Los\\nAngeles is not a constituent member of the Union, but it\\nacted at the direction of the State and had no authority over\\nthe payments at issue. But when faced with a different\\ndichotomy in Hess, the Court emphasized that the most\\nimportant factor was whether judgments against PATH\\nwould be paid by the State: “the vulnerability of the State’s\\npurse [is] the most salient factor in Eleventh Amendment\\ndeterminations.” Id. at 48; see also id. at 48–49 (citing cases\\nfor the “prevailing view” that the state-treasury factor is\\n“generally accorded . . . dispositive weight”); id. at 51\\n(stating that “the Eleventh Amendment’s core concern is not\\nimplicated” if the State is not “in fact obligated to bear and\\npay the . . . indebtedness of the enterprise”). 8\\n\\n\\n\\n\\n 8\\n The dissent read the holding even more broadly:\\n\\n In place of the various factors recognized in Lake\\n Country Estates, Inc. v. Tahoe Regional Planning\\n Agency, 440 U.S. 391, 99 S. Ct. 1171, 59 L.Ed.2d 401\\n (1979), for determining arm-of-the-state status, we\\n may now substitute a single overriding criterion,\\n vulnerability of the state treasury. If a State does not\\n fund judgments against an entity, that entity is not\\n\\f20 RAY V. COUNTY OF LOS ANGELES\\n\\n After noting that the bistate entity “was financially self-\\nsufficient,” generated “its own revenues,” and paid “its own\\ndebts,” the Court held that “[r]equiring the [bistate entity] to\\nanswer in federal court to injured railroad workers who\\nassert a federal statutory right, under the FELA, to recover\\ndamages does not touch the concerns—the States’ solvency\\nand dignity—that underpin the Eleventh Amendment.” Id.\\nat 52. The same is true here. Mitchell and Hess both\\nemphasize the state-treasury factor. Hess thus fully supports\\nand does not undermine Mitchell. 9\\n\\n The County argues that Hess emphasized the amount of\\ncontrol that a state maintains over an entity, a factor\\nsupposedly not mentioned in Mitchell and one that,\\naccording to the County, favors Eleventh Amendment\\nimmunity here. First, as we mentioned above, the second\\nMitchell factor does include a “control” inquiry—it just\\ndoesn’t make that factor dispositive. In addition, Hess\\npointed out that “[g]auging actual control . . . can be a\\n‘perilous inquiry,’ [and] ‘an uncertain and unreliable\\nexercise.’” 513 U.S. at 47 (quoting Note, 92 Colum. L. Rev.\\n1243, 1284 (1992)). The Court therefore doubted not only\\nthe efficacy but also the utility of a “control” analysis, and it\\n\\n\\n\\n within the ambit of the Eleventh Amendment, and\\n suits in federal court may proceed unimpeded.\\n\\nId. at 55 (O’Connor, J., dissenting).\\n 9\\n Los Angeles makes a legitimate point about the unfairness of the\\nresult here. But that unfairness springs from the State and its\\nimplementing legislation, not the Eleventh Amendment. Los Angeles\\nmust air its grievance, if at all, in Sacramento.\\n\\f RAY V. COUNTY OF LOS ANGELES 21\\n\\ndid not suggest that control was a favored, much less\\ndispositive, factor in the Eleventh Amendment analysis. 10\\n\\n Hess clearly stated that “rendering control dispositive\\ndoes not home in on the impetus for the Eleventh\\nAmendment: the prevention of federal-court judgments that\\nmust be paid out of a State’s treasury.” Id. at 48. And, in\\nspecifically discussing the control factor, the Court noted\\nthat even though “‘political subdivisions exist solely at the\\nwhim and behest of their State,’ . . . cities and counties do\\nnot enjoy Eleventh Amendment immunity.” Id. at 47\\n(quoting Port Auth. Trans-Hudson Corp. v. Feeney, 495 U.S.\\n299, 313 (1990)).\\n\\n Finally, the County insists that Hess compels us to\\nconsider the State’s dignity, a factor not mentioned in\\nMitchell. Hess noted that the State’s “solvency and dignity\\n. . . underpin the Eleventh Amendment.” 513 U.S. at 52.\\nThat is undoubtedly true. But the State is no longer a party\\nto this action, and it will not be responsible for an adverse\\njudgment against the County. Allowing this action against\\nLos Angeles does not injure California’s dignity. 11\\n\\n\\n\\n 10\\n The control discussed in Hess seems to have gone to overall\\ncontrol over the entity, not just control within the context of the particular\\nfunction at issue: “PATH urges that we find good reason to classify the\\nPort Authority as a state agency for Eleventh Amendment purposes\\nbased on the control New York and New Jersey wield over the\\nAuthority. . . . But ultimate control of every state-created entity resides\\nwith the State, for the State may destroy or reshape any unit it creates.”\\nId. at 47. Thus, looking at the State’s overall control over the County as\\na county would not help the County’s position here.\\n 11\\n And, although it would not have altered our analysis, we note that\\nCalifornia has not sought to file an amicus brief (below or on appeal)\\n\\f22 RAY V. COUNTY OF LOS ANGELES\\n\\n The Supreme Court decided Hess about five years after\\nwe decided Mitchell. And although Hess arose in a different\\ncontext than Mitchell-Hess addressed a bistate entity, not a\\ncounty—nothing in Hess so undermines Mitchell that we\\nhave the power to overrule it. More importantly, even if we\\nused Hess rather than Mitchell to guide our analysis, we\\nwould reach the same result.\\n\\n When a non-state entity invokes Eleventh Amendment\\nimmunity, the most important factor for determining\\nwhether the entity is an arm of the state remains the state-\\ntreasury factor—that is, whether the state will be liable for a\\nmoney judgment against the non-state entity. That factor,\\nand all but one of the other Mitchell factors, dictates the\\nresult here. The Eleventh Amendment does not bar\\nPlaintiffs’ suit against Los Angeles.\\n\\n The effective date of the rule is January 1, 2015.\\n\\n We next consider whether the effective date of the rule\\nis the original effective date of January 1, 2015 or some date\\nafter the D.C. Circuit reversed the district court’s vacatur.\\nThe County argues that the rule cannot have an effective date\\nthat is earlier than the date on which the D.C. Circuit\\nreversed the district court’s vacatur. Plaintiffs argue that the\\nlegal effect of the vacatur is to reinstate the original January\\n1, 2015 effective date. We agree with Plaintiffs and hold\\nthat the effective date of the rule is January 1, 2015.\\n\\n\\n\\n\\narguing either that the County is entitled to Eleventh Amendment\\nimmunity or that this case threatens California’s dignity.\\n\\f RAY V. COUNTY OF LOS ANGELES 23\\n\\n A January 1, 2015 effective date is not\\n impermissibly retroactive.\\n\\n The County argues that a January 1, 2015 effective date\\nis impermissibly retroactive. Plaintiffs argue that the D.C.\\nCircuit’s decision, not the rule, applies retroactively, because\\nthe D.C. Circuit was “explaining what the law always was,”\\nand thus reinstating the original effective date is merely a\\nreturn to the status quo ante.\\n\\n When an appellate court applies “a rule of federal law to\\nthe parties before it,” that interpretation “must be given full\\nretroactive effect in all cases still open on direct review and\\nas to all events, regardless of whether such events predate or\\npostdate [the] announcement of the rule.” Harper v. Va.\\nDep’t of Taxation, 509 U.S. 86, 97 (1993). That is because\\n“when a court delivers a ruling, even if it is unforeseen, the\\nlaw has not changed. Rather, the court is explaining what\\nthe law always was.” Jones Stevedoring Co. v. Dir., Office\\nof Workers’ Comp. Programs, 133 F.3d 683, 688 (9th Cir.\\n1997).\\n\\n When the D.C. Circuit held that the DOL had the\\nrulemaking authority to promulgate the new rule and that its\\nnew rule was a reasonable exercise of that authority, see Weil\\nII, 799 F.3d at 1090, it did not change the law but merely\\nexplained what the law always was—the district court’s\\nerroneous contrary holding notwithstanding.\\n\\n Two cases support our holding. In GTE South, Inc. v.\\nMorrison, 199 F.3d 733, 738, 740 (4th Cir. 1999), the Fourth\\nCircuit addressed an issue much like the one we face:\\ndetermining the effective date of certain pricing rules,\\npromulgated by the FCC, that the Eighth Circuit stayed and\\nthen vacated before their effective date. The Supreme Court\\nlater reversed the Eighth Circuit. See AT & T Corp. v. Iowa\\n\\f24 RAY V. COUNTY OF LOS ANGELES\\n\\nUtilities Bd., 525 U.S. 366, 385 (1999). The Morrison panel\\nheld that “the Supreme Court’s determination that the FCC\\nhas jurisdiction to issue pricing rules would appear to\\ncompel the conclusion that the FCC always had such\\njurisdiction and that the rules apply as of the effective date\\noriginally scheduled.” 199 F.3d at 740 (emphasis added).\\nThe Fourth Circuit emphasized that its holding was not\\nunfair to the parties who argued for a later effective date\\nbecause they had “ample notice” of the original effective\\ndate and “surely knew that the FCC’s authority to issue\\npricing rules might ultimately be upheld by the Supreme\\nCourt.” Id. at 741.\\n\\n In US West Communication, Inc. v. Jennings, 304 F.3d\\n950, 955 (9th Cir. 2002), we considered a similar question:\\nwhether the regulations that the Fourth Circuit considered in\\nMorrison applied to conduct that occurred during the period\\nof vacatur. Finding the Fourth Circuit’s reasoning in\\nMorrison persuasive and applicable, we noted that the\\nSupreme Court’s determination that the regulations were\\nvalid meant that we should apply them “to all . . . agreements\\narbitrated under the Act, including agreements arbitrated\\nbefore the rules were reinstated.” Id. at 957 (emphasis\\nadded). Relying on Morrison, we held that applying the\\nreinstated regulations to conduct that occurred during the\\nperiod of vacatur would not give the regulations an\\nimpermissible retroactive effect. Id. at 958.\\n\\n Morrison and Jennings are analogous to this case\\nbecause both involved determining how to apply rules or\\nregulations that were vacated but ultimately reinstated on\\nappeal. Indeed, Morrison commented not only on the\\nretroactivity of the Supreme Court’s reversal but also on the\\neffective date of the regulations, holding that the intervening\\n\\f RAY V. COUNTY OF LOS ANGELES 25\\n\\nvacatur did not alter the original effective date of the pricing\\nrules. 199 F.3d at 740.\\n\\n Thus, Morrison and Jennings guide our analysis here.\\nThe D.C. Circuit’s holding that the DOL had the authority to\\npromulgate the new rule and that the rule was reasonable\\napplies retroactively. As in Jennings, the regulations apply\\nas of the original effective date. To hold otherwise could\\nencourage dilatory appellate litigation. If an erroneously\\nvacated rule or regulation were not effective until sometime\\nafter the mandate issued in a later appeal, then a party might\\ndrag out the appellate process to avoid compliance for as\\nlong as possible. Put differently, an erroneous vacatur\\ncannot postpone a rule’s effective date until an appellate\\ncourt corrects the error sometime in the future. And, as the\\nFourth Circuit noted in Morrison, in a case like this everyone\\nknows that the lower court decision might be reversed on\\nappeal.\\n\\n The State and its counties knew from October 13, 2013,\\nwhen the DOL first announced its final rule, that January 1,\\n2015 was the rule’s effective date. See Application of the\\nFair Labor Standards Act to Domestic Service, 78 Fed. Reg.\\nat 60,454. The State and its counties had a full fifteen\\nmonths to comply with the final rule—indeed the State\\ninitially said that it would comply with the original effective\\ndate, but it changed course after the Weil I court vacated the\\nrule. That decision may have been reasonable, but it created\\na monetary risk, as the State and its counties were well aware\\nthat an appellate court might uphold the regulations on\\nappeal.\\n\\n The district court held that to apply the Weil II decision\\nretroactively would be to “reinstate[] an earlier and judicially\\nvacated effective date and retroactively creat[e] liability for\\nviolations of the reinstated regulations as if the District\\n\\f26 RAY V. COUNTY OF LOS ANGELES\\n\\nCourt’s vacation of the regulations had never occurred.”\\nThat is exactly correct. And although the district court found\\nthat to be unfair, it would be equally unfair to hold that a\\nputative collective of homecare providers is not entitled to\\nnearly a year’s worth of overtime wages just because a single\\ndistrict court issued an erroneous decision that another court\\nreversed on appeal. The State gambled that Weil I would be\\naffirmed. The effect of that gamble might be unfair to the\\nCounty, but the County must seek any recourse from the\\nState. It is not fair for the homecare providers to bear the\\nfinancial consequences of the State’s calculated risk.\\n\\n The DOL’s decision not to enforce a new rule does\\n not obviate private rights of action.\\n\\n According to the County, the DOL’s choice against\\nenforcing the rule until November 12, 2015 eliminated the\\navailability of private rights of action until that date because\\na private right of action cannot precede an agency’s\\nenforcement of a rule or regulation. We disagree.\\n\\n “An agency’s informal assurance that it will not pursue\\nenforcement cannot preclude a citizen’s suit to do so.” Ohio\\nValley Envtl. Coal. v. Fola Coal Co., LLC, 845 F.3d 133,\\n145 (4th Cir. 2017) (emphasis added). Congress created a\\nprivate right of action under the FLSA for unpaid overtime:\\n“Any employer who violates the provisions of section 206\\nor section 207 of this title shall be liable to the employee or\\nemployees affected in the amount of their . . . unpaid\\novertime compensation . . . .” 29 U.S.C. § 216(b). An\\nagency’s discretionary decision to hold off enforcement does\\nnot and cannot strip private parties of their rights to do so.\\nSee Ohio Valley, 845 F.3d at 145 (“Congress enacted the\\ncitizen suit provision of the Clean Water Act to address\\nsituations, like the one at hand, in which the traditional\\nenforcement agency declines to act.”).\\n\\f RAY V. COUNTY OF LOS ANGELES 27\\n\\n The district court’s hypothesis that the D.C. Circuit and\\nDOL “intended” that the regulation become effective “no\\nearlier than November 12, 2015” is tenuous and, in any\\nevent, irrelevant. First, the D.C. Circuit said nothing at all\\non the issue. Second, there is nothing in the several\\nstatements of the DOL, which the district court relied on, that\\nsuggest that it intended its discretionary enforcement choices\\nto preclude private enforcement. Indeed, other than by\\namending the rule, the DOL could not have precluded\\nprivate enforcement even if it wanted to.\\n\\n The rule’s original effective date remains January 1,\\n2015. If the DOL “intended” for the effective date be\\nsomething other than January 1, 2015, the DOL could have\\nsought to change that effective date through the procedures\\nset out in the Administrative Procedure Act. Were we to\\nhold to the contrary and impose our view that the DOL’s\\nexercise of discretion amended the effective date sub\\nsilentio, we would in fact be usurping the rulemaking\\nauthority of the DOL. See Nat. Res. Def. Council, Inc. v.\\nU.S. E.P.A., 683 F.2d 752, 762 (3d Cir. 1982) (holding that\\na final rule’s effective date is an “essential part” of that rule\\nand is thus subject to the rulemaking procedures of the\\nAPA).\\n\\n The effective date of the rule is January 1, 2015. 12\\n\\n\\n 12\\n Although some district courts have reached a different\\nconclusion—see, e.g., Bangoy v. Total Homecare Solutions, LLC, No.\\n1:15-CV-573, 2015 WL 12672727, at *3 (S.D. Ohio Dec. 21, 2015)\\n(holding that the plaintiffs failed to state a claim for a violation of the\\nFLSA between January 1, 2015 and “late August 2015”)—nearly all of\\nthem have reached the same result we reach here, see, e.g., Kinkead v.\\nHumana, Inc., 206 F. Supp. 3d 751, 752 (D. Conn. 2016) (holding that\\nthe effective date of the rule is January 1, 2015, “the effective date set\\n\\f28 RAY V. COUNTY OF LOS ANGELES\\n\\n CONCLUSION\\n\\n We AFFIRM the district court’s holding that the County\\nis not entitled to Eleventh Amendment immunity and\\nREVERSE the district court’s holding that the putative\\ncollective period began on November 12, 2015, holding\\ninstead that the rule’s effective date—and thus the beginning\\nof the putative collective period—is January 1, 2015. We\\nREMAND for proceedings consistent with this opinion.\\nCosts shall be awarded to Plaintiffs-Appellants.\\n\\n\\n\\n\\nforth by the agency”); Collins v. DKL Ventures, LLC, 215 F. Supp. 3d\\n1059 (D. Colo. 2016) (same); Lewis-Ramsey v. Evangelical Lutheran\\nGood Samaritan Soc’y, 215 F. Supp. 3d 805 (S.D. Iowa 2016) (same).\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432212/\", \"author_raw\": \"BENNETT, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"KIM MCLANE WARDLAW"},"judge2":{"kind":"string","value":"MARK J BENNETT"},"judge3":{"kind":"string","value":"KATHLEEN CARDONE"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4654959/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Trina Ray v. County of Los Angeles\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-22\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nTRINA RAY, individually, and on No. 17-56581\nbehalf of others similarly situated,\n Plaintiff-Appellee, D.C. No.\n 2:17-cv-04239-\n v. PA-SK\n\nCOUNTY OF LOS ANGELES,\n Defendant-Appellant.\n\n\nTRINA RAY; SASHA WALKER, No. 18-55276\nindividually, and on behalf of all\nothers similarly situated, D.C. No.\n Plaintiffs-Appellants, 2:17-cv-04239-\n PA-SK\n v.\n\nLOS ANGELES COUNTY DEPARTMENT OPINION\nOF PUBLIC SOCIAL SERVICES,\nErroneously Sued As County of Los\nAngeles,\n Defendant-Appellee.\n\n Appeal from the United States District Court\n for the Central District of California\n Percy Anderson, District Judge, Presiding\n\f2 RAY V. COUNTY OF LOS ANGELES\n\n Argued and Submitted March 7, 2019\n Pasadena, California\n\n Filed August 22, 2019\n\n Before: Kim McLane Wardlaw and Mark J. Bennett,\n Circuit Judges, and Kathleen Cardone, * District Judge.\n\n Opinion by Judge Bennett\n\n\n SUMMARY **\n\n\n Labor Law / Eleventh Amendment Immunity\n\n The panel affirmed the district court’s order denying a\ndefendant county’s motion to dismiss, on Eleventh\nAmendment immunity grounds, a putative collective action\nunder the Fair Labor Standards Act; reversed the district\ncourt’s order regarding the putative collective period; and\nremanded.\n\n Plaintiff homecare providers were employed through\nCalifornia’s In-Home Supportive Services program, which\nis implemented and run by the State and its counties. In\nOctober 2013, the Department of Labor promulgated a new\nrule providing that homecare providers would be entitled to\novertime pay under the FLSA. The final rule had an\n\n *\n The Honorable Kathleen Cardone, United States District Judge for\nthe Western District of Texas, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f RAY V. COUNTY OF LOS ANGELES 3\n\neffective date of January 1, 2015. In 2014, the District Court\nfor the District of Columbia vacated the rule. On August 21,\n2015, the D.C. Circuit reversed and ordered the district court\nto enter summary judgment for the Department of Labor. On\nSeptember 14, 2015, the Department of Labor announced\nthat it would not bring enforcement actions against any\nemployer for violations of the new rule for 30 days after\nissuance of the mandate of the D.C. Circuit. On October 27,\n2015, the Department of Labor said it would not begin\nenforcing the new rule until November 12, 2015. The State\nbegan paying overtime wages on February 1, 2016.\n\n Affirming in part, the panel held that the County of Los\nAngeles was not entitled to Eleventh Amendment immunity.\nThe panel assumed without deciding that a county might be\nentitled to immunity if acting as an arm of the state. The\npanel held that, under the five-part Mitchell test, the County\nwas not an arm of the State when it administered the IHSS\nprogram because the state-treasury factor, which is the most\nimportant, and all but one of the other Mitchell factors\nweighed against immunity. The panel held that a later\nSupreme Court case, Hess v. Port Auth. Trans-Hudson\nCorp., 513 U.S. 30 (1994), did not undermine Mitchell such\nthat it should be overruled.\n\n Reversing in part, the panel held that the effective date\nof the Department of Labor’s rule was January 1, 2015,\nbecause the legal effect of the D.C. Circuit’s vacatur was to\nreinstate the original effective date. The panel held that the\nDepartment of Labor’s choice against enforcing the rule\nuntil November 12, 2015, did not eliminate the availability\nof private rights of action until that date. Accordingly, the\nbeginning of the putative collective period was January 1,\n2015.\n\f4 RAY V. COUNTY OF LOS ANGELES\n\n COUNSEL\n\nJennifer Mira Hashmall (argued) and Jeffrey B. White,\nMiller Barondess LLP, Los Angeles, California, for\nDefendant-Appellant/Cross-Appellee.\n\nMatthew C. Helland (argued) and Daniel S. Brome, Nichols\nKaster LLP, San Francisco, California; Philip Bohrer,\nBohrer Brady LLC, Baton Rouge, Louisiana; for Plaintiff-\nAppellee/Cross-Appellants.\n\n\n OPINION\n\nBENNETT, Circuit Judge:\n\n This case concerns whether a county is an arm of the\nstate and thus entitled to Eleventh Amendment immunity\nwhen it shares responsibility with the state for implementing\na state-wide homecare program. We also consider the\neffective date of regulations that (1) a district court vacated\nbefore their original effective date; (2) an appellate court\nupheld, reversing the district court; and (3) the agency then\ndecided not to enforce until a date after the original effective\ndate. We agree with the district court that the County of Los\nAngeles is not entitled to Eleventh Amendment immunity\nbut disagree as to the effective date of the regulations, which\nwe hold is the original effective date of January 1, 2015. We\nthus affirm in part, reverse in part, and remand.\n\n FACTS\n\n California’s In-Home Supportive Services program\n(“IHSS program” or “the program”) provides in-home\nsupportive services to eligible low-income elderly, blind, or\ndisabled persons. Homecare providers help recipients with\n\f RAY V. COUNTY OF LOS ANGELES 5\n\ndaily activities like housework, meal preparation, and\npersonal care. The program serves hundreds of thousands of\nrecipients. In the County of Los Angeles alone there are\nabout 170,000 homecare providers and more than 200,000\nrecipients. California implements the program through\nregulations promulgated by the California Department of\nSocial Services (CDSS), and the program is administered in\npart by California counties. Plaintiffs are current or former\nLos Angeles IHSS homecare providers.\n\n The State and its counties share responsibility for\nimplementing and running the IHSS program. The CDSS\nensures that “in-home supportive services [are] provided in\na uniform manner in every county,” Cal. Welf. & Inst. Code\n§ 12301(a), and it must “adopt regulations establishing a\nuniform range of services available to all eligible recipients\nbased upon individual needs,” id. § 12301.1(a). The State\nalso procures and implements a “Case Management\nInformation and Payroll System.” Id. § 12317(b).\n\n But counties have some oversight of the IHSS program\nas well. They, like the State, may terminate homecare\nproviders. See id. § 12300.4(b)(5). And counties evaluate\nrecipients and ensure quality compliance. See id. § 12301.1.\nCounties also “ensure that services are provided to all\neligible recipients.” Id. § 12302. Plaintiffs claim that\nalthough they receive paychecks from the State, the County\nis responsible for a “share” of their wages. For example, if\na county imposes “any increase in provider wages or benefits\n[that] is locally negotiated,” then “the county shall use\ncounty-only funds” to fund that increase. Id. § 12306.1(a).\nEach county also determines whether its providers may\nexceed the maximum number of hours set by the CDSS. See\nid. § 12300.4(d)(3).\n\f6 RAY V. COUNTY OF LOS ANGELES\n\n As employers of the homecare providers, the State and\nCounty must comply with the Fair Labor Standards Act’s\n(FLSA) overtime wage requirements. See 29 U.S.C.\n§ 207(a)(1). But that wasn’t always the case.\n\n In 1974, Congress created a “companionship exemption”\nto the FLSA for employees “employed in domestic service\nemployment to provide companionship services for\nindividuals who (because of age or infirmity) are unable to\ncare for themselves.” See id. § 213 (a)(15); Fair Labor\nStandards Amendments of 1974, Pub. L. No. 93-259,\n88 Stat. 55. This exemption applied to homecare providers\nlike Plaintiffs.\n\n In October 2013, however, the Department of Labor\n(DOL) promulgated a new rule that changed the definition\nof “companionship services” so that homecare providers like\nPlaintiffs would be entitled to overtime pay under the FLSA.\nSee Application of the Fair Labor Standards Act to Domestic\nService, 78 Fed. Reg. 60,454, 60,454 (Oct. 1, 2013)\n(codified at 29 C.F.R. pt. 552). The final rule had an\neffective date of January 1, 2015. See id.\n\n Before the rule’s effective date, a group of “trade\nassociations that represent businesses employing workers”\nsubject to the FLSA exemption filed a lawsuit in the District\nCourt for the District of Columbia. See Home Care Ass’n of\nAm. v. Weil, 76 F. Supp. 3d 138, 142 (D.D.C. 2014) (Weil I).\nThe plaintiffs claimed that the rule was arbitrary and\ncapricious and thus sought to enjoin its implementation. Id.\nat 139. At step one of its Chevron analysis, the district court\nfound that Congress had “clearly spoken” on the issue. Id.\nat 146. The district court then vacated the rule, id. at 148,\nand the DOL appealed.\n\f RAY V. COUNTY OF LOS ANGELES 7\n\n On August 21, 2015, the D.C. Circuit reversed and\nordered the district court to enter summary judgment for the\nDOL. Home Care Ass’n of Am. v. Weil, 799 F.3d 1084, 1087\n(D.C. Cir. 2015) (Weil II). Although the DOL prevailed, on\nSeptember 14, 2015 it announced that it would “not bring\nenforcement actions against any employer for violations of\nFLSA obligations resulting from the amended domestic\nservice regulations for 30 days after the date the mandate\nissues.” 1 Application of the Fair Labor Standards Act to\nDomestic Service; Announcement of 30-Day Period of Non-\nEnforcement, 80 Fed. Reg. 55,029, 55,029 (Sept. 14, 2015)\n(codified at 29 C.F.R. pt. 552). The Weil II mandate issued\non October 13, 2015.\n\n On October 27, 2015, the DOL said that it would not\nbegin enforcing the final rule until November 12, 2015.\n\n\n 1\n The DOL also stated:\n\n This 30-day non-enforcement policy does not replace\n or affect the timeline of the Department’s existing\n time-limited non-enforcement policy announced in\n October 2014. 79 FR 60974. Under that policy,\n through December 31, 2015, the Department will\n exercise prosecutorial discretion in determining\n whether to bring enforcement actions, with particular\n consideration given to the extent to which States and\n other entities have made good faith efforts to bring\n their home care programs into compliance with the\n FLSA since the promulgation of the Final Rule. The\n Department will also continue to provide intensive\n technical assistance to the regulated community, as it\n has since promulgation of the Final Rule.\n\nApplication of the Fair Labor Standards Act to Domestic Service;\nAnnouncement of 30-Day Period of Non-Enforcement, 80 Fed. Reg. at\n55,029.\n\f8 RAY V. COUNTY OF LOS ANGELES\n\nAnd, echoing its September 14, 2015 statement, the DOL\nagain said that\n\n from November 12, 2015 through December\n 31, 2015, [it would] exercise prosecutorial\n discretion in determining whether to bring\n enforcement actions, with particular\n consideration given to the extent to which\n States and other entities have made good faith\n efforts to bring their home care programs into\n compliance with the FLSA since the\n promulgation of the Final Rule.\n\nApplication of the Fair Labor Standards Act to Domestic\nService; Dates of Previously Announced 30-Day Period of\nNon-Enforcement, 80 Fed. Reg. 65,646, 65,646 (Oct. 27,\n2015) (codified at 29 C.F.R. pt. 552).\n\n Before the Weil I decision, California (through the\nCDSS) began taking steps to “meet the January 1, 2015,\nimplementation date,” including modifying its systems to\n“process and calculate overtime compensation.” But after\nthe Weil I decision, the CDSS decided that it would not\nimplement overtime payments “until further notice.” After\nWeil II, the CDSS again said that it would comply with the\novertime requirements—but not until February 1, 2016.\n\n In June 2017, Ray filed a putative collective action, 2\nunder Section 216(b) of the FLSA, against the State of\nCalifornia and the County of Los Angeles. Ray’s complaint\nsought relief for herself and the putative collective for\n\n 2\n Collective actions are provided for in the FLSA and are different\nfrom class actions, see Campbell v. City of L.A., 903 F.3d 1090, 1101\n(9th Cir. 2018), but the differences are not relevant to this appeal.\n\f RAY V. COUNTY OF LOS ANGELES 9\n\nunpaid overtime wages between January 1, 2015—the rule’s\noriginal effective date—and February 1, 2016, the date on\nwhich the State began paying overtime wages.\n\n As relevant here, the County moved to dismiss the\ncomplaint on Eleventh Amendment immunity grounds. 3 In\nthe alternative, the County moved to strike all references in\nthe complaint to overtime wages allegedly earned before\nOctober 13, 2015—the date on which the mandate issued in\nWeil II.\n\n The district court first held that the County had no\nEleventh Amendment immunity. The district court noted\nthat the Supreme Court has long refused to grant Eleventh\nAmendment immunity to counties and that the Court has\nalready held that California counties are not arms of the\nState. The district court then assumed arguendo that a\ncounty could be an arm of the State under the five-factor test\nthat we set out in Mitchell v. Los Angeles Community\nCollege District, 861 F.2d 198 (9th Cir. 1988) for\ndetermining whether an entity is an arm of the state for\npurposes of Eleventh Amendment immunity. The district\ncourt found that only one of the five factors favored the\nCounty, and thus it held that the County enjoyed no Eleventh\nAmendment immunity.\n\n The district court then “reject[ed] Plaintiffs’ efforts to\nenforce the FLSA companionship exemption regulations\nretroactively to January 1, 2015.” Instead, it held “that the\nputative collective period extends from November 12, 2015,\nthrough January 31, 2016,” and not before. The court said\n\n 3\n Early on, Ray voluntarily dismissed the CDSS as a defendant, and\nPlaintiffs did not name the State as a defendant in the now-operative\ncomplaint.\n\f10 RAY V. COUNTY OF LOS ANGELES\n\nthat although the Weil II decision applied retroactively, that\ndecision was merely that the DOL could amend the FLSA\nand that those amendments were not arbitrary and\ncapricious. This, the district court held, differed from “the\nretroactive application of the amended regulations\nthemselves.” The district court reasoned:\n\n The rule of law announced by the D.C.\n Circuit is given retroactive effect by allowing\n DOL to reinstate those regulations without\n having to begin a new rule-making process.\n That is not the same thing as reinstating an\n earlier and judicially vacated effective date\n and retroactively creating liability for\n violations of the reinstated regulations as if\n the District Court’s vacation of the\n regulations had never occurred.\n\n The district court also found it “compelling” that both the\nD.C. Circuit and the DOL “intended” that the regulation\nbecome effective “no earlier than November 12, 2015.” As\nevidence of this intent, the district court pointed to the\nDOL’s decision not to enforce the new regulations before\nthat date.\n\n Finally, the district court found that its holding was\nconsistent “with the general rule that a private right of action\nshould ordinarily not exist when the applicable rule could\nnot be enforced by the relevant enforcement agency.”\n\n The County filed an interlocutory appeal as to the denial\nof Eleventh Amendment immunity. The district court\ngranted Plaintiffs’ motion to certify for interlocutory appeal\nthe district court’s holding that the putative collective period\nbegan on November 12, 2015, and we granted Plaintiffs’\nrequest to appeal that holding.\n\f RAY V. COUNTY OF LOS ANGELES 11\n\n DISCUSSION\n\n We review de novo the denial of Eleventh Amendment\nimmunity. Cal. ex rel. Lockyer v. Dynegy, Inc., 375 F.3d\n831, 843 n.12 (9th Cir. 2004). We construe the motion to\nstrike as a motion to dismiss in part, and thus we review the\neffective date holding de novo because it essentially\ndismissed Plaintiffs’ overtime claims for the period between\nJanuary 1, 2015 and November 12, 2015. See Yamaguchi v.\nU.S. Dep’t of the Air Force, 109 F.3d 1475, 1482 (9th Cir.\n1997).\n\n The County is not entitled to Eleventh Amendment\n immunity.\n\n Plaintiffs first argue that Eleventh Amendment\nimmunity is never available to counties. The County argues\nthat it enjoys Eleventh Amendment immunity when acting\nas an “arm of the State.”\n\n Federal courts have long declined to extend Eleventh\nAmendment immunity to counties. 4 Indeed, the Supreme\n\n 4\n See, e.g., Lake Country Estates, Inc. v. Tahoe Reg’l Planning\nAgency, 440 U.S. 391, 401 (1979) (“[T]he Court has consistently refused\nto construe the Amendment to afford protection to political subdivisions\nsuch as counties and municipalities, even though such entities exercise a\nslice of state power.” (internal quotation marks omitted)); Lincoln Cty.\nv. Luning, 133 U.S. 529, 530 (1890) (holding that the Eleventh\nAmendment does not bar a suit against a county, though the principle\nadvanced has changed over time); Del Campo v. Kennedy, 517 F.3d\n1070, 1075–76 (9th Cir. 2008) (“State sovereign immunity . . . does not\nextend to counties and similar municipal corporations, even though they\nshare some portion of state power.” (internal quotation marks omitted)\n(quoting Mt. Healthy City Sch. Dist. Bd. of Educ. v. Doyle, 429 U.S. 274,\n280 (1977))).\n\f12 RAY V. COUNTY OF LOS ANGELES\n\nCourt once said that Eleventh Amendment immunity does\nnot extend to municipal corporations. Mt. Healthy, 429 U.S.\nat 280. But thirty years later, the Supreme Court suggested\nthat it was at least possible for a county to receive Eleventh\nAmendment immunity. In Northern Insurance Company of\nNew York v. Chatham County, 547 U.S. 189, 190 (2006),\nwhich involved a county-operated drawbridge, the Court\nstated that a county might be entitled to Eleventh\nAmendment immunity if it were “acting as an arm of the\nState, as delineated by this Court’s precedents, in operating\nthe drawbridge.” 5\n\n The Court cited several cases for this proposition. First,\nAlden v. Maine: “The second important limit to the principle\nof sovereign immunity is that it bars suits against States but\nnot lesser entities. The immunity does not extend to suits\nprosecuted against a municipal corporation or other\ngovernmental entity which is not an arm of the State.”\n527 U.S. 706, 756 (1999). This sentence means one of two\nthings: either (1) that Eleventh Amendment immunity does\nnot extend to municipal corporations because they are not\narms of the state or (2) that Eleventh Amendment immunity\ndoes not extend to a municipal corporation unless it is acting,\nin a particular circumstance, as an arm of the state. Alden in\nturn cites Mt. Healthy, in which the Court considered\nwhether “the Mt. Healthy Board of Education is to be treated\nas an arm of the State partaking of the State’s Eleventh\n\n 5\n At least one circuit has relied on this language and held that\ncounties might be entitled to Eleventh Amendment immunity. See\nFuesting v. Lafayette Par. Bayou Vermilion Dist., 470 F.3d 576, 579 (5th\nCir. 2006) (“[A] municipality can be immune from suit if it was ‘acting\nas an arm of the State, as delineated by [the Supreme] Court’s\nprecedent’” (alteration in original) (quoting Chatham, 547 U.S. at 194)).\nBut, to our knowledge, no court has ever actually extended Eleventh\nAmendment immunity to a county.\n\f RAY V. COUNTY OF LOS ANGELES 13\n\nAmendment immunity, or is instead to be treated as a\nmunicipal corporation or other political subdivision to which\nthe Eleventh Amendment does not extend.” Mt. Healthy,\n429 U.S. at 280. That citation suggests the former reading.\n\n The Chatham Court also cited Lake Country Estates, but\nwhile that case noted that “some agencies exercising state\npower have been permitted to invoke the [Eleventh]\nAmendment in order to protect the state treasury from\nliability that would have had essentially the same practical\nconsequences as a judgment against the State itself,” it also\nstated that “the Court has consistently refused to construe the\nAmendment to afford protection to political subdivisions\nsuch as counties and municipalities, even though such\nentities exercise a ‘slice of state power.’” Lake Country\nEstates, 440 U.S. at 400–01. Although these passages seem\nto support Plaintiffs’ argument that counties never enjoy\nEleventh Amendment immunity, it is not for us to clarify\nChatham’s apparently contrary statement.\n\n The Chatham Court ultimately found it dispositive that\nthe County there had conceded below that it had no Eleventh\nAmendment immunity and that the question on which\ncertiorari was granted assumed that conclusion. Given that\nthe Supreme Court appears to have left open the possibility\nthat a county could be entitled to Eleventh Amendment\nimmunity in some cases, we decline to hold to the contrary.\nWe therefore assume without deciding that, consistent with\nthe Court’s language in Chatham, a county might be entitled\nto Eleventh Amendment immunity if acting as an arm of the\nstate.\n\n The County is not an arm of the State here.\n\n In Mitchell, we set out five factors for determining\nwhether a government entity is an arm of its state for\n\f14 RAY V. COUNTY OF LOS ANGELES\n\nEleventh Amendment immunity purposes: (1) “whether a\nmoney judgment would be satisfied out of state funds”;\n(2) “whether the entity performs central governmental\nfunctions”; (3) “whether the entity may sue or be sued”;\n(4) “whether the entity has the power to take property in its\nown name or only the name of the state”; and (5) “the\ncorporate status of the entity.” 861 F.2d at 201. “To\ndetermine these factors, the court looks to the way state law\ntreats the entity.” Id.\n\n a. First Mitchell factor\n\n “The first Mitchell factor—whether a money judgment\n. . . would be satisfied out of state funds—is the most\nimportant.” Sato v. Orange Cty. Dep’t of Educ., 861 F.3d\n923, 929 (9th Cir. 2017); see also Beentjes v. Placer Cty. Air\nPollution Control Dist., 397 F.3d 775, 785 (9th Cir. 2005)\n(noting that the first Mitchell factor is “the one given the\nmost weight”). The County conceded, both below and on\nappeal, that it cannot show that a money judgment would be\npaid directly with State funds. 6 Thus, this factor weighs\nagainst Eleventh Amendment immunity.\n\n b. Second Mitchell factor\n\n As to the second Mitchell factor—whether the County\nperforms central governmental functions—we must\ndetermine whether the County addresses “a matter of\nstatewide rather than local or municipal concern, and the\nextent to which the state exercises centralized governmental\ncontrol over the entity.” Beentjes, 397 F.3d at 782 (internal\n 6\n The parties discuss at length how the County and the State allocate\nthe costs of the program, but that is not relevant—what matters is who\nwould be responsible for satisfying a money judgment against the\nCounty, not who pays for the program.\n\f RAY V. COUNTY OF LOS ANGELES 15\n\nquotation marks omitted) (first quoting Belanger v. Madera\nUnified Sch. Dist., 963 F.2d 248, 253 (9th Cir. 1992); then\nquoting Savage v. Glendale Union High Sch., Dist. No. 205,\nMaricopa Cty., 343 F.3d 1036, 1044 (9th Cir. 2003)).\n\n To begin, it is unclear whether the second Mitchell factor\nconcerns whether the County performs central government\nfunctions in general or whether the County performs central\ngovernment functions in carrying out the particular function\nat issue—here implementing the IHSS program.\n\n As the district court correctly noted, the closest analogue\nin our case law is Streit v. County of Los Angeles, 236 F.3d\n552 (9th Cir. 2001). There, the Los Angeles County\nSheriff’s Department (LASD) would check its systems,\nbefore releasing a prisoner, to see if the prisoner was wanted\nby another law enforcement agency. Id. at 556. This\nextended the period of incarceration one or two days past the\nprisoners’ release dates. Id. The plaintiffs alleged that the\nCounty delayed their release during these checks, in\nviolation of their civil rights. Id. The LASD argued that\nbecause it was an arm of the state, it was not a “person” that\ncould be liable for damages under § 1983. Id. at 557.\n\n We looked at the LASD’s performance of the particular\nfunction at issue—implementing the pre-release policy—not\nthe LASD’s general function as a sheriff’s department. See\nid. at 567. We held that “conducting the AJIS checks is not\na central government function.” Id. (emphasis added).\nThus, it appears from Streit that we look to whether the\nCounty, in performing the particular function at issue,\nperforms a central government function. This fits with the\nCourt’s statement in Chatham that the county there might\nhave been entitled to Eleventh Amendment immunity if it\nwere “acting as an arm of the State, as delineated by this\n\f16 RAY V. COUNTY OF LOS ANGELES\n\nCourt’s precedents, in operating [a] drawbridge.”\nChatham, 547 U.S. at 194 (emphasis added).\n\n i. A matter of statewide rather than local or\n municipal concern\n\n The in-home care of the elderly and disabled is a matter\nof both statewide and local concern. Plaintiffs are residents\nof California, and the IHSS program is a statewide program\nimplemented through State legislation that provides care to\nhundreds of thousands of California residents. But Plaintiffs\nare also, of course, residents of Los Angeles County, and the\nCounty has an interest in the program and the care provided\nin Los Angeles.\n\n ii. The extent to which the state exercises\n centralized governmental control over the\n entity\n\n Here we consider the extent to which the County, in\nimplementing the program, has “discretionary powers” and\n“substantial autonomy in carrying out [its] duties.” Beentjes,\n397 F.3d at 783.\n\n The County may negotiate, implement, and pay for pay\nraises. See Cal. Welf. & Inst. Code § 12306.1. The County\nmay also allow its providers to exceed the maximum number\nof hours that the CDSS has set. See id. § 12300.4(d)(3).\nThus, the County has discretion to make some important\nchoices on its own.\n\n But the County contends—and Plaintiffs do not\ndispute—that it has no discretion over the action (or\ninaction) that subjected it to potential liability here: payment\nof overtime wages under the FLSA. In taking the actions\nthat have subjected it to potential liability, the County had\n\f RAY V. COUNTY OF LOS ANGELES 17\n\nneither “discretionary powers” nor “substantial autonomy”\nin carrying out its duties.\n\n We think this clearly tips the scales in the County’s favor\nas to this factor. The County had no choice in the matter of\nthe overtime wages, as the State mandated the payment start\ndate. We therefore hold that the second Mitchell factor\nfavors Eleventh Amendment immunity.\n\n c. Third, fourth, and fifth Mitchell factors\n\n The County does not dispute that it can sue and be sued\n(third Mitchell factor), that it has the power to take property\nin its own name (fourth Mitchell factor), or that it has an\nindependent corporate status 7 separate from the State (fifth\nMitchell factor). Thus, these three Mitchell factors weigh\nagainst Eleventh Amendment immunity.\n\n * * *\n\n In sum, the first Mitchell factor is the most important,\nand it weighs against Eleventh Amendment immunity. So\ndo the third, fourth, and fifth Mitchell factors. Only the\nsecond factor favors immunity. We therefore hold that,\nunder Mitchell, the County is not an arm of the State when it\n\n\n\n\n 7\n The fifth Mitchell factor asks whether the entity has “independent\ncorporate status,” Holz v. Nenana City Pub. Sch. Dist., 347 F.3d 1176,\n1188 (9th Cir. 2003), or is, instead, merely an agency of the state without\nan identity that is separate from the state, Beentjes, 397 F.3d at 785. Here\nthe County does not dispute its independent corporate status, as the\nSupreme Court has already held that California counties have\nindependent corporate status and are not agents of the State of California.\nSee Moor v. Alameda Cty., 411 U.S. 693, 719 (1973).\n\f18 RAY V. COUNTY OF LOS ANGELES\n\nadministers the IHSS program, and thus it has no Eleventh\nAmendment immunity barring this action.\n\n The Supreme Court has not overruled or\n undermined Mitchell.\n\n The County argues that we should overrule Mitchell\nbecause a later Supreme Court case, Hess v. Port Authority\nTrans-Hudson Corporation, 513 U.S. 30 (1994),\nundermined it. As a three-judge panel, if we find that\nintervening Supreme Court authority is clearly\nirreconcilable with our own precedent, we must consider\nourselves bound by the intervening higher authority and\nconsider our precedent effectively overruled. See Miller v.\nGammie, 335 F.3d 889, 900 (9th Cir. 2003). Because Hess\nis not clearly irreconcilable with Mitchell, we reject the\nCounty’s argument.\n\n In Hess, the Court held that a Congressionally approved\nbistate entity—the Port Authority Trans-Hudson\nCorporation (PATH), created to improve coordination of the\n“terminal, transportation and other facilities of commerce in,\nabout and through the port of New York”—did not have\nEleventh Amendment immunity. 513 U.S at 35, 52–53\n(citation omitted). The County argues that Hess established\n“indicators of immunity” that undermine the Mitchell test.\nWe disagree.\n\n The Hess Court noted that “current Eleventh\nAmendment jurisprudence emphasizes the integrity retained\nby each State in our federal system.” Id. at 39. The Court\nthen emphasized the difference between PATH and the\nStates of the Union: “The States, as separate sovereigns, are\nthe constituent elements of the Union. Bistate entities, in\ncontrast, typically are creations of three discrete sovereigns:\ntwo States and the Federal Government.” Id. at 40.\n\f RAY V. COUNTY OF LOS ANGELES 19\n\n The Court stated that “[p]ointing away from Eleventh\nAmendment immunity, the States lack financial\nresponsibility” for the bistate entity. Id. at 45. Here,\nCalifornia similarly lacks financial responsibility for the\nCounty generally, but Plaintiffs allege that although\nCalifornia writes their checks, the County pays a share of\ntheir wages and sets their hours of work.\n\n In Hess, “indicators of immunity point[ed] in different\ndirections.” Id. at 47. Perhaps they do here as well. Los\nAngeles is not a constituent member of the Union, but it\nacted at the direction of the State and had no authority over\nthe payments at issue. But when faced with a different\ndichotomy in Hess, the Court emphasized that the most\nimportant factor was whether judgments against PATH\nwould be paid by the State: “the vulnerability of the State’s\npurse [is] the most salient factor in Eleventh Amendment\ndeterminations.” Id. at 48; see also id. at 48–49 (citing cases\nfor the “prevailing view” that the state-treasury factor is\n“generally accorded . . . dispositive weight”); id. at 51\n(stating that “the Eleventh Amendment’s core concern is not\nimplicated” if the State is not “in fact obligated to bear and\npay the . . . indebtedness of the enterprise”). 8\n\n\n\n\n 8\n The dissent read the holding even more broadly:\n\n In place of the various factors recognized in Lake\n Country Estates, Inc. v. Tahoe Regional Planning\n Agency, 440 U.S. 391, 99 S. Ct. 1171, 59 L.Ed.2d 401\n (1979), for determining arm-of-the-state status, we\n may now substitute a single overriding criterion,\n vulnerability of the state treasury. If a State does not\n fund judgments against an entity, that entity is not\n\f20 RAY V. COUNTY OF LOS ANGELES\n\n After noting that the bistate entity “was financially self-\nsufficient,” generated “its own revenues,” and paid “its own\ndebts,” the Court held that “[r]equiring the [bistate entity] to\nanswer in federal court to injured railroad workers who\nassert a federal statutory right, under the FELA, to recover\ndamages does not touch the concerns—the States’ solvency\nand dignity—that underpin the Eleventh Amendment.” Id.\nat 52. The same is true here. Mitchell and Hess both\nemphasize the state-treasury factor. Hess thus fully supports\nand does not undermine Mitchell. 9\n\n The County argues that Hess emphasized the amount of\ncontrol that a state maintains over an entity, a factor\nsupposedly not mentioned in Mitchell and one that,\naccording to the County, favors Eleventh Amendment\nimmunity here. First, as we mentioned above, the second\nMitchell factor does include a “control” inquiry—it just\ndoesn’t make that factor dispositive. In addition, Hess\npointed out that “[g]auging actual control . . . can be a\n‘perilous inquiry,’ [and] ‘an uncertain and unreliable\nexercise.’” 513 U.S. at 47 (quoting Note, 92 Colum. L. Rev.\n1243, 1284 (1992)). The Court therefore doubted not only\nthe efficacy but also the utility of a “control” analysis, and it\n\n\n\n within the ambit of the Eleventh Amendment, and\n suits in federal court may proceed unimpeded.\n\nId. at 55 (O’Connor, J., dissenting).\n 9\n Los Angeles makes a legitimate point about the unfairness of the\nresult here. But that unfairness springs from the State and its\nimplementing legislation, not the Eleventh Amendment. Los Angeles\nmust air its grievance, if at all, in Sacramento.\n\f RAY V. COUNTY OF LOS ANGELES 21\n\ndid not suggest that control was a favored, much less\ndispositive, factor in the Eleventh Amendment analysis. 10\n\n Hess clearly stated that “rendering control dispositive\ndoes not home in on the impetus for the Eleventh\nAmendment: the prevention of federal-court judgments that\nmust be paid out of a State’s treasury.” Id. at 48. And, in\nspecifically discussing the control factor, the Court noted\nthat even though “‘political subdivisions exist solely at the\nwhim and behest of their State,’ . . . cities and counties do\nnot enjoy Eleventh Amendment immunity.” Id. at 47\n(quoting Port Auth. Trans-Hudson Corp. v. Feeney, 495 U.S.\n299, 313 (1990)).\n\n Finally, the County insists that Hess compels us to\nconsider the State’s dignity, a factor not mentioned in\nMitchell. Hess noted that the State’s “solvency and dignity\n. . . underpin the Eleventh Amendment.” 513 U.S. at 52.\nThat is undoubtedly true. But the State is no longer a party\nto this action, and it will not be responsible for an adverse\njudgment against the County. Allowing this action against\nLos Angeles does not injure California’s dignity. 11\n\n\n\n 10\n The control discussed in Hess seems to have gone to overall\ncontrol over the entity, not just control within the context of the particular\nfunction at issue: “PATH urges that we find good reason to classify the\nPort Authority as a state agency for Eleventh Amendment purposes\nbased on the control New York and New Jersey wield over the\nAuthority. . . . But ultimate control of every state-created entity resides\nwith the State, for the State may destroy or reshape any unit it creates.”\nId. at 47. Thus, looking at the State’s overall control over the County as\na county would not help the County’s position here.\n 11\n And, although it would not have altered our analysis, we note that\nCalifornia has not sought to file an amicus brief (below or on appeal)\n\f22 RAY V. COUNTY OF LOS ANGELES\n\n The Supreme Court decided Hess about five years after\nwe decided Mitchell. And although Hess arose in a different\ncontext than Mitchell-Hess addressed a bistate entity, not a\ncounty—nothing in Hess so undermines Mitchell that we\nhave the power to overrule it. More importantly, even if we\nused Hess rather than Mitchell to guide our analysis, we\nwould reach the same result.\n\n When a non-state entity invokes Eleventh Amendment\nimmunity, the most important factor for determining\nwhether the entity is an arm of the state remains the state-\ntreasury factor—that is, whether the state will be liable for a\nmoney judgment against the non-state entity. That factor,\nand all but one of the other Mitchell factors, dictates the\nresult here. The Eleventh Amendment does not bar\nPlaintiffs’ suit against Los Angeles.\n\n The effective date of the rule is January 1, 2015.\n\n We next consider whether the effective date of the rule\nis the original effective date of January 1, 2015 or some date\nafter the D.C. Circuit reversed the district court’s vacatur.\nThe County argues that the rule cannot have an effective date\nthat is earlier than the date on which the D.C. Circuit\nreversed the district court’s vacatur. Plaintiffs argue that the\nlegal effect of the vacatur is to reinstate the original January\n1, 2015 effective date. We agree with Plaintiffs and hold\nthat the effective date of the rule is January 1, 2015.\n\n\n\n\narguing either that the County is entitled to Eleventh Amendment\nimmunity or that this case threatens California’s dignity.\n\f RAY V. COUNTY OF LOS ANGELES 23\n\n A January 1, 2015 effective date is not\n impermissibly retroactive.\n\n The County argues that a January 1, 2015 effective date\nis impermissibly retroactive. Plaintiffs argue that the D.C.\nCircuit’s decision, not the rule, applies retroactively, because\nthe D.C. Circuit was “explaining what the law always was,”\nand thus reinstating the original effective date is merely a\nreturn to the status quo ante.\n\n When an appellate court applies “a rule of federal law to\nthe parties before it,” that interpretation “must be given full\nretroactive effect in all cases still open on direct review and\nas to all events, regardless of whether such events predate or\npostdate [the] announcement of the rule.” Harper v. Va.\nDep’t of Taxation, 509 U.S. 86, 97 (1993). That is because\n“when a court delivers a ruling, even if it is unforeseen, the\nlaw has not changed. Rather, the court is explaining what\nthe law always was.” Jones Stevedoring Co. v. Dir., Office\nof Workers’ Comp. Programs, 133 F.3d 683, 688 (9th Cir.\n1997).\n\n When the D.C. Circuit held that the DOL had the\nrulemaking authority to promulgate the new rule and that its\nnew rule was a reasonable exercise of that authority, see Weil\nII, 799 F.3d at 1090, it did not change the law but merely\nexplained what the law always was—the district court’s\nerroneous contrary holding notwithstanding.\n\n Two cases support our holding. In GTE South, Inc. v.\nMorrison, 199 F.3d 733, 738, 740 (4th Cir. 1999), the Fourth\nCircuit addressed an issue much like the one we face:\ndetermining the effective date of certain pricing rules,\npromulgated by the FCC, that the Eighth Circuit stayed and\nthen vacated before their effective date. The Supreme Court\nlater reversed the Eighth Circuit. See AT & T Corp. v. Iowa\n\f24 RAY V. COUNTY OF LOS ANGELES\n\nUtilities Bd., 525 U.S. 366, 385 (1999). The Morrison panel\nheld that “the Supreme Court’s determination that the FCC\nhas jurisdiction to issue pricing rules would appear to\ncompel the conclusion that the FCC always had such\njurisdiction and that the rules apply as of the effective date\noriginally scheduled.” 199 F.3d at 740 (emphasis added).\nThe Fourth Circuit emphasized that its holding was not\nunfair to the parties who argued for a later effective date\nbecause they had “ample notice” of the original effective\ndate and “surely knew that the FCC’s authority to issue\npricing rules might ultimately be upheld by the Supreme\nCourt.” Id. at 741.\n\n In US West Communication, Inc. v. Jennings, 304 F.3d\n950, 955 (9th Cir. 2002), we considered a similar question:\nwhether the regulations that the Fourth Circuit considered in\nMorrison applied to conduct that occurred during the period\nof vacatur. Finding the Fourth Circuit’s reasoning in\nMorrison persuasive and applicable, we noted that the\nSupreme Court’s determination that the regulations were\nvalid meant that we should apply them “to all . . . agreements\narbitrated under the Act, including agreements arbitrated\nbefore the rules were reinstated.” Id. at 957 (emphasis\nadded). Relying on Morrison, we held that applying the\nreinstated regulations to conduct that occurred during the\nperiod of vacatur would not give the regulations an\nimpermissible retroactive effect. Id. at 958.\n\n Morrison and Jennings are analogous to this case\nbecause both involved determining how to apply rules or\nregulations that were vacated but ultimately reinstated on\nappeal. Indeed, Morrison commented not only on the\nretroactivity of the Supreme Court’s reversal but also on the\neffective date of the regulations, holding that the intervening\n\f RAY V. COUNTY OF LOS ANGELES 25\n\nvacatur did not alter the original effective date of the pricing\nrules. 199 F.3d at 740.\n\n Thus, Morrison and Jennings guide our analysis here.\nThe D.C. Circuit’s holding that the DOL had the authority to\npromulgate the new rule and that the rule was reasonable\napplies retroactively. As in Jennings, the regulations apply\nas of the original effective date. To hold otherwise could\nencourage dilatory appellate litigation. If an erroneously\nvacated rule or regulation were not effective until sometime\nafter the mandate issued in a later appeal, then a party might\ndrag out the appellate process to avoid compliance for as\nlong as possible. Put differently, an erroneous vacatur\ncannot postpone a rule’s effective date until an appellate\ncourt corrects the error sometime in the future. And, as the\nFourth Circuit noted in Morrison, in a case like this everyone\nknows that the lower court decision might be reversed on\nappeal.\n\n The State and its counties knew from October 13, 2013,\nwhen the DOL first announced its final rule, that January 1,\n2015 was the rule’s effective date. See Application of the\nFair Labor Standards Act to Domestic Service, 78 Fed. Reg.\nat 60,454. The State and its counties had a full fifteen\nmonths to comply with the final rule—indeed the State\ninitially said that it would comply with the original effective\ndate, but it changed course after the Weil I court vacated the\nrule. That decision may have been reasonable, but it created\na monetary risk, as the State and its counties were well aware\nthat an appellate court might uphold the regulations on\nappeal.\n\n The district court held that to apply the Weil II decision\nretroactively would be to “reinstate[] an earlier and judicially\nvacated effective date and retroactively creat[e] liability for\nviolations of the reinstated regulations as if the District\n\f26 RAY V. COUNTY OF LOS ANGELES\n\nCourt’s vacation of the regulations had never occurred.”\nThat is exactly correct. And although the district court found\nthat to be unfair, it would be equally unfair to hold that a\nputative collective of homecare providers is not entitled to\nnearly a year’s worth of overtime wages just because a single\ndistrict court issued an erroneous decision that another court\nreversed on appeal. The State gambled that Weil I would be\naffirmed. The effect of that gamble might be unfair to the\nCounty, but the County must seek any recourse from the\nState. It is not fair for the homecare providers to bear the\nfinancial consequences of the State’s calculated risk.\n\n The DOL’s decision not to enforce a new rule does\n not obviate private rights of action.\n\n According to the County, the DOL’s choice against\nenforcing the rule until November 12, 2015 eliminated the\navailability of private rights of action until that date because\na private right of action cannot precede an agency’s\nenforcement of a rule or regulation. We disagree.\n\n “An agency’s informal assurance that it will not pursue\nenforcement cannot preclude a citizen’s suit to do so.” Ohio\nValley Envtl. Coal. v. Fola Coal Co., LLC, 845 F.3d 133,\n145 (4th Cir. 2017) (emphasis added). Congress created a\nprivate right of action under the FLSA for unpaid overtime:\n“Any employer who violates the provisions of section 206\nor section 207 of this title shall be liable to the employee or\nemployees affected in the amount of their . . . unpaid\novertime compensation . . . .” 29 U.S.C. § 216(b). An\nagency’s discretionary decision to hold off enforcement does\nnot and cannot strip private parties of their rights to do so.\nSee Ohio Valley, 845 F.3d at 145 (“Congress enacted the\ncitizen suit provision of the Clean Water Act to address\nsituations, like the one at hand, in which the traditional\nenforcement agency declines to act.”).\n\f RAY V. COUNTY OF LOS ANGELES 27\n\n The district court’s hypothesis that the D.C. Circuit and\nDOL “intended” that the regulation become effective “no\nearlier than November 12, 2015” is tenuous and, in any\nevent, irrelevant. First, the D.C. Circuit said nothing at all\non the issue. Second, there is nothing in the several\nstatements of the DOL, which the district court relied on, that\nsuggest that it intended its discretionary enforcement choices\nto preclude private enforcement. Indeed, other than by\namending the rule, the DOL could not have precluded\nprivate enforcement even if it wanted to.\n\n The rule’s original effective date remains January 1,\n2015. If the DOL “intended” for the effective date be\nsomething other than January 1, 2015, the DOL could have\nsought to change that effective date through the procedures\nset out in the Administrative Procedure Act. Were we to\nhold to the contrary and impose our view that the DOL’s\nexercise of discretion amended the effective date sub\nsilentio, we would in fact be usurping the rulemaking\nauthority of the DOL. See Nat. Res. Def. Council, Inc. v.\nU.S. E.P.A., 683 F.2d 752, 762 (3d Cir. 1982) (holding that\na final rule’s effective date is an “essential part” of that rule\nand is thus subject to the rulemaking procedures of the\nAPA).\n\n The effective date of the rule is January 1, 2015. 12\n\n\n 12\n Although some district courts have reached a different\nconclusion—see, e.g., Bangoy v. Total Homecare Solutions, LLC, No.\n1:15-CV-573, 2015 WL 12672727, at *3 (S.D. Ohio Dec. 21, 2015)\n(holding that the plaintiffs failed to state a claim for a violation of the\nFLSA between January 1, 2015 and “late August 2015”)—nearly all of\nthem have reached the same result we reach here, see, e.g., Kinkead v.\nHumana, Inc., 206 F. Supp. 3d 751, 752 (D. Conn. 2016) (holding that\nthe effective date of the rule is January 1, 2015, “the effective date set\n\f28 RAY V. COUNTY OF LOS ANGELES\n\n CONCLUSION\n\n We AFFIRM the district court’s holding that the County\nis not entitled to Eleventh Amendment immunity and\nREVERSE the district court’s holding that the putative\ncollective period began on November 12, 2015, holding\ninstead that the rule’s effective date—and thus the beginning\nof the putative collective period—is January 1, 2015. We\nREMAND for proceedings consistent with this opinion.\nCosts shall be awarded to Plaintiffs-Appellants.\n\n\n\n\nforth by the agency”); Collins v. DKL Ventures, LLC, 215 F. Supp. 3d\n1059 (D. Colo. 2016) (same); Lewis-Ramsey v. Evangelical Lutheran\nGood Samaritan Soc’y, 215 F. Supp. 3d 805 (S.D. Iowa 2016) (same).\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Trina Ray v. County of Los Angeles\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-22\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nTRINA RAY, individually, and on No. 17-56581\\nbehalf of others similarly situated,\\n Plaintiff-Appellee, D.C. No.\\n 2:17-cv-04239-\\n v. PA-SK\\n\\nCOUNTY OF LOS ANGELES,\\n Defendant-Appellant.\\n\\n\\nTRINA RAY; SASHA WALKER, No. 18-55276\\nindividually, and on behalf of all\\nothers similarly situated, D.C. No.\\n Plaintiffs-Appellants, 2:17-cv-04239-\\n PA-SK\\n v.\\n\\nLOS ANGELES COUNTY DEPARTMENT OPINION\\nOF PUBLIC SOCIAL SERVICES,\\nErroneously Sued As County of Los\\nAngeles,\\n Defendant-Appellee.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Percy Anderson, District Judge, Presiding\\n\\f2 RAY V. COUNTY OF LOS ANGELES\\n\\n Argued and Submitted March 7, 2019\\n Pasadena, California\\n\\n Filed August 22, 2019\\n\\n Before: Kim McLane Wardlaw and Mark J. Bennett,\\n Circuit Judges, and Kathleen Cardone, * District Judge.\\n\\n Opinion by Judge Bennett\\n\\n\\n SUMMARY **\\n\\n\\n Labor Law / Eleventh Amendment Immunity\\n\\n The panel affirmed the district court’s order denying a\\ndefendant county’s motion to dismiss, on Eleventh\\nAmendment immunity grounds, a putative collective action\\nunder the Fair Labor Standards Act; reversed the district\\ncourt’s order regarding the putative collective period; and\\nremanded.\\n\\n Plaintiff homecare providers were employed through\\nCalifornia’s In-Home Supportive Services program, which\\nis implemented and run by the State and its counties. In\\nOctober 2013, the Department of Labor promulgated a new\\nrule providing that homecare providers would be entitled to\\novertime pay under the FLSA. The final rule had an\\n\\n *\\n The Honorable Kathleen Cardone, United States District Judge for\\nthe Western District of Texas, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f RAY V. COUNTY OF LOS ANGELES 3\\n\\neffective date of January 1, 2015. In 2014, the District Court\\nfor the District of Columbia vacated the rule. On August 21,\\n2015, the D.C. Circuit reversed and ordered the district court\\nto enter summary judgment for the Department of Labor. On\\nSeptember 14, 2015, the Department of Labor announced\\nthat it would not bring enforcement actions against any\\nemployer for violations of the new rule for 30 days after\\nissuance of the mandate of the D.C. Circuit. On October 27,\\n2015, the Department of Labor said it would not begin\\nenforcing the new rule until November 12, 2015. The State\\nbegan paying overtime wages on February 1, 2016.\\n\\n Affirming in part, the panel held that the County of Los\\nAngeles was not entitled to Eleventh Amendment immunity.\\nThe panel assumed without deciding that a county might be\\nentitled to immunity if acting as an arm of the state. The\\npanel held that, under the five-part Mitchell test, the County\\nwas not an arm of the State when it administered the IHSS\\nprogram because the state-treasury factor, which is the most\\nimportant, and all but one of the other Mitchell factors\\nweighed against immunity. The panel held that a later\\nSupreme Court case, Hess v. Port Auth. Trans-Hudson\\nCorp., 513 U.S. 30 (1994), did not undermine Mitchell such\\nthat it should be overruled.\\n\\n Reversing in part, the panel held that the effective date\\nof the Department of Labor’s rule was January 1, 2015,\\nbecause the legal effect of the D.C. Circuit’s vacatur was to\\nreinstate the original effective date. The panel held that the\\nDepartment of Labor’s choice against enforcing the rule\\nuntil November 12, 2015, did not eliminate the availability\\nof private rights of action until that date. Accordingly, the\\nbeginning of the putative collective period was January 1,\\n2015.\\n\\f4 RAY V. COUNTY OF LOS ANGELES\\n\\n COUNSEL\\n\\nJennifer Mira Hashmall (argued) and Jeffrey B. White,\\nMiller Barondess LLP, Los Angeles, California, for\\nDefendant-Appellant/Cross-Appellee.\\n\\nMatthew C. Helland (argued) and Daniel S. Brome, Nichols\\nKaster LLP, San Francisco, California; Philip Bohrer,\\nBohrer Brady LLC, Baton Rouge, Louisiana; for Plaintiff-\\nAppellee/Cross-Appellants.\\n\\n\\n OPINION\\n\\nBENNETT, Circuit Judge:\\n\\n This case concerns whether a county is an arm of the\\nstate and thus entitled to Eleventh Amendment immunity\\nwhen it shares responsibility with the state for implementing\\na state-wide homecare program. We also consider the\\neffective date of regulations that (1) a district court vacated\\nbefore their original effective date; (2) an appellate court\\nupheld, reversing the district court; and (3) the agency then\\ndecided not to enforce until a date after the original effective\\ndate. We agree with the district court that the County of Los\\nAngeles is not entitled to Eleventh Amendment immunity\\nbut disagree as to the effective date of the regulations, which\\nwe hold is the original effective date of January 1, 2015. We\\nthus affirm in part, reverse in part, and remand.\\n\\n FACTS\\n\\n California’s In-Home Supportive Services program\\n(“IHSS program” or “the program”) provides in-home\\nsupportive services to eligible low-income elderly, blind, or\\ndisabled persons. Homecare providers help recipients with\\n\\f RAY V. COUNTY OF LOS ANGELES 5\\n\\ndaily activities like housework, meal preparation, and\\npersonal care. The program serves hundreds of thousands of\\nrecipients. In the County of Los Angeles alone there are\\nabout 170,000 homecare providers and more than 200,000\\nrecipients. California implements the program through\\nregulations promulgated by the California Department of\\nSocial Services (CDSS), and the program is administered in\\npart by California counties. Plaintiffs are current or former\\nLos Angeles IHSS homecare providers.\\n\\n The State and its counties share responsibility for\\nimplementing and running the IHSS program. The CDSS\\nensures that “in-home supportive services [are] provided in\\na uniform manner in every county,” Cal. Welf. & Inst. Code\\n§ 12301(a), and it must “adopt regulations establishing a\\nuniform range of services available to all eligible recipients\\nbased upon individual needs,” id. § 12301.1(a). The State\\nalso procures and implements a “Case Management\\nInformation and Payroll System.” Id. § 12317(b).\\n\\n But counties have some oversight of the IHSS program\\nas well. They, like the State, may terminate homecare\\nproviders. See id. § 12300.4(b)(5). And counties evaluate\\nrecipients and ensure quality compliance. See id. § 12301.1.\\nCounties also “ensure that services are provided to all\\neligible recipients.” Id. § 12302. Plaintiffs claim that\\nalthough they receive paychecks from the State, the County\\nis responsible for a “share” of their wages. For example, if\\na county imposes “any increase in provider wages or benefits\\n[that] is locally negotiated,” then “the county shall use\\ncounty-only funds” to fund that increase. Id. § 12306.1(a).\\nEach county also determines whether its providers may\\nexceed the maximum number of hours set by the CDSS. See\\nid. § 12300.4(d)(3).\\n\\f6 RAY V. COUNTY OF LOS ANGELES\\n\\n As employers of the homecare providers, the State and\\nCounty must comply with the Fair Labor Standards Act’s\\n(FLSA) overtime wage requirements. See 29 U.S.C.\\n§ 207(a)(1). But that wasn’t always the case.\\n\\n In 1974, Congress created a “companionship exemption”\\nto the FLSA for employees “employed in domestic service\\nemployment to provide companionship services for\\nindividuals who (because of age or infirmity) are unable to\\ncare for themselves.” See id. § 213 (a)(15); Fair Labor\\nStandards Amendments of 1974, Pub. L. No. 93-259,\\n88 Stat. 55. This exemption applied to homecare providers\\nlike Plaintiffs.\\n\\n In October 2013, however, the Department of Labor\\n(DOL) promulgated a new rule that changed the definition\\nof “companionship services” so that homecare providers like\\nPlaintiffs would be entitled to overtime pay under the FLSA.\\nSee Application of the Fair Labor Standards Act to Domestic\\nService, 78 Fed. Reg. 60,454, 60,454 (Oct. 1, 2013)\\n(codified at 29 C.F.R. pt. 552). The final rule had an\\neffective date of January 1, 2015. See id.\\n\\n Before the rule’s effective date, a group of “trade\\nassociations that represent businesses employing workers”\\nsubject to the FLSA exemption filed a lawsuit in the District\\nCourt for the District of Columbia. See Home Care Ass’n of\\nAm. v. Weil, 76 F. Supp. 3d 138, 142 (D.D.C. 2014) (Weil I).\\nThe plaintiffs claimed that the rule was arbitrary and\\ncapricious and thus sought to enjoin its implementation. Id.\\nat 139. At step one of its Chevron analysis, the district court\\nfound that Congress had “clearly spoken” on the issue. Id.\\nat 146. The district court then vacated the rule, id. at 148,\\nand the DOL appealed.\\n\\f RAY V. COUNTY OF LOS ANGELES 7\\n\\n On August 21, 2015, the D.C. Circuit reversed and\\nordered the district court to enter summary judgment for the\\nDOL. Home Care Ass’n of Am. v. Weil, 799 F.3d 1084, 1087\\n(D.C. Cir. 2015) (Weil II). Although the DOL prevailed, on\\nSeptember 14, 2015 it announced that it would “not bring\\nenforcement actions against any employer for violations of\\nFLSA obligations resulting from the amended domestic\\nservice regulations for 30 days after the date the mandate\\nissues.” 1 Application of the Fair Labor Standards Act to\\nDomestic Service; Announcement of 30-Day Period of Non-\\nEnforcement, 80 Fed. Reg. 55,029, 55,029 (Sept. 14, 2015)\\n(codified at 29 C.F.R. pt. 552). The Weil II mandate issued\\non October 13, 2015.\\n\\n On October 27, 2015, the DOL said that it would not\\nbegin enforcing the final rule until November 12, 2015.\\n\\n\\n 1\\n The DOL also stated:\\n\\n This 30-day non-enforcement policy does not replace\\n or affect the timeline of the Department’s existing\\n time-limited non-enforcement policy announced in\\n October 2014. 79 FR 60974. Under that policy,\\n through December 31, 2015, the Department will\\n exercise prosecutorial discretion in determining\\n whether to bring enforcement actions, with particular\\n consideration given to the extent to which States and\\n other entities have made good faith efforts to bring\\n their home care programs into compliance with the\\n FLSA since the promulgation of the Final Rule. The\\n Department will also continue to provide intensive\\n technical assistance to the regulated community, as it\\n has since promulgation of the Final Rule.\\n\\nApplication of the Fair Labor Standards Act to Domestic Service;\\nAnnouncement of 30-Day Period of Non-Enforcement, 80 Fed. Reg. at\\n55,029.\\n\\f8 RAY V. COUNTY OF LOS ANGELES\\n\\nAnd, echoing its September 14, 2015 statement, the DOL\\nagain said that\\n\\n from November 12, 2015 through December\\n 31, 2015, [it would] exercise prosecutorial\\n discretion in determining whether to bring\\n enforcement actions, with particular\\n consideration given to the extent to which\\n States and other entities have made good faith\\n efforts to bring their home care programs into\\n compliance with the FLSA since the\\n promulgation of the Final Rule.\\n\\nApplication of the Fair Labor Standards Act to Domestic\\nService; Dates of Previously Announced 30-Day Period of\\nNon-Enforcement, 80 Fed. Reg. 65,646, 65,646 (Oct. 27,\\n2015) (codified at 29 C.F.R. pt. 552).\\n\\n Before the Weil I decision, California (through the\\nCDSS) began taking steps to “meet the January 1, 2015,\\nimplementation date,” including modifying its systems to\\n“process and calculate overtime compensation.” But after\\nthe Weil I decision, the CDSS decided that it would not\\nimplement overtime payments “until further notice.” After\\nWeil II, the CDSS again said that it would comply with the\\novertime requirements—but not until February 1, 2016.\\n\\n In June 2017, Ray filed a putative collective action, 2\\nunder Section 216(b) of the FLSA, against the State of\\nCalifornia and the County of Los Angeles. Ray’s complaint\\nsought relief for herself and the putative collective for\\n\\n 2\\n Collective actions are provided for in the FLSA and are different\\nfrom class actions, see Campbell v. City of L.A., 903 F.3d 1090, 1101\\n(9th Cir. 2018), but the differences are not relevant to this appeal.\\n\\f RAY V. COUNTY OF LOS ANGELES 9\\n\\nunpaid overtime wages between January 1, 2015—the rule’s\\noriginal effective date—and February 1, 2016, the date on\\nwhich the State began paying overtime wages.\\n\\n As relevant here, the County moved to dismiss the\\ncomplaint on Eleventh Amendment immunity grounds. 3 In\\nthe alternative, the County moved to strike all references in\\nthe complaint to overtime wages allegedly earned before\\nOctober 13, 2015—the date on which the mandate issued in\\nWeil II.\\n\\n The district court first held that the County had no\\nEleventh Amendment immunity. The district court noted\\nthat the Supreme Court has long refused to grant Eleventh\\nAmendment immunity to counties and that the Court has\\nalready held that California counties are not arms of the\\nState. The district court then assumed arguendo that a\\ncounty could be an arm of the State under the five-factor test\\nthat we set out in Mitchell v. Los Angeles Community\\nCollege District, 861 F.2d 198 (9th Cir. 1988) for\\ndetermining whether an entity is an arm of the state for\\npurposes of Eleventh Amendment immunity. The district\\ncourt found that only one of the five factors favored the\\nCounty, and thus it held that the County enjoyed no Eleventh\\nAmendment immunity.\\n\\n The district court then “reject[ed] Plaintiffs’ efforts to\\nenforce the FLSA companionship exemption regulations\\nretroactively to January 1, 2015.” Instead, it held “that the\\nputative collective period extends from November 12, 2015,\\nthrough January 31, 2016,” and not before. The court said\\n\\n 3\\n Early on, Ray voluntarily dismissed the CDSS as a defendant, and\\nPlaintiffs did not name the State as a defendant in the now-operative\\ncomplaint.\\n\\f10 RAY V. COUNTY OF LOS ANGELES\\n\\nthat although the Weil II decision applied retroactively, that\\ndecision was merely that the DOL could amend the FLSA\\nand that those amendments were not arbitrary and\\ncapricious. This, the district court held, differed from “the\\nretroactive application of the amended regulations\\nthemselves.” The district court reasoned:\\n\\n The rule of law announced by the D.C.\\n Circuit is given retroactive effect by allowing\\n DOL to reinstate those regulations without\\n having to begin a new rule-making process.\\n That is not the same thing as reinstating an\\n earlier and judicially vacated effective date\\n and retroactively creating liability for\\n violations of the reinstated regulations as if\\n the District Court’s vacation of the\\n regulations had never occurred.\\n\\n The district court also found it “compelling” that both the\\nD.C. Circuit and the DOL “intended” that the regulation\\nbecome effective “no earlier than November 12, 2015.” As\\nevidence of this intent, the district court pointed to the\\nDOL’s decision not to enforce the new regulations before\\nthat date.\\n\\n Finally, the district court found that its holding was\\nconsistent “with the general rule that a private right of action\\nshould ordinarily not exist when the applicable rule could\\nnot be enforced by the relevant enforcement agency.”\\n\\n The County filed an interlocutory appeal as to the denial\\nof Eleventh Amendment immunity. The district court\\ngranted Plaintiffs’ motion to certify for interlocutory appeal\\nthe district court’s holding that the putative collective period\\nbegan on November 12, 2015, and we granted Plaintiffs’\\nrequest to appeal that holding.\\n\\f RAY V. COUNTY OF LOS ANGELES 11\\n\\n DISCUSSION\\n\\n We review de novo the denial of Eleventh Amendment\\nimmunity. Cal. ex rel. Lockyer v. Dynegy, Inc., 375 F.3d\\n831, 843 n.12 (9th Cir. 2004). We construe the motion to\\nstrike as a motion to dismiss in part, and thus we review the\\neffective date holding de novo because it essentially\\ndismissed Plaintiffs’ overtime claims for the period between\\nJanuary 1, 2015 and November 12, 2015. See Yamaguchi v.\\nU.S. Dep’t of the Air Force, 109 F.3d 1475, 1482 (9th Cir.\\n1997).\\n\\n The County is not entitled to Eleventh Amendment\\n immunity.\\n\\n Plaintiffs first argue that Eleventh Amendment\\nimmunity is never available to counties. The County argues\\nthat it enjoys Eleventh Amendment immunity when acting\\nas an “arm of the State.”\\n\\n Federal courts have long declined to extend Eleventh\\nAmendment immunity to counties. 4 Indeed, the Supreme\\n\\n 4\\n See, e.g., Lake Country Estates, Inc. v. Tahoe Reg’l Planning\\nAgency, 440 U.S. 391, 401 (1979) (“[T]he Court has consistently refused\\nto construe the Amendment to afford protection to political subdivisions\\nsuch as counties and municipalities, even though such entities exercise a\\nslice of state power.” (internal quotation marks omitted)); Lincoln Cty.\\nv. Luning, 133 U.S. 529, 530 (1890) (holding that the Eleventh\\nAmendment does not bar a suit against a county, though the principle\\nadvanced has changed over time); Del Campo v. Kennedy, 517 F.3d\\n1070, 1075–76 (9th Cir. 2008) (“State sovereign immunity . . . does not\\nextend to counties and similar municipal corporations, even though they\\nshare some portion of state power.” (internal quotation marks omitted)\\n(quoting Mt. Healthy City Sch. Dist. Bd. of Educ. v. Doyle, 429 U.S. 274,\\n280 (1977))).\\n\\f12 RAY V. COUNTY OF LOS ANGELES\\n\\nCourt once said that Eleventh Amendment immunity does\\nnot extend to municipal corporations. Mt. Healthy, 429 U.S.\\nat 280. But thirty years later, the Supreme Court suggested\\nthat it was at least possible for a county to receive Eleventh\\nAmendment immunity. In Northern Insurance Company of\\nNew York v. Chatham County, 547 U.S. 189, 190 (2006),\\nwhich involved a county-operated drawbridge, the Court\\nstated that a county might be entitled to Eleventh\\nAmendment immunity if it were “acting as an arm of the\\nState, as delineated by this Court’s precedents, in operating\\nthe drawbridge.” 5\\n\\n The Court cited several cases for this proposition. First,\\nAlden v. Maine: “The second important limit to the principle\\nof sovereign immunity is that it bars suits against States but\\nnot lesser entities. The immunity does not extend to suits\\nprosecuted against a municipal corporation or other\\ngovernmental entity which is not an arm of the State.”\\n527 U.S. 706, 756 (1999). This sentence means one of two\\nthings: either (1) that Eleventh Amendment immunity does\\nnot extend to municipal corporations because they are not\\narms of the state or (2) that Eleventh Amendment immunity\\ndoes not extend to a municipal corporation unless it is acting,\\nin a particular circumstance, as an arm of the state. Alden in\\nturn cites Mt. Healthy, in which the Court considered\\nwhether “the Mt. Healthy Board of Education is to be treated\\nas an arm of the State partaking of the State’s Eleventh\\n\\n 5\\n At least one circuit has relied on this language and held that\\ncounties might be entitled to Eleventh Amendment immunity. See\\nFuesting v. Lafayette Par. Bayou Vermilion Dist., 470 F.3d 576, 579 (5th\\nCir. 2006) (“[A] municipality can be immune from suit if it was ‘acting\\nas an arm of the State, as delineated by [the Supreme] Court’s\\nprecedent’” (alteration in original) (quoting Chatham, 547 U.S. at 194)).\\nBut, to our knowledge, no court has ever actually extended Eleventh\\nAmendment immunity to a county.\\n\\f RAY V. COUNTY OF LOS ANGELES 13\\n\\nAmendment immunity, or is instead to be treated as a\\nmunicipal corporation or other political subdivision to which\\nthe Eleventh Amendment does not extend.” Mt. Healthy,\\n429 U.S. at 280. That citation suggests the former reading.\\n\\n The Chatham Court also cited Lake Country Estates, but\\nwhile that case noted that “some agencies exercising state\\npower have been permitted to invoke the [Eleventh]\\nAmendment in order to protect the state treasury from\\nliability that would have had essentially the same practical\\nconsequences as a judgment against the State itself,” it also\\nstated that “the Court has consistently refused to construe the\\nAmendment to afford protection to political subdivisions\\nsuch as counties and municipalities, even though such\\nentities exercise a ‘slice of state power.’” Lake Country\\nEstates, 440 U.S. at 400–01. Although these passages seem\\nto support Plaintiffs’ argument that counties never enjoy\\nEleventh Amendment immunity, it is not for us to clarify\\nChatham’s apparently contrary statement.\\n\\n The Chatham Court ultimately found it dispositive that\\nthe County there had conceded below that it had no Eleventh\\nAmendment immunity and that the question on which\\ncertiorari was granted assumed that conclusion. Given that\\nthe Supreme Court appears to have left open the possibility\\nthat a county could be entitled to Eleventh Amendment\\nimmunity in some cases, we decline to hold to the contrary.\\nWe therefore assume without deciding that, consistent with\\nthe Court’s language in Chatham, a county might be entitled\\nto Eleventh Amendment immunity if acting as an arm of the\\nstate.\\n\\n The County is not an arm of the State here.\\n\\n In Mitchell, we set out five factors for determining\\nwhether a government entity is an arm of its state for\\n\\f14 RAY V. COUNTY OF LOS ANGELES\\n\\nEleventh Amendment immunity purposes: (1) “whether a\\nmoney judgment would be satisfied out of state funds”;\\n(2) “whether the entity performs central governmental\\nfunctions”; (3) “whether the entity may sue or be sued”;\\n(4) “whether the entity has the power to take property in its\\nown name or only the name of the state”; and (5) “the\\ncorporate status of the entity.” 861 F.2d at 201. “To\\ndetermine these factors, the court looks to the way state law\\ntreats the entity.” Id.\\n\\n a. First Mitchell factor\\n\\n “The first Mitchell factor—whether a money judgment\\n. . . would be satisfied out of state funds—is the most\\nimportant.” Sato v. Orange Cty. Dep’t of Educ., 861 F.3d\\n923, 929 (9th Cir. 2017); see also Beentjes v. Placer Cty. Air\\nPollution Control Dist., 397 F.3d 775, 785 (9th Cir. 2005)\\n(noting that the first Mitchell factor is “the one given the\\nmost weight”). The County conceded, both below and on\\nappeal, that it cannot show that a money judgment would be\\npaid directly with State funds. 6 Thus, this factor weighs\\nagainst Eleventh Amendment immunity.\\n\\n b. Second Mitchell factor\\n\\n As to the second Mitchell factor—whether the County\\nperforms central governmental functions—we must\\ndetermine whether the County addresses “a matter of\\nstatewide rather than local or municipal concern, and the\\nextent to which the state exercises centralized governmental\\ncontrol over the entity.” Beentjes, 397 F.3d at 782 (internal\\n 6\\n The parties discuss at length how the County and the State allocate\\nthe costs of the program, but that is not relevant—what matters is who\\nwould be responsible for satisfying a money judgment against the\\nCounty, not who pays for the program.\\n\\f RAY V. COUNTY OF LOS ANGELES 15\\n\\nquotation marks omitted) (first quoting Belanger v. Madera\\nUnified Sch. Dist., 963 F.2d 248, 253 (9th Cir. 1992); then\\nquoting Savage v. Glendale Union High Sch., Dist. No. 205,\\nMaricopa Cty., 343 F.3d 1036, 1044 (9th Cir. 2003)).\\n\\n To begin, it is unclear whether the second Mitchell factor\\nconcerns whether the County performs central government\\nfunctions in general or whether the County performs central\\ngovernment functions in carrying out the particular function\\nat issue—here implementing the IHSS program.\\n\\n As the district court correctly noted, the closest analogue\\nin our case law is Streit v. County of Los Angeles, 236 F.3d\\n552 (9th Cir. 2001). There, the Los Angeles County\\nSheriff’s Department (LASD) would check its systems,\\nbefore releasing a prisoner, to see if the prisoner was wanted\\nby another law enforcement agency. Id. at 556. This\\nextended the period of incarceration one or two days past the\\nprisoners’ release dates. Id. The plaintiffs alleged that the\\nCounty delayed their release during these checks, in\\nviolation of their civil rights. Id. The LASD argued that\\nbecause it was an arm of the state, it was not a “person” that\\ncould be liable for damages under § 1983. Id. at 557.\\n\\n We looked at the LASD’s performance of the particular\\nfunction at issue—implementing the pre-release policy—not\\nthe LASD’s general function as a sheriff’s department. See\\nid. at 567. We held that “conducting the AJIS checks is not\\na central government function.” Id. (emphasis added).\\nThus, it appears from Streit that we look to whether the\\nCounty, in performing the particular function at issue,\\nperforms a central government function. This fits with the\\nCourt’s statement in Chatham that the county there might\\nhave been entitled to Eleventh Amendment immunity if it\\nwere “acting as an arm of the State, as delineated by this\\n\\f16 RAY V. COUNTY OF LOS ANGELES\\n\\nCourt’s precedents, in operating [a] drawbridge.”\\nChatham, 547 U.S. at 194 (emphasis added).\\n\\n i. A matter of statewide rather than local or\\n municipal concern\\n\\n The in-home care of the elderly and disabled is a matter\\nof both statewide and local concern. Plaintiffs are residents\\nof California, and the IHSS program is a statewide program\\nimplemented through State legislation that provides care to\\nhundreds of thousands of California residents. But Plaintiffs\\nare also, of course, residents of Los Angeles County, and the\\nCounty has an interest in the program and the care provided\\nin Los Angeles.\\n\\n ii. The extent to which the state exercises\\n centralized governmental control over the\\n entity\\n\\n Here we consider the extent to which the County, in\\nimplementing the program, has “discretionary powers” and\\n“substantial autonomy in carrying out [its] duties.” Beentjes,\\n397 F.3d at 783.\\n\\n The County may negotiate, implement, and pay for pay\\nraises. See Cal. Welf. & Inst. Code § 12306.1. The County\\nmay also allow its providers to exceed the maximum number\\nof hours that the CDSS has set. See id. § 12300.4(d)(3).\\nThus, the County has discretion to make some important\\nchoices on its own.\\n\\n But the County contends—and Plaintiffs do not\\ndispute—that it has no discretion over the action (or\\ninaction) that subjected it to potential liability here: payment\\nof overtime wages under the FLSA. In taking the actions\\nthat have subjected it to potential liability, the County had\\n\\f RAY V. COUNTY OF LOS ANGELES 17\\n\\nneither “discretionary powers” nor “substantial autonomy”\\nin carrying out its duties.\\n\\n We think this clearly tips the scales in the County’s favor\\nas to this factor. The County had no choice in the matter of\\nthe overtime wages, as the State mandated the payment start\\ndate. We therefore hold that the second Mitchell factor\\nfavors Eleventh Amendment immunity.\\n\\n c. Third, fourth, and fifth Mitchell factors\\n\\n The County does not dispute that it can sue and be sued\\n(third Mitchell factor), that it has the power to take property\\nin its own name (fourth Mitchell factor), or that it has an\\nindependent corporate status 7 separate from the State (fifth\\nMitchell factor). Thus, these three Mitchell factors weigh\\nagainst Eleventh Amendment immunity.\\n\\n * * *\\n\\n In sum, the first Mitchell factor is the most important,\\nand it weighs against Eleventh Amendment immunity. So\\ndo the third, fourth, and fifth Mitchell factors. Only the\\nsecond factor favors immunity. We therefore hold that,\\nunder Mitchell, the County is not an arm of the State when it\\n\\n\\n\\n\\n 7\\n The fifth Mitchell factor asks whether the entity has “independent\\ncorporate status,” Holz v. Nenana City Pub. Sch. Dist., 347 F.3d 1176,\\n1188 (9th Cir. 2003), or is, instead, merely an agency of the state without\\nan identity that is separate from the state, Beentjes, 397 F.3d at 785. Here\\nthe County does not dispute its independent corporate status, as the\\nSupreme Court has already held that California counties have\\nindependent corporate status and are not agents of the State of California.\\nSee Moor v. Alameda Cty., 411 U.S. 693, 719 (1973).\\n\\f18 RAY V. COUNTY OF LOS ANGELES\\n\\nadministers the IHSS program, and thus it has no Eleventh\\nAmendment immunity barring this action.\\n\\n The Supreme Court has not overruled or\\n undermined Mitchell.\\n\\n The County argues that we should overrule Mitchell\\nbecause a later Supreme Court case, Hess v. Port Authority\\nTrans-Hudson Corporation, 513 U.S. 30 (1994),\\nundermined it. As a three-judge panel, if we find that\\nintervening Supreme Court authority is clearly\\nirreconcilable with our own precedent, we must consider\\nourselves bound by the intervening higher authority and\\nconsider our precedent effectively overruled. See Miller v.\\nGammie, 335 F.3d 889, 900 (9th Cir. 2003). Because Hess\\nis not clearly irreconcilable with Mitchell, we reject the\\nCounty’s argument.\\n\\n In Hess, the Court held that a Congressionally approved\\nbistate entity—the Port Authority Trans-Hudson\\nCorporation (PATH), created to improve coordination of the\\n“terminal, transportation and other facilities of commerce in,\\nabout and through the port of New York”—did not have\\nEleventh Amendment immunity. 513 U.S at 35, 52–53\\n(citation omitted). The County argues that Hess established\\n“indicators of immunity” that undermine the Mitchell test.\\nWe disagree.\\n\\n The Hess Court noted that “current Eleventh\\nAmendment jurisprudence emphasizes the integrity retained\\nby each State in our federal system.” Id. at 39. The Court\\nthen emphasized the difference between PATH and the\\nStates of the Union: “The States, as separate sovereigns, are\\nthe constituent elements of the Union. Bistate entities, in\\ncontrast, typically are creations of three discrete sovereigns:\\ntwo States and the Federal Government.” Id. at 40.\\n\\f RAY V. COUNTY OF LOS ANGELES 19\\n\\n The Court stated that “[p]ointing away from Eleventh\\nAmendment immunity, the States lack financial\\nresponsibility” for the bistate entity. Id. at 45. Here,\\nCalifornia similarly lacks financial responsibility for the\\nCounty generally, but Plaintiffs allege that although\\nCalifornia writes their checks, the County pays a share of\\ntheir wages and sets their hours of work.\\n\\n In Hess, “indicators of immunity point[ed] in different\\ndirections.” Id. at 47. Perhaps they do here as well. Los\\nAngeles is not a constituent member of the Union, but it\\nacted at the direction of the State and had no authority over\\nthe payments at issue. But when faced with a different\\ndichotomy in Hess, the Court emphasized that the most\\nimportant factor was whether judgments against PATH\\nwould be paid by the State: “the vulnerability of the State’s\\npurse [is] the most salient factor in Eleventh Amendment\\ndeterminations.” Id. at 48; see also id. at 48–49 (citing cases\\nfor the “prevailing view” that the state-treasury factor is\\n“generally accorded . . . dispositive weight”); id. at 51\\n(stating that “the Eleventh Amendment’s core concern is not\\nimplicated” if the State is not “in fact obligated to bear and\\npay the . . . indebtedness of the enterprise”). 8\\n\\n\\n\\n\\n 8\\n The dissent read the holding even more broadly:\\n\\n In place of the various factors recognized in Lake\\n Country Estates, Inc. v. Tahoe Regional Planning\\n Agency, 440 U.S. 391, 99 S. Ct. 1171, 59 L.Ed.2d 401\\n (1979), for determining arm-of-the-state status, we\\n may now substitute a single overriding criterion,\\n vulnerability of the state treasury. If a State does not\\n fund judgments against an entity, that entity is not\\n\\f20 RAY V. COUNTY OF LOS ANGELES\\n\\n After noting that the bistate entity “was financially self-\\nsufficient,” generated “its own revenues,” and paid “its own\\ndebts,” the Court held that “[r]equiring the [bistate entity] to\\nanswer in federal court to injured railroad workers who\\nassert a federal statutory right, under the FELA, to recover\\ndamages does not touch the concerns—the States’ solvency\\nand dignity—that underpin the Eleventh Amendment.” Id.\\nat 52. The same is true here. Mitchell and Hess both\\nemphasize the state-treasury factor. Hess thus fully supports\\nand does not undermine Mitchell. 9\\n\\n The County argues that Hess emphasized the amount of\\ncontrol that a state maintains over an entity, a factor\\nsupposedly not mentioned in Mitchell and one that,\\naccording to the County, favors Eleventh Amendment\\nimmunity here. First, as we mentioned above, the second\\nMitchell factor does include a “control” inquiry—it just\\ndoesn’t make that factor dispositive. In addition, Hess\\npointed out that “[g]auging actual control . . . can be a\\n‘perilous inquiry,’ [and] ‘an uncertain and unreliable\\nexercise.’” 513 U.S. at 47 (quoting Note, 92 Colum. L. Rev.\\n1243, 1284 (1992)). The Court therefore doubted not only\\nthe efficacy but also the utility of a “control” analysis, and it\\n\\n\\n\\n within the ambit of the Eleventh Amendment, and\\n suits in federal court may proceed unimpeded.\\n\\nId. at 55 (O’Connor, J., dissenting).\\n 9\\n Los Angeles makes a legitimate point about the unfairness of the\\nresult here. But that unfairness springs from the State and its\\nimplementing legislation, not the Eleventh Amendment. Los Angeles\\nmust air its grievance, if at all, in Sacramento.\\n\\f RAY V. COUNTY OF LOS ANGELES 21\\n\\ndid not suggest that control was a favored, much less\\ndispositive, factor in the Eleventh Amendment analysis. 10\\n\\n Hess clearly stated that “rendering control dispositive\\ndoes not home in on the impetus for the Eleventh\\nAmendment: the prevention of federal-court judgments that\\nmust be paid out of a State’s treasury.” Id. at 48. And, in\\nspecifically discussing the control factor, the Court noted\\nthat even though “‘political subdivisions exist solely at the\\nwhim and behest of their State,’ . . . cities and counties do\\nnot enjoy Eleventh Amendment immunity.” Id. at 47\\n(quoting Port Auth. Trans-Hudson Corp. v. Feeney, 495 U.S.\\n299, 313 (1990)).\\n\\n Finally, the County insists that Hess compels us to\\nconsider the State’s dignity, a factor not mentioned in\\nMitchell. Hess noted that the State’s “solvency and dignity\\n. . . underpin the Eleventh Amendment.” 513 U.S. at 52.\\nThat is undoubtedly true. But the State is no longer a party\\nto this action, and it will not be responsible for an adverse\\njudgment against the County. Allowing this action against\\nLos Angeles does not injure California’s dignity. 11\\n\\n\\n\\n 10\\n The control discussed in Hess seems to have gone to overall\\ncontrol over the entity, not just control within the context of the particular\\nfunction at issue: “PATH urges that we find good reason to classify the\\nPort Authority as a state agency for Eleventh Amendment purposes\\nbased on the control New York and New Jersey wield over the\\nAuthority. . . . But ultimate control of every state-created entity resides\\nwith the State, for the State may destroy or reshape any unit it creates.”\\nId. at 47. Thus, looking at the State’s overall control over the County as\\na county would not help the County’s position here.\\n 11\\n And, although it would not have altered our analysis, we note that\\nCalifornia has not sought to file an amicus brief (below or on appeal)\\n\\f22 RAY V. COUNTY OF LOS ANGELES\\n\\n The Supreme Court decided Hess about five years after\\nwe decided Mitchell. And although Hess arose in a different\\ncontext than Mitchell-Hess addressed a bistate entity, not a\\ncounty—nothing in Hess so undermines Mitchell that we\\nhave the power to overrule it. More importantly, even if we\\nused Hess rather than Mitchell to guide our analysis, we\\nwould reach the same result.\\n\\n When a non-state entity invokes Eleventh Amendment\\nimmunity, the most important factor for determining\\nwhether the entity is an arm of the state remains the state-\\ntreasury factor—that is, whether the state will be liable for a\\nmoney judgment against the non-state entity. That factor,\\nand all but one of the other Mitchell factors, dictates the\\nresult here. The Eleventh Amendment does not bar\\nPlaintiffs’ suit against Los Angeles.\\n\\n The effective date of the rule is January 1, 2015.\\n\\n We next consider whether the effective date of the rule\\nis the original effective date of January 1, 2015 or some date\\nafter the D.C. Circuit reversed the district court’s vacatur.\\nThe County argues that the rule cannot have an effective date\\nthat is earlier than the date on which the D.C. Circuit\\nreversed the district court’s vacatur. Plaintiffs argue that the\\nlegal effect of the vacatur is to reinstate the original January\\n1, 2015 effective date. We agree with Plaintiffs and hold\\nthat the effective date of the rule is January 1, 2015.\\n\\n\\n\\n\\narguing either that the County is entitled to Eleventh Amendment\\nimmunity or that this case threatens California’s dignity.\\n\\f RAY V. COUNTY OF LOS ANGELES 23\\n\\n A January 1, 2015 effective date is not\\n impermissibly retroactive.\\n\\n The County argues that a January 1, 2015 effective date\\nis impermissibly retroactive. Plaintiffs argue that the D.C.\\nCircuit’s decision, not the rule, applies retroactively, because\\nthe D.C. Circuit was “explaining what the law always was,”\\nand thus reinstating the original effective date is merely a\\nreturn to the status quo ante.\\n\\n When an appellate court applies “a rule of federal law to\\nthe parties before it,” that interpretation “must be given full\\nretroactive effect in all cases still open on direct review and\\nas to all events, regardless of whether such events predate or\\npostdate [the] announcement of the rule.” Harper v. Va.\\nDep’t of Taxation, 509 U.S. 86, 97 (1993). That is because\\n“when a court delivers a ruling, even if it is unforeseen, the\\nlaw has not changed. Rather, the court is explaining what\\nthe law always was.” Jones Stevedoring Co. v. Dir., Office\\nof Workers’ Comp. Programs, 133 F.3d 683, 688 (9th Cir.\\n1997).\\n\\n When the D.C. Circuit held that the DOL had the\\nrulemaking authority to promulgate the new rule and that its\\nnew rule was a reasonable exercise of that authority, see Weil\\nII, 799 F.3d at 1090, it did not change the law but merely\\nexplained what the law always was—the district court’s\\nerroneous contrary holding notwithstanding.\\n\\n Two cases support our holding. In GTE South, Inc. v.\\nMorrison, 199 F.3d 733, 738, 740 (4th Cir. 1999), the Fourth\\nCircuit addressed an issue much like the one we face:\\ndetermining the effective date of certain pricing rules,\\npromulgated by the FCC, that the Eighth Circuit stayed and\\nthen vacated before their effective date. The Supreme Court\\nlater reversed the Eighth Circuit. See AT & T Corp. v. Iowa\\n\\f24 RAY V. COUNTY OF LOS ANGELES\\n\\nUtilities Bd., 525 U.S. 366, 385 (1999). The Morrison panel\\nheld that “the Supreme Court’s determination that the FCC\\nhas jurisdiction to issue pricing rules would appear to\\ncompel the conclusion that the FCC always had such\\njurisdiction and that the rules apply as of the effective date\\noriginally scheduled.” 199 F.3d at 740 (emphasis added).\\nThe Fourth Circuit emphasized that its holding was not\\nunfair to the parties who argued for a later effective date\\nbecause they had “ample notice” of the original effective\\ndate and “surely knew that the FCC’s authority to issue\\npricing rules might ultimately be upheld by the Supreme\\nCourt.” Id. at 741.\\n\\n In US West Communication, Inc. v. Jennings, 304 F.3d\\n950, 955 (9th Cir. 2002), we considered a similar question:\\nwhether the regulations that the Fourth Circuit considered in\\nMorrison applied to conduct that occurred during the period\\nof vacatur. Finding the Fourth Circuit’s reasoning in\\nMorrison persuasive and applicable, we noted that the\\nSupreme Court’s determination that the regulations were\\nvalid meant that we should apply them “to all . . . agreements\\narbitrated under the Act, including agreements arbitrated\\nbefore the rules were reinstated.” Id. at 957 (emphasis\\nadded). Relying on Morrison, we held that applying the\\nreinstated regulations to conduct that occurred during the\\nperiod of vacatur would not give the regulations an\\nimpermissible retroactive effect. Id. at 958.\\n\\n Morrison and Jennings are analogous to this case\\nbecause both involved determining how to apply rules or\\nregulations that were vacated but ultimately reinstated on\\nappeal. Indeed, Morrison commented not only on the\\nretroactivity of the Supreme Court’s reversal but also on the\\neffective date of the regulations, holding that the intervening\\n\\f RAY V. COUNTY OF LOS ANGELES 25\\n\\nvacatur did not alter the original effective date of the pricing\\nrules. 199 F.3d at 740.\\n\\n Thus, Morrison and Jennings guide our analysis here.\\nThe D.C. Circuit’s holding that the DOL had the authority to\\npromulgate the new rule and that the rule was reasonable\\napplies retroactively. As in Jennings, the regulations apply\\nas of the original effective date. To hold otherwise could\\nencourage dilatory appellate litigation. If an erroneously\\nvacated rule or regulation were not effective until sometime\\nafter the mandate issued in a later appeal, then a party might\\ndrag out the appellate process to avoid compliance for as\\nlong as possible. Put differently, an erroneous vacatur\\ncannot postpone a rule’s effective date until an appellate\\ncourt corrects the error sometime in the future. And, as the\\nFourth Circuit noted in Morrison, in a case like this everyone\\nknows that the lower court decision might be reversed on\\nappeal.\\n\\n The State and its counties knew from October 13, 2013,\\nwhen the DOL first announced its final rule, that January 1,\\n2015 was the rule’s effective date. See Application of the\\nFair Labor Standards Act to Domestic Service, 78 Fed. Reg.\\nat 60,454. The State and its counties had a full fifteen\\nmonths to comply with the final rule—indeed the State\\ninitially said that it would comply with the original effective\\ndate, but it changed course after the Weil I court vacated the\\nrule. That decision may have been reasonable, but it created\\na monetary risk, as the State and its counties were well aware\\nthat an appellate court might uphold the regulations on\\nappeal.\\n\\n The district court held that to apply the Weil II decision\\nretroactively would be to “reinstate[] an earlier and judicially\\nvacated effective date and retroactively creat[e] liability for\\nviolations of the reinstated regulations as if the District\\n\\f26 RAY V. COUNTY OF LOS ANGELES\\n\\nCourt’s vacation of the regulations had never occurred.”\\nThat is exactly correct. And although the district court found\\nthat to be unfair, it would be equally unfair to hold that a\\nputative collective of homecare providers is not entitled to\\nnearly a year’s worth of overtime wages just because a single\\ndistrict court issued an erroneous decision that another court\\nreversed on appeal. The State gambled that Weil I would be\\naffirmed. The effect of that gamble might be unfair to the\\nCounty, but the County must seek any recourse from the\\nState. It is not fair for the homecare providers to bear the\\nfinancial consequences of the State’s calculated risk.\\n\\n The DOL’s decision not to enforce a new rule does\\n not obviate private rights of action.\\n\\n According to the County, the DOL’s choice against\\nenforcing the rule until November 12, 2015 eliminated the\\navailability of private rights of action until that date because\\na private right of action cannot precede an agency’s\\nenforcement of a rule or regulation. We disagree.\\n\\n “An agency’s informal assurance that it will not pursue\\nenforcement cannot preclude a citizen’s suit to do so.” Ohio\\nValley Envtl. Coal. v. Fola Coal Co., LLC, 845 F.3d 133,\\n145 (4th Cir. 2017) (emphasis added). Congress created a\\nprivate right of action under the FLSA for unpaid overtime:\\n“Any employer who violates the provisions of section 206\\nor section 207 of this title shall be liable to the employee or\\nemployees affected in the amount of their . . . unpaid\\novertime compensation . . . .” 29 U.S.C. § 216(b). An\\nagency’s discretionary decision to hold off enforcement does\\nnot and cannot strip private parties of their rights to do so.\\nSee Ohio Valley, 845 F.3d at 145 (“Congress enacted the\\ncitizen suit provision of the Clean Water Act to address\\nsituations, like the one at hand, in which the traditional\\nenforcement agency declines to act.”).\\n\\f RAY V. COUNTY OF LOS ANGELES 27\\n\\n The district court’s hypothesis that the D.C. Circuit and\\nDOL “intended” that the regulation become effective “no\\nearlier than November 12, 2015” is tenuous and, in any\\nevent, irrelevant. First, the D.C. Circuit said nothing at all\\non the issue. Second, there is nothing in the several\\nstatements of the DOL, which the district court relied on, that\\nsuggest that it intended its discretionary enforcement choices\\nto preclude private enforcement. Indeed, other than by\\namending the rule, the DOL could not have precluded\\nprivate enforcement even if it wanted to.\\n\\n The rule’s original effective date remains January 1,\\n2015. If the DOL “intended” for the effective date be\\nsomething other than January 1, 2015, the DOL could have\\nsought to change that effective date through the procedures\\nset out in the Administrative Procedure Act. Were we to\\nhold to the contrary and impose our view that the DOL’s\\nexercise of discretion amended the effective date sub\\nsilentio, we would in fact be usurping the rulemaking\\nauthority of the DOL. See Nat. Res. Def. Council, Inc. v.\\nU.S. E.P.A., 683 F.2d 752, 762 (3d Cir. 1982) (holding that\\na final rule’s effective date is an “essential part” of that rule\\nand is thus subject to the rulemaking procedures of the\\nAPA).\\n\\n The effective date of the rule is January 1, 2015. 12\\n\\n\\n 12\\n Although some district courts have reached a different\\nconclusion—see, e.g., Bangoy v. Total Homecare Solutions, LLC, No.\\n1:15-CV-573, 2015 WL 12672727, at *3 (S.D. Ohio Dec. 21, 2015)\\n(holding that the plaintiffs failed to state a claim for a violation of the\\nFLSA between January 1, 2015 and “late August 2015”)—nearly all of\\nthem have reached the same result we reach here, see, e.g., Kinkead v.\\nHumana, Inc., 206 F. Supp. 3d 751, 752 (D. Conn. 2016) (holding that\\nthe effective date of the rule is January 1, 2015, “the effective date set\\n\\f28 RAY V. COUNTY OF LOS ANGELES\\n\\n CONCLUSION\\n\\n We AFFIRM the district court’s holding that the County\\nis not entitled to Eleventh Amendment immunity and\\nREVERSE the district court’s holding that the putative\\ncollective period began on November 12, 2015, holding\\ninstead that the rule’s effective date—and thus the beginning\\nof the putative collective period—is January 1, 2015. We\\nREMAND for proceedings consistent with this opinion.\\nCosts shall be awarded to Plaintiffs-Appellants.\\n\\n\\n\\n\\nforth by the agency”); Collins v. DKL Ventures, LLC, 215 F. Supp. 3d\\n1059 (D. Colo. 2016) (same); Lewis-Ramsey v. Evangelical Lutheran\\nGood Samaritan Soc’y, 215 F. Supp. 3d 805 (S.D. Iowa 2016) (same).\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4092,"cells":{"id":{"kind":"number","value":4654960,"string":"4,654,960"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Fidelity National Financial v. Colin Friedman"},"decision_date":{"kind":"string","value":"2019-08-22"},"docket_number":{"kind":"string","value":"17-15913"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: A. Wallace Tashima and Mary H. Murguia, Circuit Judges, and Robert N. Chatigny,* District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"TASHIMA, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nFIDELITY NATIONAL FINANCIAL, No. 17-15913\\nINC., a Delaware corporation;\\nFIDELITY EXPRESS NETWORK, INC., a D.C. No.\\nCalifornia corporation, CV 15-2288 DJH\\n Plaintiffs-Appellants,\\n\\n v. OPINION\\n\\nCOLIN H. FRIEDMAN, individually\\nand as trustee of Friedman Family\\ntrust UDT Dated 7/23/87; HEDY\\nKRAMER FRIEDMAN, individually\\nand as trustee of Friedman Family\\ntrust UDT Dated 7/23/87; FARID\\nMESHKATAI, an individual; ANITA\\nKRAMER MESHKATAI, individually\\nand as trustee of Anita Kramer\\nLiving Trust Dated 7/23/87,\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Diane J. Humetewa, District Judge, Presiding\\n\\n Argued and Submitted October 12, 2018\\n San Francisco, California\\n\\n Filed August 22, 2019\\n\\f2 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n Before: A. Wallace Tashima and Mary H. Murguia,\\n Circuit Judges, and Robert N. Chatigny,* District Judge.\\n\\n Opinion by Judge Tashima\\n\\n\\n SUMMARY**\\n\\n\\n Registration of Judgments\\n\\n The panel reversed the district court’s order vacating a\\nregistered judgment and remanded.\\n\\n Plaintiffs obtained a civil fraud judgment in California\\nfederal court and registered this California judgment in the\\nDistrict of Arizona pursuant to 28 U.S.C. § 1963. Plaintiffs\\nlater attempted to renew the Arizona judgment, but the\\ndistrict court ruled that the renewal or re-registration was void\\nas untimely. Plaintiffs next registered the California\\njudgment in Washington federal court and then registered the\\nnewly-obtained Washington judgment in the District of\\nArizona. The district court granted defendants’ motion under\\nFed. R. Civ. P. 60(b) to vacate the newly-registered second\\nArizona judgment as void. In a prior appeal, the court of\\nappeals reversed, holding that registering the California\\njudgment in Washington created a “new” Washington\\njudgment that could be re-registered in another state under\\n\\n\\n *\\n The Honorable Robert N. Chatigny, United States District Judge for\\nthe District of Connecticut, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 3\\n\\n§ 1963. On remand, the district court again granted relief\\nunder Rule 60(b) on the ground that the Washington\\njudgment was void because the Western District of\\nWashington lacked personal jurisdiction over the defendants\\nat the time of registration.\\n\\n The panel reversed, holding that a court need not have\\npersonal jurisdiction over a judgment debtor in order to\\n“merely register” a previously obtained judgment pursuant to\\n§ 1963. The panel held that § 1963 itself does not require that\\nthe court have personal jurisdiction. Further, the Due Process\\nClause does not require that a court in which a judgment\\ncreditor registers a pre-existing federal judgment have\\npersonal jurisdiction over judgment debtors at the time of\\nregistration. The panel held that once a federal court of\\ncompetent jurisdiction has determined the parties’ substantive\\nrights and entered a judgment following a proceeding that\\ncomports with due process, that federal judgment should be\\nenforceable in any other federal district by way of the federal\\njudgment registration statute.\\n\\n\\n COUNSEL\\n\\nThomas H. Case (argued) and Michael G. King, Hennelly &\\nGrossfeld LLP, Marina del Rey, California, for Plaintiffs-\\nAppellants.\\n\\nDavid M. Bass (argued), David M. Bass & Associates Inc.,\\nLos Angeles, California; Dominica J. Minore, The Law\\nOffices of Dominica J. Minore P.C., Scottsdale Arizona; for\\nDefendants-Appellees.\\n\\f4 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n OPINION\\n\\nTASHIMA, Circuit Judge:\\n\\n In order to facilitate the enforcement of federal\\njudgments, 28 U.S.C. § 1963 provides that a judgment\\nentered in a federal court may be registered in any other\\nfederal district by “filing a certified copy of the judgment” in\\nthat district. In this case we address, as a matter of first\\nimpression in our Circuit, whether personal jurisdiction over\\nthe judgment debtors in the district of registration is required\\nfor such registration of a judgment. We hold that it is not,\\nbecause neither § 1963 nor due process imposes such a\\npersonal jurisdiction requirement. We therefore reverse the\\norder and judgment of the district court, and remand.\\n\\n BACKGROUND\\n\\n In 2002, Plaintiffs-Appellants Fidelity National Financial,\\nInc., and Fidelity Express Network, Inc. (collectively,\\n“Fidelity”), obtained a multimillion dollar civil fraud\\njudgment (the “California Judgment”) against Defendants-\\nAppellees the Friedmans and Meshkatais (collectively,\\n“Defendants”) in the U.S. District Court for the Central\\nDistrict of California. This judgment became final on May\\n15, 2003, after this Court dismissed Defendants’ appeal from\\nthe judgment.\\n\\n While Defendants’ appeal in the original case was\\npending, Fidelity registered the California Judgment in the\\nDistrict of Arizona pursuant to the federal registration statute,\\n28 U.S.C. § 1963. In 2007, Fidelity attempted to renew the\\nArizona judgment. However, on March 2, 2012, the District\\nCourt of Arizona ruled that Fidelity’s 2007 renewal or re-\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 5\\n\\nregistration of the Arizona registered judgment were void as\\nuntimely, because the judgment had already expired under\\nArizona’s five-year statute of limitations for the enforcement\\nof judgments.\\n\\n Unable to enforce the Arizona registered judgment or re-\\nregister the original California Judgment in Arizona, Fidelity\\ncame up with a creative alternative. Fidelity registered the\\nCalifornia Judgment in the Western District of Washington\\n(the “Washington Judgment”) Fidelity then registered the\\nnewly-obtained Washington Judgment in the District of\\nArizona (the “Second Arizona Judgment”).\\n\\n Several months later, Defendants moved the Arizona\\nDistrict Court under Federal Rule of Civil Procedure 60(b) to\\nvacate the newly-registered Second Arizona Judgment as\\nvoid, arguing that § 1963 did not allow successive registration\\nof federal judgments. That is, Defendants argued that\\nregistering the California Judgment in Washington did not\\ncreate a new Washington judgment that could then be\\nregistered in Arizona under § 1963. Defendants also argued\\nthat, in any case, the Washington Judgment was invalid for\\nlack of personal jurisdiction over Defendants. The Arizona\\ndistrict court granted Defendants’ motion and vacated the\\nSecond Arizona Judgment, holding that § 1963 did not allow\\nsuccessive registration of judgments—in other words, that\\nonly an original judgment, such as the California Judgment in\\nthis case, may be registered in another district under § 1963.\\nSee Fid. Nat’l Fin., Inc. v. Friedman, 939 F. Supp. 2d 974,\\n979–87 (D. Ariz. 2013). The district court did not reach\\nDefendants’ second argument regarding lack of due process\\nand personal jurisdiction. Id. at 986–87.\\n\\f6 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n Fidelity appealed, and this Court reversed. Fid. Nat’l\\nFin., Inc. v. Friedman, 803 F.3d 999 (9th Cir. 2015)\\n(“Fidelity I”). We held that registering the California\\nJudgment in Washington created a “new” Washington\\njudgment that, like any other Washington judgment, could be\\nre-registered in another state under the plain terms of § 1963.\\nId. at 1003. We remanded the case to the Arizona district\\ncourt on that basis, without reaching Defendants’ alternative\\nargument that the Washington Judgment was void for lack of\\njurisdiction over Defendants. See id. at 1003 n.3.\\n\\n On remand, the district court again granted Defendants’\\nRule 60(b) motion for relief from judgment, after allowing\\nsupplemental briefing on Defendants’ contention that the\\nWashington Judgment was void because the Western District\\nof Washington lacked personal jurisdiction over Defendants\\nat the time of registration. See Fid. Nat’l Fin. Inc. v.\\nFriedman, No. CV-15-2288-PHX-DJH, 2017 WL 6049376\\n(D. Ariz. May 1, 2017). The district court held that\\nregistration of a judgment pursuant to § 1963 requires that the\\ncourt of registration have personal jurisdiction over the\\njudgment debtors. See id. at *6–7. In reaching this\\nconclusion, the district court first noted that a judgment is\\nvoid if the court that “rendered” the judgment lacked\\njurisdiction over the parties. Id. at *5. It then reasoned that\\nbecause we suggested in Fidelity I that registering a judgment\\nunder § 1963 creates a “new” judgment, the court of\\nregistration can be said to have “rendered” a judgment such\\nthat the normal jurisdictional requirements apply. See id. at\\n*5–7. Because Defendants had no assets or other contacts in\\nWashington, the district court concluded that the Washington\\ncourt lacked personal jurisdiction over Defendants;\\nconsequently, that the Washington Judgment was void and\\ncould not have been validly registered in Arizona. Id. at *4,\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 7\\n\\n*7. As a result, the district court vacated the Second Arizona\\nJudgment. Fidelity again appealed.\\n\\n STANDARD OF REVIEW\\n\\n This Court “review[s] de novo . . . a district court’s ruling\\nupon a Rule 60(b)(4) motion . . . because the question of the\\nvalidity of a judgment is a legal one.” Export Grp. v. Reef\\nIndus., Inc., 54 F.3d 1466, 1469 (9th Cir. 1995); see also\\nFidelity I, 803 F.3d at 1001.\\n\\n DISCUSSION\\n\\n On appeal, Fidelity asserts that the district court\\nimproperly granted relief from judgment because a court need\\nnot have personal jurisdiction over a judgment debtor in order\\nto “merely register” a previously obtained judgment pursuant\\nto § 1963. For the reasons explained below, we agree.\\n\\n First, neither the relevant statute’s plain language nor its\\npurpose supports a personal jurisdiction requirement for\\nregistration of a judgment. Section 1963 provides:\\n\\n A judgment in an action for the recovery of\\n money or property entered in any court of\\n appeals, district court, bankruptcy court, or in\\n the Court of International Trade may be\\n registered by filing a certified copy of the\\n judgment in any other district . . . when the\\n judgment has become final by appeal or\\n expiration of the time for appeal or when\\n ordered by the court that entered the judgment\\n for good cause shown. . . . A judgment so\\n registered shall have the same effect as a\\n\\f8 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n judgment of the district court of the district\\n where registered and may be enforced in like\\n manner.\\n\\n28 U.S.C. § 1963. Nothing in this provision limits the district\\ncourts in which a judgment may be registered to only those\\nthat can assert personal jurisdiction over the judgment\\ndebtors. Instead, the provision’s text is extremely broad,\\nallowing a judgment entered in “any . . . district court” to be\\nregistered “in any other district.” Id. Furthermore, the fact\\nthat registering a judgment in another district simply requires\\n“filing a certified copy of the judgment” suggests that the\\nregistration process is intended to be simple, essentially an\\nadministrative task that does not require any additional\\njudicial action. See id.\\n\\n Giving effect to this broad statutory language also accords\\nwith the provision’s purpose, which is “to simplify and\\nfacilitate collection on valid judgments.” Fidelity I, 803 F.3d\\nat 1003 (citation omitted). Section 1963 aims to spare\\ncreditors and debtors “the additional cost and harassment of\\nfurther litigation which would otherwise be required by way\\nof an action on the judgment in a district court other than that\\nwhere the judgment was originally obtained.” S. Rep. No.\\n83-1917 (1954), as reprinted in 1954 U.S.C.C.A.N. 3142; see\\nalso Stanford v. Utley, 341 F. 2d 265, 270 (8th Cir. 1965)\\n(“[T]he purposes of § 1963 were to simplify and facilitate the\\nenforcement of federal judgments, . . . to eliminate the\\nnecessity and expense of a second lawsuit, and to avoid the\\nimpediments, such as diversity of citizenship, which new and\\ndistinct federal litigation might otherwise encounter.”).\\nAdhering to the statutory text’s expansive license to register\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 9\\n\\njudgments in other districts effectuates those purposes;1\\nreading a non-existent jurisdictional requirement into the\\nstatute would contravene Congress’ intent by placing limits\\non registration that would make the process more onerous and\\npotentially require additional litigation regarding\\njurisdiction.2 Thus, we hold that § 1963 itself does not\\n\\n 1\\n Defendants point out that Fidelity’s registration of the California\\nJudgment in Washington was not for the purpose of enforcing the\\njudgment in Washington, but rather for the purpose of creating a new\\njudgment that could subsequently be registered in Arizona and thus\\nprovide an end-run around Arizona’s statute of limitations on the\\nenforcement of judgments. Defendants’ argument that imposing a\\njurisdictional requirement for registration under § 1963 is appropriate\\nsimply because, as applied to the facts of this case, such a requirement\\nwould not contravene the statute’s underlying purpose of facilitating\\nenforcement in the state of registration and would close a seeming\\nloophole, is unpersuasive. In Fidelity I, this Court rejected Defendants’\\ninvitation to “refuse to give effect to Congress’ chosen words” simply\\nbecause doing so “potentially allows plaintiffs to register a judgment that\\nhas previously expired under a state’s statute of limitations,” noting that\\nsuch an effect is “irrelevant in view of the plain language of § 1963.” 803\\nF.3d at 1003.\\n 2\\n For example, a personal jurisdiction requirement could hamper a\\njudgment creditor’s ability to register and enforce a judgment when a\\njudgment debtor has no contacts with a state beyond assets that are\\ncontrolled by a third party in the state. Cf. U.S.I. Props. Corp. v. M.D.\\nConst. Co., 230 F.3d 489, 497 (1st Cir. 2000). In fact, the Supreme Court\\nhas explicitly recognized that in some cases, the “presence of the\\ndefendant’s property in a state . . . alone would not support . . .\\njurisdiction” under the minimum contacts test for personal jurisdiction.\\nShaffer v. Heitner, 433 U.S. 186, 209 (1977). Thus, imposing a minimum\\ncontacts jurisdictional limit on the registration of judgments, as\\nDefendants suggest is required, risks allowing a judgment debtor to “avoid\\npayment of his obligations by the expedient of removing his assets to a\\nplace where he is not subject to an in personam suit.” Id. This would be\\ndirectly at odds with § 1963’s purpose of facilitating collection on valid\\njudgments. See Fidelity I, 803 F.3d at 1003.\\n\\f10 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\nrequire that a court have personal jurisdiction over a\\njudgment debtor in order to register an existing judgment.\\n\\n We next turn to the due process basis of the district\\ncourt’s ruling. The district court noted that under this Court’s\\nruling in Fidelity I, registering a judgment in another district\\npursuant to § 1963 creates a “new judgment” that is treated as\\nif it had been rendered in the new district and that “[a]\\njudgment is void ‘if the court that rendered the judgment\\nlacked jurisdiction over the parties . . . .’” Fid. Nat’l Fin.\\nInc., 2017 WL 6049376, at *5 (quoting In re Ctr. Wholesale,\\n759 F.2d 1440, 1448 (9th Cir. 1985)). This reasoning,\\nhowever, is flawed because it incorrectly presumes that\\nconstitutional due process protections in the form of a\\npersonal jurisdiction requirement attach based on the\\nappearance of a judgment, rather than based on the\\ncharacteristics of the process that results in that judgment.\\nWe must therefore ask this crucial constitutional question:\\ndoes the Due Process Clause require that a court in which a\\njudgment creditor registers a pre-existing federal judgment\\nhave personal jurisdiction over judgment debtors at the time\\nof registration? We hold in the negative.\\n\\n A long line of Supreme Court cases reflects that the\\npersonal jurisdiction requirement of due process is grounded\\nin protecting litigants from being unfairly dragged into a\\nfaraway court to defend a suit.3 See, e.g., World-Wide\\n\\n\\n 3\\n Another function of the personal jurisdiction requirement is\\npreserving the balance of powers between coequal states. See World-Wide\\nVolkswagen Corp. v. Woodson, 444 U.S. 286, 292 (1980) (“[The\\nminimum contacts test] acts to ensure that the States through their courts,\\ndo not reach out beyond the limits imposed on them by their status as\\ncoequal sovereigns in a federal system.”). But that issue is not of concern\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 11\\n\\nVolkswagen Corp. v. Woodson, 444 U.S. 286, 291–92 (1980)\\n(“The concept of minimum contacts . . . protects the\\ndefendant against the burdens of litigating in a distant or\\ninconvenient forum.”). The constitutional standard focuses\\non “whether the ‘quality and nature’ of the defendant’s\\nactivity is such that it is ‘reasonable’ and ‘fair’ to require him\\nto conduct his defense in that State,” i.e., whether the\\ndefendant has sufficient “contacts with the forum state such\\nthat the maintenance of the suit does not offend ‘traditional\\nnotions of fair play and substantial justice.’” Kulko v.\\nSuperior Court, 436 U.S. 84, 92 (1978) (quoting Int’l Shoe\\nCo. v. Washington, 326 U.S. 310, 316–17 (1945)).\\n\\n A key distinction here, however, is that the process of\\nregistering a federal judgment in another federal district\\npursuant to § 1963 does not involve “maintenance of a suit”\\nor “[conducting a] defense.” See id. Instead, registration\\nsimply requires “filing a certified copy of the judgment.”4\\n\\n\\n\\n\\nhere, because we are dealing with judgments issued by federal, not state,\\ncourts.\\n 4\\n In fact, as discussed above, legislative history documents show that\\n§ 1963 was specifically intended to avoid additional litigation. The report\\nissued by the Senate Judiciary Committee explains:\\n\\n [R]egistration of judgments in other districts is a\\n modern legal device. It assists judgment creditors by\\n making it possible for them to pursue the property of a\\n debtor in satisfaction of a judgment by the ordinary\\n process of levying execution on a judgment in any\\n district where the judgment is registered. The result is\\n that both creditors and debtors are relieved of the\\n\\f12 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\nSee 28 U.S.C. § 1963. Defendants make no substantive\\nargument about why, even in the absence of the central\\nconcern animating the due-process-based personal\\njurisdiction requirement, courts must nonetheless have\\npersonal jurisdiction to register judgments under 28 U.S.C.\\n§ 1963. Constitutional due process requirements depend on\\nthe procedures at issue, not semantics—just because a court\\nmight be said to “render” a “new” judgment pursuant to\\nregistration does not mean that no meaningful distinction\\nexists between, on the one hand, subjecting a person to a\\nlawsuit in which claims are litigated in an ongoing\\nproceeding before a court that may determine or alter\\nsubstantive rights when it renders a final judgment, and, on\\nthe other hand, registering a pre-existing judgment, a process\\n\\n\\n additional cost and harassment of further litigation\\n which would otherwise be required by way of an action\\n on the judgment in a district other than that where the\\n judgment was originally obtained.\\n\\nS. Rep. No. 1917 (1954) (emphasis added), as reprinted in 1954\\nU.S.C.C.A.N. 3142, 3142; see also Air Transport Ass’n of Am. v. Prof’l\\nAir Traffic Controllers Org. (PATCO) (In re Prof’l Air Traffic\\nControllers Org. (PATCO)), 699 F.2d 539, 544 (D.C. Cir. 1983)\\n(“Registration is a rapid procedure that does not require the intervention\\nof a judge. It is merely a matter of having the clerk of the court in which\\nthe judgment is registered enter the pertinent provisions of the . . . sister\\ncourt’s judgment on the judgment docket.” (internal quotation marks and\\nalterations omitted)); Ohio Hoist Mfg. Co. v. LiRocchi, 490 F.2d 105, 107\\n(6th Cir. 1974) (“This statute has been held to have been adopted to\\nprotect both judgment creditors and judgment debtors from the additional\\ncost and harassment of further litigation which otherwise would be\\nincident to an action on the judgment in a foreign district.”); Stanford,\\n341 F. 2d at 270 (“[T]he purposes of § 1963 were . . . to eliminate the\\nnecessity and expense of a second lawsuit, and to avoid the impediments,\\nsuch as diversity of citizenship, which new and distinct federal litigation\\nmight otherwise encounter.”).\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 13\\n\\nwhich does not require any party to appear in court and in\\nwhich no judicial action is taken. See Shaffer, 433 U.S. at\\n210 n.36 (holding that, once a judgment is validly rendered\\nagainst a debtor, the judgment creditor may sue to satisfy the\\ndebt with property in a state that lacks personal jurisdiction\\nover the judgment debtor).\\n\\n Registration of a judgment pursuant to § 1963 can only\\noccur after a court of competent jurisdiction has rendered the\\noriginal judgment (otherwise there would be no judgment to\\nregister). Of course, the court that adjudicated the parties’\\nclaims and issued the original judgment must of course\\nsatisfy due process requirements such as ensuring personal\\njurisdiction. Thus, registration will only happen after full due\\nprocess has been provided during the original adjudication on\\nthe merits that determined the parties’ substantive rights and\\nobligations. We see no reason why—based on case law,\\npolicy, or otherwise—why the simple act of subsequently\\nregistering a judgment alters a debtor’s substantive rights\\nsuch that a due process right is triggered.5 To the contrary,\\nregistration itself does not change the amount of money or\\nproperty owed; it only facilitates collection of a pre-existing\\njudgment.\\n\\n\\n 5\\n Moreover, the practical reality is that judgment debtors must have\\nsome sort of presence or property in the state before judgment creditors\\ncan attempt to enforce a judgment, because without a judgment debtor’s\\npresence or property in the state, there is nothing against which the\\njudgment can be “enforced.” The district court elided this fact, concluding\\nthat any potential distinction between registration and enforcement was\\neliminated by this Court’s determination in Fidelity I that registration of\\na judgment has the same effect as creating a new judgment. We see no\\nbasis for such a conclusion. Enforcement involves additional steps\\nbeyond registration, such as moving to execute on a judgment debtor’s\\nassets, and nothing we said in Fidelity I changed this.\\n\\f14 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n We therefore hold that once a federal court of competent\\njurisdiction has determined the parties’ substantive rights and\\nentered a judgment following a proceeding that accords with\\ndue process, that federal judgment should be enforceable in\\nany other federal district by way of the federal judgment\\nregistration statute.\\n\\n CONCLUSION\\n\\n For the foregoing reasons, we hold that valid registration\\nof a federal judgment under § 1963 does not require that the\\ncourt of registration have personal jurisdiction over the\\njudgment debtors. The statute permits a judgment creditor to\\nregister a federal judgment “in any other district.” 28 U.S.C.\\n§ 1963 (emphasis added). Because the Due Process Clause\\ndoes not require that this broad statutory language be cabined\\nby a jurisdictional limitation, “any . . . district” includes\\ndistricts in which personal jurisdiction over judgment debtors\\nmay be lacking. As a result, the Washington Judgment was\\nnot void for lack of jurisdiction, and the district court erred by\\nvacating the Second Arizona Judgment on that basis.\\n\\n REVERSED and REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432213/\", \"author_raw\": \"TASHIMA, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"A WALLACE TASHIMA"},"judge2":{"kind":"string","value":"MARY H MURGUIA"},"judge3":{"kind":"string","value":"ROBERT N CHATIGNY"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4654960/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Fidelity National Financial v. Colin Friedman\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-22\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nFIDELITY NATIONAL FINANCIAL, No. 17-15913\nINC., a Delaware corporation;\nFIDELITY EXPRESS NETWORK, INC., a D.C. No.\nCalifornia corporation, CV 15-2288 DJH\n Plaintiffs-Appellants,\n\n v. OPINION\n\nCOLIN H. FRIEDMAN, individually\nand as trustee of Friedman Family\ntrust UDT Dated 7/23/87; HEDY\nKRAMER FRIEDMAN, individually\nand as trustee of Friedman Family\ntrust UDT Dated 7/23/87; FARID\nMESHKATAI, an individual; ANITA\nKRAMER MESHKATAI, individually\nand as trustee of Anita Kramer\nLiving Trust Dated 7/23/87,\n Defendants-Appellees.\n\n\n Appeal from the United States District Court\n for the District of Arizona\n Diane J. Humetewa, District Judge, Presiding\n\n Argued and Submitted October 12, 2018\n San Francisco, California\n\n Filed August 22, 2019\n\f2 FIDELITY NAT’L FIN. V. FRIEDMAN\n\n Before: A. Wallace Tashima and Mary H. Murguia,\n Circuit Judges, and Robert N. Chatigny,* District Judge.\n\n Opinion by Judge Tashima\n\n\n SUMMARY**\n\n\n Registration of Judgments\n\n The panel reversed the district court’s order vacating a\nregistered judgment and remanded.\n\n Plaintiffs obtained a civil fraud judgment in California\nfederal court and registered this California judgment in the\nDistrict of Arizona pursuant to 28 U.S.C. § 1963. Plaintiffs\nlater attempted to renew the Arizona judgment, but the\ndistrict court ruled that the renewal or re-registration was void\nas untimely. Plaintiffs next registered the California\njudgment in Washington federal court and then registered the\nnewly-obtained Washington judgment in the District of\nArizona. The district court granted defendants’ motion under\nFed. R. Civ. P. 60(b) to vacate the newly-registered second\nArizona judgment as void. In a prior appeal, the court of\nappeals reversed, holding that registering the California\njudgment in Washington created a “new” Washington\njudgment that could be re-registered in another state under\n\n\n *\n The Honorable Robert N. Chatigny, United States District Judge for\nthe District of Connecticut, sitting by designation.\n **\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f FIDELITY NAT’L FIN. V. FRIEDMAN 3\n\n§ 1963. On remand, the district court again granted relief\nunder Rule 60(b) on the ground that the Washington\njudgment was void because the Western District of\nWashington lacked personal jurisdiction over the defendants\nat the time of registration.\n\n The panel reversed, holding that a court need not have\npersonal jurisdiction over a judgment debtor in order to\n“merely register” a previously obtained judgment pursuant to\n§ 1963. The panel held that § 1963 itself does not require that\nthe court have personal jurisdiction. Further, the Due Process\nClause does not require that a court in which a judgment\ncreditor registers a pre-existing federal judgment have\npersonal jurisdiction over judgment debtors at the time of\nregistration. The panel held that once a federal court of\ncompetent jurisdiction has determined the parties’ substantive\nrights and entered a judgment following a proceeding that\ncomports with due process, that federal judgment should be\nenforceable in any other federal district by way of the federal\njudgment registration statute.\n\n\n COUNSEL\n\nThomas H. Case (argued) and Michael G. King, Hennelly &\nGrossfeld LLP, Marina del Rey, California, for Plaintiffs-\nAppellants.\n\nDavid M. Bass (argued), David M. Bass & Associates Inc.,\nLos Angeles, California; Dominica J. Minore, The Law\nOffices of Dominica J. Minore P.C., Scottsdale Arizona; for\nDefendants-Appellees.\n\f4 FIDELITY NAT’L FIN. V. FRIEDMAN\n\n OPINION\n\nTASHIMA, Circuit Judge:\n\n In order to facilitate the enforcement of federal\njudgments, 28 U.S.C. § 1963 provides that a judgment\nentered in a federal court may be registered in any other\nfederal district by “filing a certified copy of the judgment” in\nthat district. In this case we address, as a matter of first\nimpression in our Circuit, whether personal jurisdiction over\nthe judgment debtors in the district of registration is required\nfor such registration of a judgment. We hold that it is not,\nbecause neither § 1963 nor due process imposes such a\npersonal jurisdiction requirement. We therefore reverse the\norder and judgment of the district court, and remand.\n\n BACKGROUND\n\n In 2002, Plaintiffs-Appellants Fidelity National Financial,\nInc., and Fidelity Express Network, Inc. (collectively,\n“Fidelity”), obtained a multimillion dollar civil fraud\njudgment (the “California Judgment”) against Defendants-\nAppellees the Friedmans and Meshkatais (collectively,\n“Defendants”) in the U.S. District Court for the Central\nDistrict of California. This judgment became final on May\n15, 2003, after this Court dismissed Defendants’ appeal from\nthe judgment.\n\n While Defendants’ appeal in the original case was\npending, Fidelity registered the California Judgment in the\nDistrict of Arizona pursuant to the federal registration statute,\n28 U.S.C. § 1963. In 2007, Fidelity attempted to renew the\nArizona judgment. However, on March 2, 2012, the District\nCourt of Arizona ruled that Fidelity’s 2007 renewal or re-\n\f FIDELITY NAT’L FIN. V. FRIEDMAN 5\n\nregistration of the Arizona registered judgment were void as\nuntimely, because the judgment had already expired under\nArizona’s five-year statute of limitations for the enforcement\nof judgments.\n\n Unable to enforce the Arizona registered judgment or re-\nregister the original California Judgment in Arizona, Fidelity\ncame up with a creative alternative. Fidelity registered the\nCalifornia Judgment in the Western District of Washington\n(the “Washington Judgment”) Fidelity then registered the\nnewly-obtained Washington Judgment in the District of\nArizona (the “Second Arizona Judgment”).\n\n Several months later, Defendants moved the Arizona\nDistrict Court under Federal Rule of Civil Procedure 60(b) to\nvacate the newly-registered Second Arizona Judgment as\nvoid, arguing that § 1963 did not allow successive registration\nof federal judgments. That is, Defendants argued that\nregistering the California Judgment in Washington did not\ncreate a new Washington judgment that could then be\nregistered in Arizona under § 1963. Defendants also argued\nthat, in any case, the Washington Judgment was invalid for\nlack of personal jurisdiction over Defendants. The Arizona\ndistrict court granted Defendants’ motion and vacated the\nSecond Arizona Judgment, holding that § 1963 did not allow\nsuccessive registration of judgments—in other words, that\nonly an original judgment, such as the California Judgment in\nthis case, may be registered in another district under § 1963.\nSee Fid. Nat’l Fin., Inc. v. Friedman, 939 F. Supp. 2d 974,\n979–87 (D. Ariz. 2013). The district court did not reach\nDefendants’ second argument regarding lack of due process\nand personal jurisdiction. Id. at 986–87.\n\f6 FIDELITY NAT’L FIN. V. FRIEDMAN\n\n Fidelity appealed, and this Court reversed. Fid. Nat’l\nFin., Inc. v. Friedman, 803 F.3d 999 (9th Cir. 2015)\n(“Fidelity I”). We held that registering the California\nJudgment in Washington created a “new” Washington\njudgment that, like any other Washington judgment, could be\nre-registered in another state under the plain terms of § 1963.\nId. at 1003. We remanded the case to the Arizona district\ncourt on that basis, without reaching Defendants’ alternative\nargument that the Washington Judgment was void for lack of\njurisdiction over Defendants. See id. at 1003 n.3.\n\n On remand, the district court again granted Defendants’\nRule 60(b) motion for relief from judgment, after allowing\nsupplemental briefing on Defendants’ contention that the\nWashington Judgment was void because the Western District\nof Washington lacked personal jurisdiction over Defendants\nat the time of registration. See Fid. Nat’l Fin. Inc. v.\nFriedman, No. CV-15-2288-PHX-DJH, 2017 WL 6049376\n(D. Ariz. May 1, 2017). The district court held that\nregistration of a judgment pursuant to § 1963 requires that the\ncourt of registration have personal jurisdiction over the\njudgment debtors. See id. at *6–7. In reaching this\nconclusion, the district court first noted that a judgment is\nvoid if the court that “rendered” the judgment lacked\njurisdiction over the parties. Id. at *5. It then reasoned that\nbecause we suggested in Fidelity I that registering a judgment\nunder § 1963 creates a “new” judgment, the court of\nregistration can be said to have “rendered” a judgment such\nthat the normal jurisdictional requirements apply. See id. at\n*5–7. Because Defendants had no assets or other contacts in\nWashington, the district court concluded that the Washington\ncourt lacked personal jurisdiction over Defendants;\nconsequently, that the Washington Judgment was void and\ncould not have been validly registered in Arizona. Id. at *4,\n\f FIDELITY NAT’L FIN. V. FRIEDMAN 7\n\n*7. As a result, the district court vacated the Second Arizona\nJudgment. Fidelity again appealed.\n\n STANDARD OF REVIEW\n\n This Court “review[s] de novo . . . a district court’s ruling\nupon a Rule 60(b)(4) motion . . . because the question of the\nvalidity of a judgment is a legal one.” Export Grp. v. Reef\nIndus., Inc., 54 F.3d 1466, 1469 (9th Cir. 1995); see also\nFidelity I, 803 F.3d at 1001.\n\n DISCUSSION\n\n On appeal, Fidelity asserts that the district court\nimproperly granted relief from judgment because a court need\nnot have personal jurisdiction over a judgment debtor in order\nto “merely register” a previously obtained judgment pursuant\nto § 1963. For the reasons explained below, we agree.\n\n First, neither the relevant statute’s plain language nor its\npurpose supports a personal jurisdiction requirement for\nregistration of a judgment. Section 1963 provides:\n\n A judgment in an action for the recovery of\n money or property entered in any court of\n appeals, district court, bankruptcy court, or in\n the Court of International Trade may be\n registered by filing a certified copy of the\n judgment in any other district . . . when the\n judgment has become final by appeal or\n expiration of the time for appeal or when\n ordered by the court that entered the judgment\n for good cause shown. . . . A judgment so\n registered shall have the same effect as a\n\f8 FIDELITY NAT’L FIN. V. FRIEDMAN\n\n judgment of the district court of the district\n where registered and may be enforced in like\n manner.\n\n28 U.S.C. § 1963. Nothing in this provision limits the district\ncourts in which a judgment may be registered to only those\nthat can assert personal jurisdiction over the judgment\ndebtors. Instead, the provision’s text is extremely broad,\nallowing a judgment entered in “any . . . district court” to be\nregistered “in any other district.” Id. Furthermore, the fact\nthat registering a judgment in another district simply requires\n“filing a certified copy of the judgment” suggests that the\nregistration process is intended to be simple, essentially an\nadministrative task that does not require any additional\njudicial action. See id.\n\n Giving effect to this broad statutory language also accords\nwith the provision’s purpose, which is “to simplify and\nfacilitate collection on valid judgments.” Fidelity I, 803 F.3d\nat 1003 (citation omitted). Section 1963 aims to spare\ncreditors and debtors “the additional cost and harassment of\nfurther litigation which would otherwise be required by way\nof an action on the judgment in a district court other than that\nwhere the judgment was originally obtained.” S. Rep. No.\n83-1917 (1954), as reprinted in 1954 U.S.C.C.A.N. 3142; see\nalso Stanford v. Utley, 341 F. 2d 265, 270 (8th Cir. 1965)\n(“[T]he purposes of § 1963 were to simplify and facilitate the\nenforcement of federal judgments, . . . to eliminate the\nnecessity and expense of a second lawsuit, and to avoid the\nimpediments, such as diversity of citizenship, which new and\ndistinct federal litigation might otherwise encounter.”).\nAdhering to the statutory text’s expansive license to register\n\f FIDELITY NAT’L FIN. V. FRIEDMAN 9\n\njudgments in other districts effectuates those purposes;1\nreading a non-existent jurisdictional requirement into the\nstatute would contravene Congress’ intent by placing limits\non registration that would make the process more onerous and\npotentially require additional litigation regarding\njurisdiction.2 Thus, we hold that § 1963 itself does not\n\n 1\n Defendants point out that Fidelity’s registration of the California\nJudgment in Washington was not for the purpose of enforcing the\njudgment in Washington, but rather for the purpose of creating a new\njudgment that could subsequently be registered in Arizona and thus\nprovide an end-run around Arizona’s statute of limitations on the\nenforcement of judgments. Defendants’ argument that imposing a\njurisdictional requirement for registration under § 1963 is appropriate\nsimply because, as applied to the facts of this case, such a requirement\nwould not contravene the statute’s underlying purpose of facilitating\nenforcement in the state of registration and would close a seeming\nloophole, is unpersuasive. In Fidelity I, this Court rejected Defendants’\ninvitation to “refuse to give effect to Congress’ chosen words” simply\nbecause doing so “potentially allows plaintiffs to register a judgment that\nhas previously expired under a state’s statute of limitations,” noting that\nsuch an effect is “irrelevant in view of the plain language of § 1963.” 803\nF.3d at 1003.\n 2\n For example, a personal jurisdiction requirement could hamper a\njudgment creditor’s ability to register and enforce a judgment when a\njudgment debtor has no contacts with a state beyond assets that are\ncontrolled by a third party in the state. Cf. U.S.I. Props. Corp. v. M.D.\nConst. Co., 230 F.3d 489, 497 (1st Cir. 2000). In fact, the Supreme Court\nhas explicitly recognized that in some cases, the “presence of the\ndefendant’s property in a state . . . alone would not support . . .\njurisdiction” under the minimum contacts test for personal jurisdiction.\nShaffer v. Heitner, 433 U.S. 186, 209 (1977). Thus, imposing a minimum\ncontacts jurisdictional limit on the registration of judgments, as\nDefendants suggest is required, risks allowing a judgment debtor to “avoid\npayment of his obligations by the expedient of removing his assets to a\nplace where he is not subject to an in personam suit.” Id. This would be\ndirectly at odds with § 1963’s purpose of facilitating collection on valid\njudgments. See Fidelity I, 803 F.3d at 1003.\n\f10 FIDELITY NAT’L FIN. V. FRIEDMAN\n\nrequire that a court have personal jurisdiction over a\njudgment debtor in order to register an existing judgment.\n\n We next turn to the due process basis of the district\ncourt’s ruling. The district court noted that under this Court’s\nruling in Fidelity I, registering a judgment in another district\npursuant to § 1963 creates a “new judgment” that is treated as\nif it had been rendered in the new district and that “[a]\njudgment is void ‘if the court that rendered the judgment\nlacked jurisdiction over the parties . . . .’” Fid. Nat’l Fin.\nInc., 2017 WL 6049376, at *5 (quoting In re Ctr. Wholesale,\n759 F.2d 1440, 1448 (9th Cir. 1985)). This reasoning,\nhowever, is flawed because it incorrectly presumes that\nconstitutional due process protections in the form of a\npersonal jurisdiction requirement attach based on the\nappearance of a judgment, rather than based on the\ncharacteristics of the process that results in that judgment.\nWe must therefore ask this crucial constitutional question:\ndoes the Due Process Clause require that a court in which a\njudgment creditor registers a pre-existing federal judgment\nhave personal jurisdiction over judgment debtors at the time\nof registration? We hold in the negative.\n\n A long line of Supreme Court cases reflects that the\npersonal jurisdiction requirement of due process is grounded\nin protecting litigants from being unfairly dragged into a\nfaraway court to defend a suit.3 See, e.g., World-Wide\n\n\n 3\n Another function of the personal jurisdiction requirement is\npreserving the balance of powers between coequal states. See World-Wide\nVolkswagen Corp. v. Woodson, 444 U.S. 286, 292 (1980) (“[The\nminimum contacts test] acts to ensure that the States through their courts,\ndo not reach out beyond the limits imposed on them by their status as\ncoequal sovereigns in a federal system.”). But that issue is not of concern\n\f FIDELITY NAT’L FIN. V. FRIEDMAN 11\n\nVolkswagen Corp. v. Woodson, 444 U.S. 286, 291–92 (1980)\n(“The concept of minimum contacts . . . protects the\ndefendant against the burdens of litigating in a distant or\ninconvenient forum.”). The constitutional standard focuses\non “whether the ‘quality and nature’ of the defendant’s\nactivity is such that it is ‘reasonable’ and ‘fair’ to require him\nto conduct his defense in that State,” i.e., whether the\ndefendant has sufficient “contacts with the forum state such\nthat the maintenance of the suit does not offend ‘traditional\nnotions of fair play and substantial justice.’” Kulko v.\nSuperior Court, 436 U.S. 84, 92 (1978) (quoting Int’l Shoe\nCo. v. Washington, 326 U.S. 310, 316–17 (1945)).\n\n A key distinction here, however, is that the process of\nregistering a federal judgment in another federal district\npursuant to § 1963 does not involve “maintenance of a suit”\nor “[conducting a] defense.” See id. Instead, registration\nsimply requires “filing a certified copy of the judgment.”4\n\n\n\n\nhere, because we are dealing with judgments issued by federal, not state,\ncourts.\n 4\n In fact, as discussed above, legislative history documents show that\n§ 1963 was specifically intended to avoid additional litigation. The report\nissued by the Senate Judiciary Committee explains:\n\n [R]egistration of judgments in other districts is a\n modern legal device. It assists judgment creditors by\n making it possible for them to pursue the property of a\n debtor in satisfaction of a judgment by the ordinary\n process of levying execution on a judgment in any\n district where the judgment is registered. The result is\n that both creditors and debtors are relieved of the\n\f12 FIDELITY NAT’L FIN. V. FRIEDMAN\n\nSee 28 U.S.C. § 1963. Defendants make no substantive\nargument about why, even in the absence of the central\nconcern animating the due-process-based personal\njurisdiction requirement, courts must nonetheless have\npersonal jurisdiction to register judgments under 28 U.S.C.\n§ 1963. Constitutional due process requirements depend on\nthe procedures at issue, not semantics—just because a court\nmight be said to “render” a “new” judgment pursuant to\nregistration does not mean that no meaningful distinction\nexists between, on the one hand, subjecting a person to a\nlawsuit in which claims are litigated in an ongoing\nproceeding before a court that may determine or alter\nsubstantive rights when it renders a final judgment, and, on\nthe other hand, registering a pre-existing judgment, a process\n\n\n additional cost and harassment of further litigation\n which would otherwise be required by way of an action\n on the judgment in a district other than that where the\n judgment was originally obtained.\n\nS. Rep. No. 1917 (1954) (emphasis added), as reprinted in 1954\nU.S.C.C.A.N. 3142, 3142; see also Air Transport Ass’n of Am. v. Prof’l\nAir Traffic Controllers Org. (PATCO) (In re Prof’l Air Traffic\nControllers Org. (PATCO)), 699 F.2d 539, 544 (D.C. Cir. 1983)\n(“Registration is a rapid procedure that does not require the intervention\nof a judge. It is merely a matter of having the clerk of the court in which\nthe judgment is registered enter the pertinent provisions of the . . . sister\ncourt’s judgment on the judgment docket.” (internal quotation marks and\nalterations omitted)); Ohio Hoist Mfg. Co. v. LiRocchi, 490 F.2d 105, 107\n(6th Cir. 1974) (“This statute has been held to have been adopted to\nprotect both judgment creditors and judgment debtors from the additional\ncost and harassment of further litigation which otherwise would be\nincident to an action on the judgment in a foreign district.”); Stanford,\n341 F. 2d at 270 (“[T]he purposes of § 1963 were . . . to eliminate the\nnecessity and expense of a second lawsuit, and to avoid the impediments,\nsuch as diversity of citizenship, which new and distinct federal litigation\nmight otherwise encounter.”).\n\f FIDELITY NAT’L FIN. V. FRIEDMAN 13\n\nwhich does not require any party to appear in court and in\nwhich no judicial action is taken. See Shaffer, 433 U.S. at\n210 n.36 (holding that, once a judgment is validly rendered\nagainst a debtor, the judgment creditor may sue to satisfy the\ndebt with property in a state that lacks personal jurisdiction\nover the judgment debtor).\n\n Registration of a judgment pursuant to § 1963 can only\noccur after a court of competent jurisdiction has rendered the\noriginal judgment (otherwise there would be no judgment to\nregister). Of course, the court that adjudicated the parties’\nclaims and issued the original judgment must of course\nsatisfy due process requirements such as ensuring personal\njurisdiction. Thus, registration will only happen after full due\nprocess has been provided during the original adjudication on\nthe merits that determined the parties’ substantive rights and\nobligations. We see no reason why—based on case law,\npolicy, or otherwise—why the simple act of subsequently\nregistering a judgment alters a debtor’s substantive rights\nsuch that a due process right is triggered.5 To the contrary,\nregistration itself does not change the amount of money or\nproperty owed; it only facilitates collection of a pre-existing\njudgment.\n\n\n 5\n Moreover, the practical reality is that judgment debtors must have\nsome sort of presence or property in the state before judgment creditors\ncan attempt to enforce a judgment, because without a judgment debtor’s\npresence or property in the state, there is nothing against which the\njudgment can be “enforced.” The district court elided this fact, concluding\nthat any potential distinction between registration and enforcement was\neliminated by this Court’s determination in Fidelity I that registration of\na judgment has the same effect as creating a new judgment. We see no\nbasis for such a conclusion. Enforcement involves additional steps\nbeyond registration, such as moving to execute on a judgment debtor’s\nassets, and nothing we said in Fidelity I changed this.\n\f14 FIDELITY NAT’L FIN. V. FRIEDMAN\n\n We therefore hold that once a federal court of competent\njurisdiction has determined the parties’ substantive rights and\nentered a judgment following a proceeding that accords with\ndue process, that federal judgment should be enforceable in\nany other federal district by way of the federal judgment\nregistration statute.\n\n CONCLUSION\n\n For the foregoing reasons, we hold that valid registration\nof a federal judgment under § 1963 does not require that the\ncourt of registration have personal jurisdiction over the\njudgment debtors. The statute permits a judgment creditor to\nregister a federal judgment “in any other district.” 28 U.S.C.\n§ 1963 (emphasis added). Because the Due Process Clause\ndoes not require that this broad statutory language be cabined\nby a jurisdictional limitation, “any . . . district” includes\ndistricts in which personal jurisdiction over judgment debtors\nmay be lacking. As a result, the Washington Judgment was\nnot void for lack of jurisdiction, and the district court erred by\nvacating the Second Arizona Judgment on that basis.\n\n REVERSED and REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Fidelity National Financial v. Colin Friedman\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-22\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nFIDELITY NATIONAL FINANCIAL, No. 17-15913\\nINC., a Delaware corporation;\\nFIDELITY EXPRESS NETWORK, INC., a D.C. No.\\nCalifornia corporation, CV 15-2288 DJH\\n Plaintiffs-Appellants,\\n\\n v. OPINION\\n\\nCOLIN H. FRIEDMAN, individually\\nand as trustee of Friedman Family\\ntrust UDT Dated 7/23/87; HEDY\\nKRAMER FRIEDMAN, individually\\nand as trustee of Friedman Family\\ntrust UDT Dated 7/23/87; FARID\\nMESHKATAI, an individual; ANITA\\nKRAMER MESHKATAI, individually\\nand as trustee of Anita Kramer\\nLiving Trust Dated 7/23/87,\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n Diane J. Humetewa, District Judge, Presiding\\n\\n Argued and Submitted October 12, 2018\\n San Francisco, California\\n\\n Filed August 22, 2019\\n\\f2 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n Before: A. Wallace Tashima and Mary H. Murguia,\\n Circuit Judges, and Robert N. Chatigny,* District Judge.\\n\\n Opinion by Judge Tashima\\n\\n\\n SUMMARY**\\n\\n\\n Registration of Judgments\\n\\n The panel reversed the district court’s order vacating a\\nregistered judgment and remanded.\\n\\n Plaintiffs obtained a civil fraud judgment in California\\nfederal court and registered this California judgment in the\\nDistrict of Arizona pursuant to 28 U.S.C. § 1963. Plaintiffs\\nlater attempted to renew the Arizona judgment, but the\\ndistrict court ruled that the renewal or re-registration was void\\nas untimely. Plaintiffs next registered the California\\njudgment in Washington federal court and then registered the\\nnewly-obtained Washington judgment in the District of\\nArizona. The district court granted defendants’ motion under\\nFed. R. Civ. P. 60(b) to vacate the newly-registered second\\nArizona judgment as void. In a prior appeal, the court of\\nappeals reversed, holding that registering the California\\njudgment in Washington created a “new” Washington\\njudgment that could be re-registered in another state under\\n\\n\\n *\\n The Honorable Robert N. Chatigny, United States District Judge for\\nthe District of Connecticut, sitting by designation.\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 3\\n\\n§ 1963. On remand, the district court again granted relief\\nunder Rule 60(b) on the ground that the Washington\\njudgment was void because the Western District of\\nWashington lacked personal jurisdiction over the defendants\\nat the time of registration.\\n\\n The panel reversed, holding that a court need not have\\npersonal jurisdiction over a judgment debtor in order to\\n“merely register” a previously obtained judgment pursuant to\\n§ 1963. The panel held that § 1963 itself does not require that\\nthe court have personal jurisdiction. Further, the Due Process\\nClause does not require that a court in which a judgment\\ncreditor registers a pre-existing federal judgment have\\npersonal jurisdiction over judgment debtors at the time of\\nregistration. The panel held that once a federal court of\\ncompetent jurisdiction has determined the parties’ substantive\\nrights and entered a judgment following a proceeding that\\ncomports with due process, that federal judgment should be\\nenforceable in any other federal district by way of the federal\\njudgment registration statute.\\n\\n\\n COUNSEL\\n\\nThomas H. Case (argued) and Michael G. King, Hennelly &\\nGrossfeld LLP, Marina del Rey, California, for Plaintiffs-\\nAppellants.\\n\\nDavid M. Bass (argued), David M. Bass & Associates Inc.,\\nLos Angeles, California; Dominica J. Minore, The Law\\nOffices of Dominica J. Minore P.C., Scottsdale Arizona; for\\nDefendants-Appellees.\\n\\f4 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n OPINION\\n\\nTASHIMA, Circuit Judge:\\n\\n In order to facilitate the enforcement of federal\\njudgments, 28 U.S.C. § 1963 provides that a judgment\\nentered in a federal court may be registered in any other\\nfederal district by “filing a certified copy of the judgment” in\\nthat district. In this case we address, as a matter of first\\nimpression in our Circuit, whether personal jurisdiction over\\nthe judgment debtors in the district of registration is required\\nfor such registration of a judgment. We hold that it is not,\\nbecause neither § 1963 nor due process imposes such a\\npersonal jurisdiction requirement. We therefore reverse the\\norder and judgment of the district court, and remand.\\n\\n BACKGROUND\\n\\n In 2002, Plaintiffs-Appellants Fidelity National Financial,\\nInc., and Fidelity Express Network, Inc. (collectively,\\n“Fidelity”), obtained a multimillion dollar civil fraud\\njudgment (the “California Judgment”) against Defendants-\\nAppellees the Friedmans and Meshkatais (collectively,\\n“Defendants”) in the U.S. District Court for the Central\\nDistrict of California. This judgment became final on May\\n15, 2003, after this Court dismissed Defendants’ appeal from\\nthe judgment.\\n\\n While Defendants’ appeal in the original case was\\npending, Fidelity registered the California Judgment in the\\nDistrict of Arizona pursuant to the federal registration statute,\\n28 U.S.C. § 1963. In 2007, Fidelity attempted to renew the\\nArizona judgment. However, on March 2, 2012, the District\\nCourt of Arizona ruled that Fidelity’s 2007 renewal or re-\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 5\\n\\nregistration of the Arizona registered judgment were void as\\nuntimely, because the judgment had already expired under\\nArizona’s five-year statute of limitations for the enforcement\\nof judgments.\\n\\n Unable to enforce the Arizona registered judgment or re-\\nregister the original California Judgment in Arizona, Fidelity\\ncame up with a creative alternative. Fidelity registered the\\nCalifornia Judgment in the Western District of Washington\\n(the “Washington Judgment”) Fidelity then registered the\\nnewly-obtained Washington Judgment in the District of\\nArizona (the “Second Arizona Judgment”).\\n\\n Several months later, Defendants moved the Arizona\\nDistrict Court under Federal Rule of Civil Procedure 60(b) to\\nvacate the newly-registered Second Arizona Judgment as\\nvoid, arguing that § 1963 did not allow successive registration\\nof federal judgments. That is, Defendants argued that\\nregistering the California Judgment in Washington did not\\ncreate a new Washington judgment that could then be\\nregistered in Arizona under § 1963. Defendants also argued\\nthat, in any case, the Washington Judgment was invalid for\\nlack of personal jurisdiction over Defendants. The Arizona\\ndistrict court granted Defendants’ motion and vacated the\\nSecond Arizona Judgment, holding that § 1963 did not allow\\nsuccessive registration of judgments—in other words, that\\nonly an original judgment, such as the California Judgment in\\nthis case, may be registered in another district under § 1963.\\nSee Fid. Nat’l Fin., Inc. v. Friedman, 939 F. Supp. 2d 974,\\n979–87 (D. Ariz. 2013). The district court did not reach\\nDefendants’ second argument regarding lack of due process\\nand personal jurisdiction. Id. at 986–87.\\n\\f6 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n Fidelity appealed, and this Court reversed. Fid. Nat’l\\nFin., Inc. v. Friedman, 803 F.3d 999 (9th Cir. 2015)\\n(“Fidelity I”). We held that registering the California\\nJudgment in Washington created a “new” Washington\\njudgment that, like any other Washington judgment, could be\\nre-registered in another state under the plain terms of § 1963.\\nId. at 1003. We remanded the case to the Arizona district\\ncourt on that basis, without reaching Defendants’ alternative\\nargument that the Washington Judgment was void for lack of\\njurisdiction over Defendants. See id. at 1003 n.3.\\n\\n On remand, the district court again granted Defendants’\\nRule 60(b) motion for relief from judgment, after allowing\\nsupplemental briefing on Defendants’ contention that the\\nWashington Judgment was void because the Western District\\nof Washington lacked personal jurisdiction over Defendants\\nat the time of registration. See Fid. Nat’l Fin. Inc. v.\\nFriedman, No. CV-15-2288-PHX-DJH, 2017 WL 6049376\\n(D. Ariz. May 1, 2017). The district court held that\\nregistration of a judgment pursuant to § 1963 requires that the\\ncourt of registration have personal jurisdiction over the\\njudgment debtors. See id. at *6–7. In reaching this\\nconclusion, the district court first noted that a judgment is\\nvoid if the court that “rendered” the judgment lacked\\njurisdiction over the parties. Id. at *5. It then reasoned that\\nbecause we suggested in Fidelity I that registering a judgment\\nunder § 1963 creates a “new” judgment, the court of\\nregistration can be said to have “rendered” a judgment such\\nthat the normal jurisdictional requirements apply. See id. at\\n*5–7. Because Defendants had no assets or other contacts in\\nWashington, the district court concluded that the Washington\\ncourt lacked personal jurisdiction over Defendants;\\nconsequently, that the Washington Judgment was void and\\ncould not have been validly registered in Arizona. Id. at *4,\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 7\\n\\n*7. As a result, the district court vacated the Second Arizona\\nJudgment. Fidelity again appealed.\\n\\n STANDARD OF REVIEW\\n\\n This Court “review[s] de novo . . . a district court’s ruling\\nupon a Rule 60(b)(4) motion . . . because the question of the\\nvalidity of a judgment is a legal one.” Export Grp. v. Reef\\nIndus., Inc., 54 F.3d 1466, 1469 (9th Cir. 1995); see also\\nFidelity I, 803 F.3d at 1001.\\n\\n DISCUSSION\\n\\n On appeal, Fidelity asserts that the district court\\nimproperly granted relief from judgment because a court need\\nnot have personal jurisdiction over a judgment debtor in order\\nto “merely register” a previously obtained judgment pursuant\\nto § 1963. For the reasons explained below, we agree.\\n\\n First, neither the relevant statute’s plain language nor its\\npurpose supports a personal jurisdiction requirement for\\nregistration of a judgment. Section 1963 provides:\\n\\n A judgment in an action for the recovery of\\n money or property entered in any court of\\n appeals, district court, bankruptcy court, or in\\n the Court of International Trade may be\\n registered by filing a certified copy of the\\n judgment in any other district . . . when the\\n judgment has become final by appeal or\\n expiration of the time for appeal or when\\n ordered by the court that entered the judgment\\n for good cause shown. . . . A judgment so\\n registered shall have the same effect as a\\n\\f8 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n judgment of the district court of the district\\n where registered and may be enforced in like\\n manner.\\n\\n28 U.S.C. § 1963. Nothing in this provision limits the district\\ncourts in which a judgment may be registered to only those\\nthat can assert personal jurisdiction over the judgment\\ndebtors. Instead, the provision’s text is extremely broad,\\nallowing a judgment entered in “any . . . district court” to be\\nregistered “in any other district.” Id. Furthermore, the fact\\nthat registering a judgment in another district simply requires\\n“filing a certified copy of the judgment” suggests that the\\nregistration process is intended to be simple, essentially an\\nadministrative task that does not require any additional\\njudicial action. See id.\\n\\n Giving effect to this broad statutory language also accords\\nwith the provision’s purpose, which is “to simplify and\\nfacilitate collection on valid judgments.” Fidelity I, 803 F.3d\\nat 1003 (citation omitted). Section 1963 aims to spare\\ncreditors and debtors “the additional cost and harassment of\\nfurther litigation which would otherwise be required by way\\nof an action on the judgment in a district court other than that\\nwhere the judgment was originally obtained.” S. Rep. No.\\n83-1917 (1954), as reprinted in 1954 U.S.C.C.A.N. 3142; see\\nalso Stanford v. Utley, 341 F. 2d 265, 270 (8th Cir. 1965)\\n(“[T]he purposes of § 1963 were to simplify and facilitate the\\nenforcement of federal judgments, . . . to eliminate the\\nnecessity and expense of a second lawsuit, and to avoid the\\nimpediments, such as diversity of citizenship, which new and\\ndistinct federal litigation might otherwise encounter.”).\\nAdhering to the statutory text’s expansive license to register\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 9\\n\\njudgments in other districts effectuates those purposes;1\\nreading a non-existent jurisdictional requirement into the\\nstatute would contravene Congress’ intent by placing limits\\non registration that would make the process more onerous and\\npotentially require additional litigation regarding\\njurisdiction.2 Thus, we hold that § 1963 itself does not\\n\\n 1\\n Defendants point out that Fidelity’s registration of the California\\nJudgment in Washington was not for the purpose of enforcing the\\njudgment in Washington, but rather for the purpose of creating a new\\njudgment that could subsequently be registered in Arizona and thus\\nprovide an end-run around Arizona’s statute of limitations on the\\nenforcement of judgments. Defendants’ argument that imposing a\\njurisdictional requirement for registration under § 1963 is appropriate\\nsimply because, as applied to the facts of this case, such a requirement\\nwould not contravene the statute’s underlying purpose of facilitating\\nenforcement in the state of registration and would close a seeming\\nloophole, is unpersuasive. In Fidelity I, this Court rejected Defendants’\\ninvitation to “refuse to give effect to Congress’ chosen words” simply\\nbecause doing so “potentially allows plaintiffs to register a judgment that\\nhas previously expired under a state’s statute of limitations,” noting that\\nsuch an effect is “irrelevant in view of the plain language of § 1963.” 803\\nF.3d at 1003.\\n 2\\n For example, a personal jurisdiction requirement could hamper a\\njudgment creditor’s ability to register and enforce a judgment when a\\njudgment debtor has no contacts with a state beyond assets that are\\ncontrolled by a third party in the state. Cf. U.S.I. Props. Corp. v. M.D.\\nConst. Co., 230 F.3d 489, 497 (1st Cir. 2000). In fact, the Supreme Court\\nhas explicitly recognized that in some cases, the “presence of the\\ndefendant’s property in a state . . . alone would not support . . .\\njurisdiction” under the minimum contacts test for personal jurisdiction.\\nShaffer v. Heitner, 433 U.S. 186, 209 (1977). Thus, imposing a minimum\\ncontacts jurisdictional limit on the registration of judgments, as\\nDefendants suggest is required, risks allowing a judgment debtor to “avoid\\npayment of his obligations by the expedient of removing his assets to a\\nplace where he is not subject to an in personam suit.” Id. This would be\\ndirectly at odds with § 1963’s purpose of facilitating collection on valid\\njudgments. See Fidelity I, 803 F.3d at 1003.\\n\\f10 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\nrequire that a court have personal jurisdiction over a\\njudgment debtor in order to register an existing judgment.\\n\\n We next turn to the due process basis of the district\\ncourt’s ruling. The district court noted that under this Court’s\\nruling in Fidelity I, registering a judgment in another district\\npursuant to § 1963 creates a “new judgment” that is treated as\\nif it had been rendered in the new district and that “[a]\\njudgment is void ‘if the court that rendered the judgment\\nlacked jurisdiction over the parties . . . .’” Fid. Nat’l Fin.\\nInc., 2017 WL 6049376, at *5 (quoting In re Ctr. Wholesale,\\n759 F.2d 1440, 1448 (9th Cir. 1985)). This reasoning,\\nhowever, is flawed because it incorrectly presumes that\\nconstitutional due process protections in the form of a\\npersonal jurisdiction requirement attach based on the\\nappearance of a judgment, rather than based on the\\ncharacteristics of the process that results in that judgment.\\nWe must therefore ask this crucial constitutional question:\\ndoes the Due Process Clause require that a court in which a\\njudgment creditor registers a pre-existing federal judgment\\nhave personal jurisdiction over judgment debtors at the time\\nof registration? We hold in the negative.\\n\\n A long line of Supreme Court cases reflects that the\\npersonal jurisdiction requirement of due process is grounded\\nin protecting litigants from being unfairly dragged into a\\nfaraway court to defend a suit.3 See, e.g., World-Wide\\n\\n\\n 3\\n Another function of the personal jurisdiction requirement is\\npreserving the balance of powers between coequal states. See World-Wide\\nVolkswagen Corp. v. Woodson, 444 U.S. 286, 292 (1980) (“[The\\nminimum contacts test] acts to ensure that the States through their courts,\\ndo not reach out beyond the limits imposed on them by their status as\\ncoequal sovereigns in a federal system.”). But that issue is not of concern\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 11\\n\\nVolkswagen Corp. v. Woodson, 444 U.S. 286, 291–92 (1980)\\n(“The concept of minimum contacts . . . protects the\\ndefendant against the burdens of litigating in a distant or\\ninconvenient forum.”). The constitutional standard focuses\\non “whether the ‘quality and nature’ of the defendant’s\\nactivity is such that it is ‘reasonable’ and ‘fair’ to require him\\nto conduct his defense in that State,” i.e., whether the\\ndefendant has sufficient “contacts with the forum state such\\nthat the maintenance of the suit does not offend ‘traditional\\nnotions of fair play and substantial justice.’” Kulko v.\\nSuperior Court, 436 U.S. 84, 92 (1978) (quoting Int’l Shoe\\nCo. v. Washington, 326 U.S. 310, 316–17 (1945)).\\n\\n A key distinction here, however, is that the process of\\nregistering a federal judgment in another federal district\\npursuant to § 1963 does not involve “maintenance of a suit”\\nor “[conducting a] defense.” See id. Instead, registration\\nsimply requires “filing a certified copy of the judgment.”4\\n\\n\\n\\n\\nhere, because we are dealing with judgments issued by federal, not state,\\ncourts.\\n 4\\n In fact, as discussed above, legislative history documents show that\\n§ 1963 was specifically intended to avoid additional litigation. The report\\nissued by the Senate Judiciary Committee explains:\\n\\n [R]egistration of judgments in other districts is a\\n modern legal device. It assists judgment creditors by\\n making it possible for them to pursue the property of a\\n debtor in satisfaction of a judgment by the ordinary\\n process of levying execution on a judgment in any\\n district where the judgment is registered. The result is\\n that both creditors and debtors are relieved of the\\n\\f12 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\nSee 28 U.S.C. § 1963. Defendants make no substantive\\nargument about why, even in the absence of the central\\nconcern animating the due-process-based personal\\njurisdiction requirement, courts must nonetheless have\\npersonal jurisdiction to register judgments under 28 U.S.C.\\n§ 1963. Constitutional due process requirements depend on\\nthe procedures at issue, not semantics—just because a court\\nmight be said to “render” a “new” judgment pursuant to\\nregistration does not mean that no meaningful distinction\\nexists between, on the one hand, subjecting a person to a\\nlawsuit in which claims are litigated in an ongoing\\nproceeding before a court that may determine or alter\\nsubstantive rights when it renders a final judgment, and, on\\nthe other hand, registering a pre-existing judgment, a process\\n\\n\\n additional cost and harassment of further litigation\\n which would otherwise be required by way of an action\\n on the judgment in a district other than that where the\\n judgment was originally obtained.\\n\\nS. Rep. No. 1917 (1954) (emphasis added), as reprinted in 1954\\nU.S.C.C.A.N. 3142, 3142; see also Air Transport Ass’n of Am. v. Prof’l\\nAir Traffic Controllers Org. (PATCO) (In re Prof’l Air Traffic\\nControllers Org. (PATCO)), 699 F.2d 539, 544 (D.C. Cir. 1983)\\n(“Registration is a rapid procedure that does not require the intervention\\nof a judge. It is merely a matter of having the clerk of the court in which\\nthe judgment is registered enter the pertinent provisions of the . . . sister\\ncourt’s judgment on the judgment docket.” (internal quotation marks and\\nalterations omitted)); Ohio Hoist Mfg. Co. v. LiRocchi, 490 F.2d 105, 107\\n(6th Cir. 1974) (“This statute has been held to have been adopted to\\nprotect both judgment creditors and judgment debtors from the additional\\ncost and harassment of further litigation which otherwise would be\\nincident to an action on the judgment in a foreign district.”); Stanford,\\n341 F. 2d at 270 (“[T]he purposes of § 1963 were . . . to eliminate the\\nnecessity and expense of a second lawsuit, and to avoid the impediments,\\nsuch as diversity of citizenship, which new and distinct federal litigation\\nmight otherwise encounter.”).\\n\\f FIDELITY NAT’L FIN. V. FRIEDMAN 13\\n\\nwhich does not require any party to appear in court and in\\nwhich no judicial action is taken. See Shaffer, 433 U.S. at\\n210 n.36 (holding that, once a judgment is validly rendered\\nagainst a debtor, the judgment creditor may sue to satisfy the\\ndebt with property in a state that lacks personal jurisdiction\\nover the judgment debtor).\\n\\n Registration of a judgment pursuant to § 1963 can only\\noccur after a court of competent jurisdiction has rendered the\\noriginal judgment (otherwise there would be no judgment to\\nregister). Of course, the court that adjudicated the parties’\\nclaims and issued the original judgment must of course\\nsatisfy due process requirements such as ensuring personal\\njurisdiction. Thus, registration will only happen after full due\\nprocess has been provided during the original adjudication on\\nthe merits that determined the parties’ substantive rights and\\nobligations. We see no reason why—based on case law,\\npolicy, or otherwise—why the simple act of subsequently\\nregistering a judgment alters a debtor’s substantive rights\\nsuch that a due process right is triggered.5 To the contrary,\\nregistration itself does not change the amount of money or\\nproperty owed; it only facilitates collection of a pre-existing\\njudgment.\\n\\n\\n 5\\n Moreover, the practical reality is that judgment debtors must have\\nsome sort of presence or property in the state before judgment creditors\\ncan attempt to enforce a judgment, because without a judgment debtor’s\\npresence or property in the state, there is nothing against which the\\njudgment can be “enforced.” The district court elided this fact, concluding\\nthat any potential distinction between registration and enforcement was\\neliminated by this Court’s determination in Fidelity I that registration of\\na judgment has the same effect as creating a new judgment. We see no\\nbasis for such a conclusion. Enforcement involves additional steps\\nbeyond registration, such as moving to execute on a judgment debtor’s\\nassets, and nothing we said in Fidelity I changed this.\\n\\f14 FIDELITY NAT’L FIN. V. FRIEDMAN\\n\\n We therefore hold that once a federal court of competent\\njurisdiction has determined the parties’ substantive rights and\\nentered a judgment following a proceeding that accords with\\ndue process, that federal judgment should be enforceable in\\nany other federal district by way of the federal judgment\\nregistration statute.\\n\\n CONCLUSION\\n\\n For the foregoing reasons, we hold that valid registration\\nof a federal judgment under § 1963 does not require that the\\ncourt of registration have personal jurisdiction over the\\njudgment debtors. The statute permits a judgment creditor to\\nregister a federal judgment “in any other district.” 28 U.S.C.\\n§ 1963 (emphasis added). Because the Due Process Clause\\ndoes not require that this broad statutory language be cabined\\nby a jurisdictional limitation, “any . . . district” includes\\ndistricts in which personal jurisdiction over judgment debtors\\nmay be lacking. As a result, the Washington Judgment was\\nnot void for lack of jurisdiction, and the district court erred by\\nvacating the Second Arizona Judgment on that basis.\\n\\n REVERSED and REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4093,"cells":{"id":{"kind":"number","value":4655515,"string":"4,655,515"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Ussec v. Hui Feng"},"decision_date":{"kind":"string","value":"2019-08-23"},"docket_number":{"kind":"string","value":"17-56522"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Kermit V. Lipez, * Kim McLane Wardlaw, and Andrew D. Hurwitz, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"LIPEZ, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n U.S. SECURITIES & EXCHANGE No. 17-56522\\n COMMISSION,\\n Plaintiff-Appellee, D.C. No.\\n 2:15-cv-09420-\\n v. CBM-SS\\n\\n HUI FENG; LAW OFFICES OF FENG\\n AND ASSOCIATES PC, OPINION\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Consuelo B. Marshall, District Judge, Presiding\\n\\n Argued and Submitted May 16, 2019\\n Pasadena, California\\n\\n Filed August 23, 2019\\n\\n Before: Kermit V. Lipez, * Kim McLane Wardlaw,\\n and Andrew D. Hurwitz, Circuit Judges.\\n\\n Opinion by Judge Lipez\\n\\n\\n\\n *\\n The Honorable Kermit V. Lipez, United States Circuit Judge for\\nthe First Circuit, sitting by designation.\\n\\f2 USSEC V. FENG\\n\\n SUMMARY **\\n\\n\\n Securities & Exchange Commission\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of the U.S. Securities and Exchange\\nCommission (“SEC”) in its civil complaint filed against Hui\\nFeng and his law firm, alleging securities fraud.\\n\\n The U.S. Immigrant Investor Program, also known as the\\nEB-5 program, provides legal permanent residency in the\\nUnited States to foreign nationals who invest in U.S.-based\\nprojects. Multiple foreign investors may pool their money\\nin the same enterprise, and these pooled investments are\\nmade through “regional centers” which are regulated by the\\nU.S. Citizenship and Immigration Services.\\n\\n Feng legally represented clients through the EB-5\\nprocess, and entered into marketing agreements with\\nregional centers. The basis of the agreements between the\\nregional centers and Feng’s investors were known as private\\nplacement memoranda (“PPMs”).\\n\\n The panel agreed with the district court that the EB-5\\ninvestments in this case constituted “securities” in the form\\nof investment contracts. The panel rejected Feng’s argument\\nthat the transactions were not “securities” because his clients\\ndid not expect profits from their investments. Specifically,\\nthe panel held that the PPMs’ identification of the\\ninvestments as securities, the form of the investment entity\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f USSEC V. FENG 3\\n\\nas a limited partnership, and the promise of a fixed rate of\\nreturn all indicated that the EB-5 transactions were\\nsecurities. The panel rejected Feng’s contention that the\\nadministrative fees upended the expectation of profits. The\\npanel also rejected Feng’s assertion that his clients lacked an\\nexpectation of profit because they were motivated to\\nparticipate in the EB-5 program by the promise of visas, not\\nby profit.\\n\\n Concerning the cause of action that Feng failed to\\nregister as a broker in violation of Section 15(a) of the\\nSecurities and Exchange Act of 1934, the panel agreed with\\nthe district court’s conclusion that Feng was acting as a\\nbroker and violated the registration requirement. The panel\\nheld that the district court properly made its broker\\ndetermination by utilizing the totality-of-the-circumstances\\napproach by relying on the so-called Hansen factors. The\\npanel rejected Feng’s arguments that the broker registration\\nrequirement should not apply to his circumstances.\\n\\n The panel affirmed the district court’s finding that Feng\\nengaged in securities fraud in violation of Section 17(a) of\\nthe Securities Act of 1933 and Section 10(b) of the Securities\\nExchange Act of 1934 based on two theories of fraud\\nliability: material omissions, and schemes to defraud.\\nConcerning material omissions, the panel held that where\\nFeng worked as both a broker and an immigration attorney,\\nhe had fiduciary duties to his clients, including the obligation\\nto disclose conflicts of interest. Here, there was a risk that\\nFeng’s judgment would be swayed by the promise of a\\ncommission from the regional center, and this presented a\\nconflict with Feng’s representation of a client, which he\\nfailed to disclose to his clients, and this failure was material.\\nConcerning schemes to defraud, the panel affirmed the\\ndistrict court’s findings that Feng defrauded the regional\\n\\f4 USSEC V. FENG\\n\\ncenters that refused to pay commissions to U.S.-based\\nattorneys not registered as brokers, and defrauded clients\\nwho sought a reduction in their administrative fees.\\n\\n Finally, the panel held that the district court did not abuse\\nits discretion in entering a disgorgement order. The district\\ncourt calculated that Feng received $1.268 million for\\ncommissions in connection with the EB-5 program, and\\nordered disgorgement of the entire amount.\\n\\n\\n COUNSEL\\n\\nAndrew B. Holmes (argued) and Matthew D. Taylor,\\nHolmes Taylor Scott & Jones LLP, Los Angeles, California,\\nfor Defendant-Appellant Law Offices of Feng & Associates.\\n\\nHui Feng (argued), Law Offices of Feng & Associates,\\nFlushing, New York, for Defendant-Appellant Hui Feng.\\n\\nKerry J. Dingle (argued), Senior Counsel; Jeffery A. Berger,\\nSenior Litigation Counsel; Michael A. Conley, Solicitor;\\nRobert B. Stebbins, General Counsel; Securities &\\nExchange Commission, Washington, D.C.; for Plaintiff-\\nAppellee.\\n\\f USSEC V. FENG 5\\n\\n OPINION\\n\\nLIPEZ, Circuit Judge:\\n\\n This appeal requires us to decide whether certain\\ninvestments made by participants in the U.S. Immigrant\\nInvestor Program are “securities” subject to regulation by the\\nSecurities and Exchange Commission (“SEC”) and, if so,\\nwhether appellant Hui Feng was a securities “broker”\\nrequired to register with the SEC and whether he committed\\nsecurities fraud in connection with the transactions. 1 The\\ndistrict court granted summary judgment for the SEC. We\\naffirm.\\n\\n I.\\n\\nA. The EB-5 Program\\n\\n The U.S. Immigrant Investor Program, colloquially\\nreferred to as the EB-5 program, provides legal permanent\\nresidency in the United States to foreign nationals who\\ninvest in U.S.-based projects. See 8 U.S.C.\\n§ 1153(b)(5)(A). 2 Generally, qualified immigrants may gain\\nU.S. visas through direct investment of at least $1 million in\\na new commercial enterprise that creates at least ten full-\\ntime jobs for U.S. workers. Id. § 1153(b)(5)(A), (C).\\nInvestment in a business in a “targeted employment area”\\n\\n 1\\n Both Hui Feng and his law firm, the Law Offices of Feng &\\nAssociates P.C., are appellants in this matter. The parties’ arguments do\\nnot distinguish between the two, and so, as a matter of convenience, we\\nrefer to the appellants collectively as “Feng.”\\n 2\\n By statute, the Immigrant Investor Program is the fifth preference\\nin the employment-based visa category, which gave rise to the nickname\\n“EB-5.” See 8 U.S.C. § 1153(b)(5)(A).\\n\\f6 USSEC V. FENG\\n\\nlowers the required capital investment amount to $500,000.\\nId. § 1153(b)(5)(C)(i), (ii).\\n\\n Multiple foreign investors may pool their money in the\\nsame enterprise, provided that each invests the required\\namount and “each individual investment results in the\\ncreation of at least ten full-time positions.” 8 C.F.R.\\n§ 204.6(g). Pooled investments are made through “regional\\ncenters,” which are regulated by the U.S. Citizenship and\\nImmigration Services (“USCIS”). The regional centers offer\\nspecific projects to investors and manage the pooled\\ninvestments. See id. § 204.6(e), (m).\\n\\n A foreign national investing in an enterprise must file an\\nI-526 application with the USCIS to prove that the\\ninvestment will satisfy EB-5 program requirements. Id.\\n§ 204.6(a), (j)(2). Approval results in conditional permanent\\nresident status. Two years later, the investor may remove\\nthe conditions on lawful permanent resident status by filing\\nan I-829 petition, demonstrating that the investment satisfied\\nthe EB-5 requirements and created, or will create within a\\nreasonable period, ten qualifying jobs. Id. § 216.6.\\n\\nB. Factual Background\\n\\n Feng and the SEC largely agree on the facts. To the\\nextent that they have differing views of the record, we draw\\nall justifiable inferences in the light most favorable to Feng.\\nBlankenhorn v. City of Orange, 485 F.3d 463, 470 (9th Cir.\\n2007).\\n\\n Feng conducts an immigration law practice in New York\\nCity. Between 2010 and 2016, he led approximately\\n150 clients through the EB-5 process, substantially all of\\nwhom were Chinese nationals who did not speak English.\\nFeng estimated that 80 percent of his clients found him\\n\\f USSEC V. FENG 7\\n\\nthrough his website, which posted comparisons and\\nrecommendations of available EB-5 projects. Feng holds a\\nlaw degree from Columbia University and an MBA from the\\nTuck School of Business at Dartmouth, and he touted his\\nskill at investigating and determining the most reliable\\ninvestment opportunities.\\n\\n Feng charged his clients a $10,000 to $15,000 upfront\\nfee. His services included serving as a liaison between\\nclients and regional centers, explaining the English-language\\noffering materials to his clients, negotiating with regional\\ncenters regarding administrative fees charged to the clients\\nby the centers, and compiling and submitting his clients’\\nsigned offering documents to regional centers. On three\\noccasions, investors wired funds directly to Feng, who\\ntransferred the funds to the regional centers.\\n\\n Feng also entered into marketing agreements with\\nregional centers. Through these agreements, he reserved a\\nnumber of the limited investor “spots” in certain EB-5\\nprojects, which he then promoted on his website and tried to\\nfill within a set time period. If one of Feng’s clients made\\nthe required capital contribution to one of these regional\\ncenters, and if the USCIS approved the investor’s I-526\\npetition, the regional center agreed to pay Feng commissions\\nof $15,000 to $70,000. Although hundreds of regional\\ncenters participate in the EB-5 program, Feng procured\\ninvestors only for ten centers, all of which agreed to pay\\ncommissions. Feng did not disclose to his clients the\\ncommissions he received from the regional centers unless\\nthey specifically asked about them. 3\\n\\n\\n 3\\n When the SEC’s investigation was pending, Feng revised his\\nretainer agreement to state that regional centers may pay fees to an\\n\\f8 USSEC V. FENG\\n\\n Payment of a commission is an EB-5 industry practice,\\nbut many regional centers refuse to pay a commission to a\\nU.S.-based attorney who is not registered as a broker, to\\navoid causing or aiding and abetting a violation of the broker\\nregistration requirement in Section 15(a) of the Securities\\nExchange Act of 1934. See 15 U.S.C. § 78o(a). Several\\nregional centers offer a discount or rebate of a portion of the\\nadministrative fee charged to the investor if there is no\\nreferral commission for the attorney. Feng chose neither to\\nregister as a broker, nor to forego commissions.\\n\\n When told by some regional centers that they could only\\npay commissions to overseas recruiters, Feng responded that\\nhe had referral partners in China who were finding investors.\\nHe entered into marketing agreements in which he promised\\nto relay the commissions to those individuals. He also\\narranged agreements that were directly between the regional\\ncenters and the individuals. The referral partners included\\nFeng’s wife, mother, and mother-in-law. Feng described\\nthese individuals in his deposition testimony as his\\n“nominees,” whose role was to be “just a surrogate to receive\\nthe money.” They did not do the work of recruiting\\ninvestors—Feng did that himself. After the regional centers\\npaid commissions to the individuals, they transferred the\\nfunds to Feng. Feng asserts that the regional centers only\\nasked for a foreign contact to pay and did not “care” who it\\nwas.\\n\\n In 2014, Feng also created Atlantic Business Consulting\\nLimited (“ABCL”), based in Hong Kong. According to\\nFeng, this decision was “driven in part by regional centers\\ninforming me they needed an overseas entity to pay.” ABCL\\n\\noverseas company owned by Feng upon completion of the client’s visa\\napproval.\\n\\f USSEC V. FENG 9\\n\\nentered into marketing agreements with regional centers, and\\nits only source of revenue was EB-5 commissions. Feng did\\nnot inform the regional centers that he solicited and referred\\nABCL’s investors himself. All of Feng’s employees worked\\nboth for his law office and for ABCL. Feng’s mother,\\nXiuyuan Tan, signed agreements between ABCL and\\nregional centers as ABCL’s “President,” but Feng admitted\\nshe played no role in ABCL. Feng also did not inform the\\nregional centers that he was ABCL’s beneficial owner, with\\nsole control of ABCL’s bank account. Representatives from\\nthe regional centers testified that they would have ceased\\ntheir marketing agreements if they had known about Feng’s\\nrelationship with the individual referral agents and with\\nABCL.\\n\\n In total, regional centers paid Feng and his overseas\\nentities $1.268 million in commissions for investments made\\nby Feng’s clients. At the time of the district court judgment,\\nthese entities were contractually entitled to an additional\\n$3.45 million in commissions, pending the approval of his\\nclients’ I-526 petitions.\\n\\n The basis of the agreements between the regional centers\\nand Feng’s investors was set forth in the regional centers’\\noffering materials, also known as private placement\\nmemoranda (“PPMs”). Many of the PPMs referred to the\\ninvestments as “securities” and asserted that the investments\\nwere compliant with U.S. securities laws. The regional\\ncenters structured the investments as limited partnerships, in\\nwhich the investors became limited partners and the regional\\ncenter was the general partner. The regional centers\\npromised investors a fixed, annual return on investment,\\nwhich ranged across projects from 0.5 to 5 percent of the\\ncapital contribution, and investors received Schedule K-1 tax\\nforms to report their investment income from the\\n\\f10 USSEC V. FENG\\n\\npartnership. The partnerships all made construction loans or\\notherwise financed a specified construction project. At the\\nend of the investment term, typically five to six years, the\\nregional centers promised the investors a return of their\\ncapital contribution, subject to market risks.\\n\\n The PPMs required that investors pay an administrative\\nfee, which ranged across projects from $30,000 to $50,000,\\nin addition to the capital contribution of $1 million or\\n$500,000. The PPMs expressly stated that these\\nadministrative fees were for operating and marketing costs,\\nwere not part of the capital contribution, and did not earn\\ninterest.\\n\\n Approximately 20 percent of Feng’s clients asked him to\\nseek a reduced administrative fee from the regional centers.\\nIn those instances, Feng negotiated with regional centers and\\nfacilitated contracts between those regional centers and his\\nclients for a rebate of a portion of the administrative fee.\\nFeng did not disclose to these clients, however, that the\\nadministrative fees helped to fund his commissions, nor that\\nthe regional centers offset the clients’ administrative fee\\nrebate with a reduction in the commissions to which he was\\ncontractually entitled. Indeed, Feng asked the regional\\ncenters not to inform the clients about either fact. Feng\\nexplained in his deposition that he wanted to create the\\nappearance to his clients that he was advocating on their\\nbehalf with the regional center and also to “keep as much of\\nthe marketing fee as possible”—that is, his commission—\\nand to avoid further “haggl[ing]” with clients who might ask\\nFeng to rebate more of the commission. 4\\n\\n\\n 4\\n For example, one client testified that Feng sent him a check for\\n$15,000, which Feng explained as a return of charges from the regional\\n\\f USSEC V. FENG 11\\n\\nC. Procedural Background\\n\\n The SEC filed a civil complaint against Feng and his law\\nfirm on December 7, 2015. The first and second causes of\\naction allege fraud under Section 17(a) of the Securities Act\\nof 1933, 15 U.S.C. § 77q(a), and Section 10(b) of the\\nSecurities Exchange Act of 1934, 15 U.S.C. § 78j(b) (“the\\nExchange Act”), respectively. The third cause of action\\nalleges failure to register as a broker-dealer under Section\\n15(a) of the Exchange Act, 15 U.S.C. § 78o.\\n\\n The defendants filed a motion for judgment on the\\npleadings, which the district court denied in August 2016.\\nThe parties subsequently filed cross-motions for summary\\njudgment. Feng argued that the EB-5 investments were not\\n“securities” because the investors had no expectation of\\nprofit—only of obtaining a green card. The SEC argued that\\nthe undisputed evidence showed that Feng acted as a\\n“broker” of “securities” as a matter of law. The district court\\nfound no genuine dispute of material fact and granted\\nsummary judgment for the SEC on all three causes of action.\\nIn this appeal, Feng argues that the transactions were not\\n“securities,” and so the district court erred by denying his\\nmotion for summary judgment and granting the SEC’s\\nmotion.\\n\\n We review the district court’s decision on cross-motions\\nfor summary judgment de novo. Avery v. First Resolution\\nMgmt. Corp., 568 F.3d 1018, 1021 (9th Cir. 2009).\\nSummary judgment is appropriate when there is no genuine\\ndispute of material fact and the moving party is entitled to\\njudgment as a matter of law. Fed. R. Civ. P. 56(a). We\\n\\ncenter without disclosing that Feng had received a $70,000 commission\\non the client’s $1 million investment.\\n\\f12 USSEC V. FENG\\n\\nconsider cross-motions “separately, giving the nonmoving\\nparty in each instance the benefit of all reasonable\\ninferences.” Zabriskie v. Fed. Nat’l Mortg. Ass’n, 912 F.3d\\n1192, 1196 (9th Cir. 2019) (quoting ACLU of Nev. v. City of\\nLas Vegas, 466 F.3d 784, 790–91 (9th Cir. 2006)).\\n\\n II.\\n\\n Feng contends that the district court erred in its threshold\\ndetermination that the EB-5 investments at issue were\\n“securities.” Because this finding is essential to all the\\ncauses of action, we begin by considering whether the\\ndistrict court properly made that determination as a matter of\\nlaw.\\n\\n The district court found that the EB-5 investments were\\nsecurities because they were made through investment\\ncontracts between the regional centers and the investors.\\nSection 3(a)(10) of the Securities Act defines the term\\n“security” broadly, enumerating a long list of financial\\ninstruments, including “any . . . investment contract.”\\n15 U.S.C. § 78c(a)(10). Although the Securities Act does\\nnot define the term “investment contract,” the parties agree\\nthat the applicable test, set forth in SEC v. W. J. Howey Co.,\\nrequires “a contract, transaction or scheme whereby a person\\ninvests his money in a common enterprise and is led to\\nexpect profits solely from the efforts of the promoter or a\\nthird party.” 328 U.S. 293, 298–99 (1946). Courts applying\\nHowey “conduct an objective inquiry into the character of\\nthe instrument or transaction offered based on what the\\npurchasers were ‘led to expect,’” including an analysis of the\\npromotional materials associated with the transaction.\\nWarfield v. Alaniz, 569 F.3d 1015, 1021 (9th Cir. 2009)\\n(quoting Howey, 328 U.S. at 298–99).\\n\\f USSEC V. FENG 13\\n\\n It is undisputed that Feng’s clients invested money in a\\ncommon enterprise managed by a third party. Feng argues\\nthat the transactions nonetheless were not “securities”\\nbecause the clients did not expect profits from their\\ninvestments. He asserts that his clients would have\\nrecognized that the modest fixed return they were due from\\nthe regional centers would not exceed the significant\\nadministrative fees that they paid. He further maintains that\\nthe lack of expected profit was not a concern for his clients\\nbecause their motivation for making an EB-5 investment was\\nto obtain U.S. visas. We find Feng’s characterization of the\\nEB-5 investments unpersuasive.\\n\\nA. The PPMs\\n\\n The regional centers promoted and described the EB-5\\ninvestments in the PPMs. The PPMs repeatedly referred to\\nthe investments as “securities” and explained that the\\nofferings were made pursuant to U.S. securities laws.\\nAlthough “the name given to an instrument is not\\ndispositive,” “most instruments bearing th[e] traditional\\ntitles [associated with securities] are likely to be covered by\\nthe statutes.” United Hous. Found., Inc. v. Forman, 421 U.S.\\n837, 850 (1975). Further, “the use of a traditional name such\\nas ‘stocks’ or ‘bonds’ [may] lead a purchaser justifiably to\\nassume that the federal securities laws apply.” Id. That\\ninference inevitably would be drawn “when the underlying\\ntransaction embodies some of the significant characteristics\\ntypically associated with the named instrument.” Id. at 851.\\n\\n Such characteristics are evident and undisputed here.\\nThe EB-5 transactions were structured as investments in\\nlimited partnerships, with the promise of a fixed annual\\nreturn on the investment, ranging from .5 to 5 percent.\\nInvestments in limited partnerships generally constitute\\ninvestment contracts, see SEC v. Murphy, 626 F.2d 633, 640\\n\\f14 USSEC V. FENG\\n\\n(9th Cir. 1980), and a return based on a fixed rate falls within\\nthe broad definition of “profits,” SEC v. Edwards, 540 U.S.\\n389, 394 (2004); see also id. at 396 (“[T]he commonsense\\nunderstanding of ‘profits’ in the Howey test [is] simply\\n‘financial returns on . . . investments.’” (quoting Forman,\\n421 U.S. at 853)); Warfield, 569 F.3d at 1022 (holding that\\nthe investors had an expectation of profit where the\\npromotional materials emphasized “long-term income\\nproduction potential”). The Supreme Court has reasoned\\nthat the promise of a fixed rate of return should be viewed as\\ntriggering an expectation of profits because the securities\\nlaws are intended “to regulate all of the ‘countless and\\nvariable schemes devised by those who seek the use of the\\nmoney of others on the promise of profits.’” Edwards,\\n540 U.S. at 396 (quoting Howey, 328 U.S. at 299). The\\nCourt was wary that “unscrupulous marketers of\\ninvestments” might otherwise “evade the securities laws by\\npicking a rate of return to promise.” Id. at 394–95.\\n\\n The PPMs’ identification of the investments as\\nsecurities, the form of the investment entity as a limited\\npartnership, and the promise of a fixed rate of return all\\nindicate that the EB-5 transactions were securities.\\n\\nB. The Administrative Fees\\n\\n Feng insists, however, that the promised return is\\neffectively nullified by the administrative fees. When an\\ninvestor’s administrative fee and capital contribution are\\nconsidered together, Feng contends, the investor could not\\nhave expected to make any profit. The record, however,\\ndoes not support treating the administrative fee as part of the\\ninvestment. The PPMs expressly distinguish those two\\ncategories of funds, describing the investment as the “capital\\ncontribution,” which would be “put at risk,” or utilized by,\\nthe regional center. By contrast, the PPMs explain that the\\n\\f USSEC V. FENG 15\\n\\nadministrative fee would be used to defray marketing and\\noperating expenses.\\n\\n Importantly, the separation between the capital\\ncontribution and the administrative fee is necessary to satisfy\\nthe EB-5 regulatory requirements. The USCIS requires that\\na qualifying EB-5 investment place capital “at risk for the\\npurpose of generating a return on the capital placed at risk.”\\n8 C.F.R. § 204.6(j)(2). In other words, to be eligible for the\\nEB-5 program, the offering’s terms must include a possible\\nreturn on the capital contribution. Hence, even if Feng is\\ncorrect that, as a mathematical proposition, combining the\\ncapital contribution and administrative fee would eliminate\\na return on the investment, the PPMs make clear that the two\\ncategories of funds are to be viewed separately.\\n\\n Feng himself confirmed these separate identities when\\nthe USCIS questioned whether one of his clients met the EB-\\n5 requirements. In a letter to the USCIS, Feng stated that the\\nclient’s capital contribution was separate from the\\nadministrative fee and was therefore put “at risk for the\\npurpose of generating a return.” The client, Feng asserted,\\n“made an at-risk capital investment of $500,000 . . . . No\\nportion of an investor’s contribution has been or will be\\napplied towards legal service fees or administrative fee[s],\\nmarketing costs or interest payment of [the regional center],\\nwhich are paid from the separate $50,000 administrative\\nfee.”\\n\\n Feng thus attempts to characterize the administrative\\nfees in two different ways to suit his own purposes. On the\\none hand, he explained that the administrative fee should be\\nexcluded from the “investment” to establish that the\\ninvestment satisfies the EB-5 requirement of a possible\\nreturn. On the other hand, he seeks to include the\\nadministrative fee in the securities analysis to show there is\\n\\f16 USSEC V. FENG\\n\\nno possibility of generating the return as required by the\\nHowey test. That inconsistent approach is plainly contrary\\nto Congress’s intent in enacting the securities laws, as it\\nwould empower “unscrupulous marketers of investments” to\\nstructure deals to evade the designation of a security.\\nEdwards, 540 U.S. at 394. Feng’s argument that the\\nadministrative fee upends the expectation of profits has no\\nmerit.\\n\\nC. Motivation\\n\\n We likewise find unsupportable Feng’s assertion that—\\nwhatever the form of the investment and the rate of return—\\nhis clients nonetheless lacked an expectation of profit\\nbecause they were motivated to participate in the EB-5\\nprogram by the promise of visas, not by profit. See Warfield,\\n569 F.3d at 1021 (holding that the focus of Howey is the\\n“objective inquiry” into “what the purchasers were offered\\nor promised,” although the investors’ subjective intent “may\\nhave some bearing on the issue of whether they entered into\\ninvestment contracts”). We do not doubt that the investors’\\nprimary reason to participate in the EB-5 program was to\\ngain U.S. visas. But Feng oversimplifies his clients’\\nmotivations.\\n\\n An EB-5 investor’s interest in a visa is inextricably tied\\nto the financial success of the regional center’s project. The\\nEB-5 program by design links the success of the investment\\nto the investor’s success in obtaining a visa. An EB-5\\napplicant must prove to the USCIS that an investment in fact\\nled to the creation of at least ten full-time jobs. 8 U.S.C.\\n§ 1153(b)(5)(A); 8 C.F.R. §§ 204.6, 216.6. Understanding\\nthe importance of a successful investment, Feng marketed\\nhis ability to evaluate the multitude of regional centers and\\nrecommend projects that were most likely to be sustained\\nlong enough to create the requisite jobs to qualify the\\n\\f USSEC V. FENG 17\\n\\ninvestor for the EB-5 visa. The record also contains\\ntestimony from some of Feng’s client-investors who said\\nthat the likelihood of the regional center returning their\\ncapital contribution was important to them or that they\\nsought EB-5 projects with higher rates of return. Feng’s\\nargument thus relies on a false dichotomy between the visa\\nand the success of the investment. 5\\n\\n We therefore agree with the district court that the\\ninvestments in this case constitute securities in the form of\\ninvestment contracts.\\n\\n III.\\n\\n Having resolved the threshold securities issue, we next\\nconsider Feng’s challenges to the grant of summary\\njudgment for the SEC on each of the three causes of action\\nbrought against him.\\n\\nA. Failure to Register as a Broker\\n\\n Section 15(a)(1) of the Exchange Act makes it unlawful\\nfor a “broker . . . to induce or attempt to induce the purchase\\nor sale of[] any security” without registering with the SEC.\\n15 U.S.C. § 78o(a)(1). The statute defines “broker” as “any\\n\\n\\n\\n\\n 5\\n In a recent memorandum disposition, we rejected the argument that\\nEB-5 investments do not constitute securities where the investors’\\ninterest was to obtain visas rather than profits. SEC v. Liu, 754 F. App’x\\n505, 507 (9th Cir. 2018). The facts in Liu mirror those in this case, with\\na series of $500,000 EB-5 investments and accompanying $45,000\\nadministrative fees. Id. The court held that the promise of an annual\\nreturn of only 0.25 percent was enough to establish an expectation of\\nprofit, “[e]ven if it was not [the investors’] primary motivation.” Id.\\n\\f18 USSEC V. FENG\\n\\nperson engaged in the business of effecting transactions in\\nsecurities for the account of others.” Id. § 78c(a)(4)(A).\\n\\n The district court concluded that the uncontroverted\\nevidence establishes that Feng was acting as a broker and\\nviolated the registration requirement. We agree.\\n\\n 1. The Broker Determination\\n\\n In making its broker determination, the district court\\nutilized the totality-of-the-circumstances approach that other\\ncourts have used under Section 15(a)(1), relying on a set of\\nfactors first set forth in SEC v. Hansen, No. 83 Civ. 3692,\\n1984 WL 2413, at *10 (S.D.N.Y. Apr. 6, 1984). See SEC v.\\nCollyard, 861 F.3d 760, 766 (8th Cir. 2017); SEC v. George,\\n426 F.3d 786, 797 (6th Cir. 2005); see also SEC v. Imperiali,\\nInc., 594 F. App’x 957, 961 (11th Cir. 2014) (per curiam)\\n(quoting George, 426 F.3d at 797). Courts emphasize that\\nthe so-called Hansen factors are “nonexclusive.” Collyard,\\n861 F.3d at 766.\\n\\n In performing its inquiry, the district court considered\\nwhether Feng:\\n\\n (1) is an employee of the issuer of the\\n security;\\n\\n (2) received transaction-based income such\\n as commissions rather than a salary;\\n\\n (3) sells or sold securities from other issuers;\\n\\n (4) was involved in negotiations between\\n issuers and investors;\\n\\n (5) advertis[ed] for clients;\\n\\f USSEC V. FENG 19\\n\\n (6) gave advice or made valuations regarding\\n the investment;\\n\\n (7) was an active finder of investors; and\\n\\n (8) regularly participates in securities\\n transactions.\\n\\nSEC v. Feng, No. 15-09420, 2017 WL 6551107, at *7–8\\n(C.D. Cal. Aug. 10, 2017) (footnote omitted). 6 The district\\ncourt found that, although Feng was not an employee of the\\nissuer, his conduct satisfied all the other factors. Id. at *8.\\nHe received commissions from regional centers, sought\\ninvestors and clients through his website and other online\\nforums, negotiated with regional centers about the terms of\\nthe projects, and gave advice about EB-5 projects’ likelihood\\nof success.\\n\\n Although Feng does not contest the facts underlying\\nthese conclusions, he objects to the factor-based analysis\\ngenerally, arguing that the factors are so vague that they\\nerroneously encompass individuals who are merely\\nproviding legal advice in the EB-5 context. 7 To the contrary,\\n\\n 6\\n The district court’s eight factors essentially tracked the\\nconsiderations identified in Hansen. Feng, 2017 WL 6551107, at *7–8;\\nHansen, 1984 WL 2413, at *10.\\n 7\\n Feng also argues that the district court’s factors omit the\\nrequirement that a broker must have authority or control over the\\naccounts of others. But the caselaw cited by Feng does not impose such\\na requirement as a prerequisite for finding that someone is a broker. See\\nSEC v. Kramer, 778 F. Supp. 2d 1320, 1339–40 (M.D. Fla. 2011)\\n(discussing authority over an account as one of several factors); SEC v.\\nM & A West, Inc., No. C-01-3376 VRW, 2005 WL 1514101, at *9 (N.D.\\nCal. June 20, 2005) (same); cf. SEC v. Mapp, 240 F. Supp. 3d 569, 592\\n\\f20 USSEC V. FENG\\n\\nthe application of the Hansen factors reveals that much of\\nthe work Feng performed for his clients, rather than being\\ntraditional legal work, aligns with the indicia of broker\\nactivity identified by those factors.\\n\\n For example, Feng asserts that the factor “received\\ntransaction-based income such as commissions rather than a\\nsalary” is unclear and could improperly include an attorney’s\\ncontingent fee. However, Feng charged his clients a\\nnoncontingent, upfront fee. And it is undisputed that Feng\\nand his nominees received $1.268 million in commissions\\nfrom the regional centers—not his legal clients—for\\nrecruiting investors to the regional centers’ EB-5 projects.\\nOn these facts, the “commissions” factor does not threaten\\nto mislabel traditional legal services as broker activity.\\n\\n Similarly, Feng suggests that “involve[ment] in\\nnegotiations between issuers and investors” should not\\ninclude his legal work explaining to clients aspects of legal\\ncompliance and regional center communications. Feng\\nignores, however, the evidence that he communicated\\nextensively with the regional centers, posing numerous\\nquestions about the terms of the projects and negotiating the\\namount of the administrative fees.\\n\\n In addition, Feng contends that the factor “gave advice\\nor made valuations regarding the investment” should not\\ninclude legal advice about the merits of the various EB-5\\nprojects. He claims that he was advising his clients about\\nhow to obtain legal permanent residency status, not about\\n\\n\\n(E.D. Tex. 2017) (discussing but not deciding whether control over the\\naccount of others is an element or a nondispositive factor). Further,\\nassuming that control is a pertinent factor, it is present here in the\\nmultiple instances in which client funds passed through Feng’s accounts.\\n\\f USSEC V. FENG 21\\n\\nhow to make money. The record, however, indicates that\\nFeng’s services included investment advice. He reviewed\\nand promoted certain EB-5 projects based on their likelihood\\nof success under the EB-5 requirements, with particular\\nattention to the projects’ likely ability to create the requisite\\nnumber of jobs over time to qualify the investor for an EB-5\\nvisa, as well as the likelihood the projects could return the\\ninvestors’ capital contributions. This work goes beyond the\\ntraditional work of a lawyer.\\n\\n Finally, Feng suggests that the factor “active finder of\\ninvestors” improperly conflates seeking legal clients with\\nseeking investors. This argument, too, is meritless. Feng\\nsought a particular kind of client—one who would become\\nan investor with a regional center with which he had a\\nmarketing agreement and a reserved investor slot. Feng’s\\n150 clients invested more than $65 million. In this EB-5\\ncontext, his legal clients and the regional center investors\\nwere, in fact, one and the same.\\n\\n 2. Feng’s Additional Arguments\\n\\n Feng goes beyond the Hansen analysis to argue more\\ngenerally that the broker registration requirement should not\\napply to his circumstances. First, he asserts that the\\nrequirement should apply only to individuals who trade\\nsecurities on an exchange and not to those involved in\\ntransactions between private parties—deals commonly\\nreferred to as “over-the-counter.” It is well established,\\nhowever, that “[t]he 1934 [Exchange] Act was intended\\nprincipally to protect investors against manipulation of stock\\nprices through regulation of transactions upon securities\\nexchanges and in over-the-counter markets.” Ernst & Ernst\\nv. Hochfelder, 425 U.S. 185, 195 (1976) (emphasis added)\\n(citing S. Rep. No. 792, at 1–5 (1934)). Although the\\nExchange Act exempts brokers “whose business is\\n\\f22 USSEC V. FENG\\n\\nexclusively intrastate and who do[] not make use of any\\nfacility of a national securities exchange,” Feng’s business\\nwas not “exclusively intrastate.” See 15 U.S.C. § 78o(a)(1).\\n\\n Second, Feng argues that the broker registration\\nrequirement is unnecessary in his circumstances because\\nregistered-broker duties duplicate the fiduciary duties that\\napply to the attorney-client relationship. Feng analogizes to\\nthe registration exemption for lawyers under the Investment\\nAdvisers Act of 1940, which excludes from the definition of\\nan investment adviser “any lawyer, accountant, engineer, or\\nteacher whose performance of such services is solely\\nincidental to the practice of his profession.” 15 U.S.C.\\n§ 80b-2(a)(11)(B). Feng asserts—without citation—that the\\nInvestment Advisers Act exempts attorneys because of their\\nheightened fiduciary duty to clients, and he urges us to apply\\n“this rationale” to create an attorney exception to the broker\\nregistration requirement in the Exchange Act.\\n\\n This argument fails for multiple reasons. As we have\\nalready explained, even if such an exemption existed, the\\nrecord belies any claim that Feng’s activities were “solely\\nincidental to the practice of [law].” Moreover, if an\\nattorney’s fiduciary duty were the rationale for the\\nexemption under the Investment Advisers Act, the incidental\\nnature of the investment-related work would seem irrelevant\\nbecause the fiduciary duty presumably would cover any\\nwork on behalf of clients. Feng essentially asks us to\\nperform a legislative act by adding an exemption to one\\nstatute on the basis of a different statute for which Congress\\nexpressly provided an exemption. Self-evidently, we cannot\\ndo that.\\n\\n In a last-ditch attempt to justify his failure to register,\\nFeng argues that imposing the broker registration\\nrequirement on U.S.-based attorneys would “force” them\\n\\f USSEC V. FENG 23\\n\\n“not to help these people.” This argument poses a false\\nchoice. Attorneys who provide only legal advice to clients\\nabout the EB-5 program are not required to register.\\nAttorneys who act as brokers in the EB-5 context may\\nlegally do so only if they register with the SEC.\\n\\n 3. Conclusion\\n\\n Given the undisputed facts in the record, we agree with\\nthe district court that Feng, as a matter of law, was “engaged\\nin the business of effecting transactions in securities for the\\naccount of others” and thus was required to register with the\\nSEC as a broker. 15 U.S.C. §§ 78c(4)(A), 78o(a)(1). 8\\n\\nB. The Fraud Determinations\\n\\n Feng argues that the district court erroneously found that\\nhe committed securities fraud in violation of Section 17 of\\nthe Securities Act and Section 10(b) of the Exchange Act, as\\nimplemented through Rule 10b-5. The district court’s\\ndeterminations were based on two theories of fraud liability:\\nmaterial omissions and schemes to defraud. We review each\\nin turn.\\n\\n\\n\\n\\n 8\\n Feng also appeals the denial of his motion for judgment on the\\npleadings, arguing that the terms “security” and “broker” in Section\\n15(a) of the Exchange Act are unconstitutionally vague. A word is not\\nvague when it has a “settled legal meaning.” United States v. Williams,\\n553 U.S. 285, 306 (2008). Section 15(a) was enacted 80 years ago, and\\nit has been applied countless times by the courts. There was no error in\\nthe district court’s denial of the motion for judgment on the pleadings.\\n\\f24 USSEC V. FENG\\n\\n 1. Material Omissions (Section 17(a)(2) and Rule\\n 10b-5(b))\\n\\n The antifraud provisions of the Securities Act, Section\\n17(a)(2), and the Exchange Act, Section 10(b), developed\\nthrough Rule 10b-5(b), “forbid making a material\\nmisstatement or omission in connection with the offer or sale\\nof a security by means of interstate commerce.” SEC v. Dain\\nRauscher, Inc., 254 F.3d 852, 856 (9th Cir. 2001). Under\\nthese provisions, it is unlawful for any person, in the offer or\\nsale of securities, “to obtain money or property by means of\\nany untrue statement of a material fact or any omission to\\nstate a material fact necessary in order to make the\\nstatements made, in light of the circumstances under which\\nthey were made, not misleading.” 15 U.S.C. § 77q(a)(2).\\nThe text of Rule 10b-5 largely mirrors the text of Section\\n17(a). 9 Although there are differences in the state of mind\\nrequirements for Rule 10b-5(b) and Section 17(a)(2), a\\nshowing of intentional or knowing conduct satisfies both.\\nSee Aaron v. SEC, 446 U.S. 680, 695–97 (1980) (holding\\nthat scienter is an element of Section 10(b), Rule 10b-5, and\\nSection 17(a)(1), but not Section 17(a)(2) or (3)).\\n\\n A fiduciary duty to disclose certain information renders\\nan omission of that information misleading. See United\\nStates v. Laurienti, 611 F.3d 530, 541 (9th Cir. 2010). In\\nLaurienti, the court addressed a broker’s liability under Rule\\n10b-5(b) for failure to disclose commissions to his clients.\\nThe court noted that, while the broker/client relationship\\n 9\\n Section 17(a) of the Securities Act prohibits fraud in the sale of\\nsecurities. Rule 10b-5, adopted by the SEC pursuant to its authority\\nunder Section 10(b) of the Exchange Act, incorporates the operative\\nlanguage of Section 17(a) explicitly and also extends the prohibition to\\nfraud committed in the purchase of securities. See Ernst & Ernst,\\n425 U.S. at 212 n.32.\\n\\f USSEC V. FENG 25\\n\\ndoes not itself impose a fiduciary duty, circumstances may\\ncreate “a relationship of trust and confidence.” Id. at 540.\\nOften, a broker has a fiduciary duty when he has\\ndiscretionary authority over a client’s account, “but we have\\nrecognized that particular factual circumstances may serve\\nto create a fiduciary duty between a broker and his customer\\neven in the absence of a discretionary account.” Id. (quoting\\nUnited States v. Skelly, 442 F.3d 94, 98 (2d Cir. 2006)). The\\ncourt held that, however the duty arises, “if a broker and a\\nclient have a trust relationship, . . . then the broker has an\\nobligation to disclose all facts material to that relationship.”\\nId. at 541.\\n\\n Here, as we have determined, Feng worked as both a\\nbroker and an immigration attorney. Indeed, Feng\\nemphasizes his role as an attorney and acknowledges that, as\\nsuch, he had fiduciary duties to his clients. Those duties\\nincluded the obligation to disclose conflicts of interest. See\\nN.Y. Rules of Prof’l Conduct (22 NYCRR § 1200.0) 1.4\\n(duty to communicate any information needed by the client\\nto give informed consent), 1.7 (duty to disclose a conflict),\\n1.8 (duty to disclose business relationships).\\n\\n Feng’s argument that there is no conflict because his\\nclients were not harmed applies the wrong standard. The\\nconflict of interest rule addresses the risk of harm to clients,\\nasking whether a “reasonable lawyer would conclude . . .\\nthere is a significant risk that the lawyer’s professional\\njudgment on behalf of a client will be adversely affected by\\nthe lawyer’s own financial, business, property or other\\npersonal interests.” N.Y. Rules of Prof’l Conduct 1.7(a). In\\nthese transactions, there was a risk that Feng’s judgment\\nwould be swayed by the promise of commissions—that, for\\nexample, he would prefer a regional center that paid a higher\\ncommission but had a slightly lower likelihood of success\\n\\f26 USSEC V. FENG\\n\\nover a regional center that offered a lower commission but\\nhad a better opportunity for the client.\\n\\n Indeed, when Feng was asked in his deposition whether\\nthe commission from the regional center and his\\nrepresentation of a client presented a conflict, he\\nacknowledged that it did:\\n\\n Yes. I see it as sort of like a conflict because\\n we’re in between, between regional center\\n and the client. I’m aware of getting legal fees\\n from the client, but also, even though we are\\n not issuer’s counsel, we are definitely not\\n regional center’s counsel. But regional\\n centers pay us after the 526 getting approved.\\n Even though we realize there’s such a\\n conflict, but our interest is first and foremost\\n with the clients.\\n\\nHis assertion that he shared his clients’ goal of obtaining a\\ngreen card does not eliminate the risk of the commission’s\\ninfluence on his professional judgment. Further, the rules\\nprovide that only the client, fully informed of both the\\nconflict and the risks stemming from the conflict, may make\\na judgment about whether to accept representation\\nnotwithstanding the conflict. N.Y. Rules of Prof’l Conduct\\n1.4, 1.7.\\n\\n “[E]ven in a trust relationship,” the duty is “to disclose\\nonly material facts.” See Laurienti, 611 F.3d at 541.\\nGenerally, an omission is material “if there is ‘a substantial\\nlikelihood that the disclosure of the omitted fact would have\\nbeen viewed by the reasonable investor as having\\nsignificantly altered the “total mix” of information made\\navailable.’” SEC v. Phan, 500 F.3d 895, 908 (9th Cir. 2007)\\n(quoting Basic Inc. v. Levinson, 485 U.S. 224, 231–32\\n\\f USSEC V. FENG 27\\n\\n(1988)). Here, the record includes statements of individual\\ninvestors who said that if they had known about Feng’s\\ncommissions, they would have attempted to negotiate their\\nadministrative fees. Feng also testified that he concealed his\\ncommissions to avoid anticipated client requests for rebates.\\nHe said that he thought disclosure of his commissions would\\nencourage the clients to ask for more reductions in their\\nadministrative fees, and he did not want to open the door to\\nnegotiations that would lower his clients’ costs but decrease\\nhis payout. We thus agree with the district court that Feng’s\\nfailure to disclose his conflict of interest was material. See\\nTSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 450 (1976)\\n(holding that materiality may be resolved at summary\\njudgment “if the established omissions are so obviously\\nimportant to an investor[] that reasonable minds cannot\\ndiffer on the question of materiality” (internal quotation\\nmarks omitted)).\\n\\n Finally, Feng’s testimony establishes intent to deceive.\\nFeng stated that he did not want to tell clients about his\\ncommissions from the regional centers because it would be\\ncostly. Feng’s brief similarly describes his choice not to\\ndisclose as a “business decision.”\\n\\n 2. Schemes to Defraud (Section 17(a)(1) and (3) and\\n Rule 10b-5(a) and (c))\\n\\n The securities fraud provisions also prohibit any person,\\nin the offer or sale of securities, “to employ any device,\\nscheme, or artifice to defraud” or “to engage in any\\ntransaction, practice, or course of business which operates or\\nwould operate as a fraud or deceit upon the purchaser.”\\n15 U.S.C. § 77q(a)(1), (3); see also 17 C.F.R. § 240.10b-\\n5(a), (c). The state of mind requirement varies among these\\nprovisions, but a showing of intentional or knowing conduct\\nclears all thresholds. See Aaron, 446 U.S. at 695–97.\\n\\f28 USSEC V. FENG\\n\\n The district court found that Feng defrauded the regional\\ncenters that refused to pay commissions to U.S.-based\\nattorneys not registered as brokers. Feng evaded this\\nrestriction by informing the regional centers that he was\\naffiliated with foreign referral agents and companies,\\nholding himself out to the regional centers as a middleman.\\nFeng furthered this scheme by setting up ABCL in China to\\nreceive commission payments, while maintaining control of\\nboth ABCL’s bank account and operations. Feng performed\\nthe recruitment services himself and received the\\ncommissions that regional centers believed they were paying\\nto the affiliated recruiters. Feng recruited his mother to sign\\ndocuments as ABCL’s president, even though she had no\\nactual role in the company. Representatives from the\\nregional centers testified they would have ceased this\\narrangement if they knew Feng’s relationship with either the\\nagents or ABCL.\\n\\n The district court also found that Feng engaged in a\\nscheme to defraud clients who sought a reduction in their\\nadministrative fees. Generally, as noted, regional centers\\nwere willing to rebate a portion of the administrative fee that\\nthey charged investors if the regional center could also\\nreduce the commission it paid for the recruitment of that\\ninvestor. When a client asked Feng to negotiate with the\\nregional center to reduce his administrative fee, Feng\\narranged with the regional center to lower the administrative\\nfee by reducing the commission. He took measures to\\nconceal this arrangement from his clients, however, because\\nhe wanted to “keep as much of the marketing fee as possible”\\nand avoid further negotiation with clients, while also\\ncreating the appearance to his clients that he was prevailing\\non their behalf with the regional centers. Feng asked the\\nregional centers to rebate a portion of the administrative fee\\n\\f USSEC V. FENG 29\\n\\nbut not to disclose that the funds were reallocated from a\\ncommission.\\n\\n Feng challenges on two grounds the district court’s\\nfinding that his conduct violated the Securities and Exchange\\nActs. First, Feng asserts that the scheme-to-defraud\\nallegations under Rule 10b-5(a) and (c) improperly duplicate\\nthe Rule 10b-5(b) material omission claim. We see no\\nduplication. The district court addressed the fraud against\\nthe regional centers only as a scheme to defraud, not as a\\nmaterial omission. With regard to the fraud against the\\nclients, the district court focused on Feng’s failure to\\ndisclose his conflict of interest to his clients as material\\nomissions and, separately, on his efforts to conceal the\\nsource of the rebates as a scheme to defraud.\\n\\n Second, Feng argues that his establishment of overseas\\nentities and occasional arrangement of rebates for clients\\nwere not unlawful. But, deceptive conduct that is not\\n“inherently unlawful” may form the basis of a scheme to\\ndefraud. SEC v. Wey, 246 F. Supp. 3d 894, 918 (S.D.N.Y.\\n2017). In addition, Feng’s attempt to analogize the rebate\\nscheme to a store proprietor giving a customer a discount is\\ninapt. The fraudulent conduct was his scheme to make the\\nrebate appear to come from his advocacy on behalf of the\\nclient, while concealing the relationship between the\\nreduction and his undisclosed commission.\\n\\n We accordingly affirm the district court’s finding that\\nFeng engaged in securities fraud in violation of Section 17(a)\\nand Section 10(b).\\n\\n IV.\\n\\n Finally, Feng challenges the district court’s\\ndisgorgement order. The district court calculated that Feng\\n\\f30 USSEC V. FENG\\n\\nreceived $1.268 million for commissions in connection with\\nthe EB-5 program and ordered disgorgement of the entire\\namount. Feng contests the disgorgement of fees paid to his\\noverseas entities, suggesting that some of the fees were used\\nto cover expenses. He does not dispute that he controlled\\nthose overseas entities.\\n\\n We review the court’s imposition of disgorgement for\\nabuse of discretion. SEC v. JT Wallenbrock & Assocs.,\\n440 F.3d 1109, 1113 (9th Cir. 2006). A disgorgement\\ncalculation “should include ‘all gains flowing from the\\nillegal activities,’” id. at 1114 (quoting SEC v. Cross Fin.\\nServs., 908 F. Supp. 718, 734 (C.D. Cal. 1995), and “requires\\nonly a ‘reasonable approximation of profits causally\\nconnected to the violation,’” id. at 1113–14 (quoting SEC v.\\nFirst Pac. Bancorp, 142 F.3d 1186, 1192 n. 6 (9th Cir.\\n1998)).\\n\\n Feng used his overseas entities—ABCL and his other\\nnominees—to advance his scheme to defraud the regional\\ncenters. The money paid by regional centers to Feng’s\\noverseas entities is therefore exactly the focus of the fraud\\nagainst the regional centers, such that the district court could\\nfind a “causal[] connect[ion]” through which the funds\\n“flow[] from the illegal activities.” Id. at 1114. Feng should\\nnot have been collecting these commissions in the first place,\\nand it would be unjust to permit him to retain some of the ill-\\ngotten funds to cover his expenses. See id. (“[I]t would be\\nunjust to permit the defendants to offset against the investor\\ndollars they received the expenses of running the very\\nbusiness they created to defraud those investors into giving\\nthe defendants the money in the first place.”). We\\naccordingly find no abuse of discretion in the disgorgement\\norder.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432768/\", \"author_raw\": \"LIPEZ, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"KERMIT V LIPEZ"},"judge2":{"kind":"string","value":"KIM MCLANE WARDLAW"},"judge3":{"kind":"string","value":"ANDREW D HURWITZ"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4655515/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Ussec v. Hui Feng\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-23\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n U.S. SECURITIES & EXCHANGE No. 17-56522\n COMMISSION,\n Plaintiff-Appellee, D.C. No.\n 2:15-cv-09420-\n v. CBM-SS\n\n HUI FENG; LAW OFFICES OF FENG\n AND ASSOCIATES PC, OPINION\n Defendants-Appellants.\n\n Appeal from the United States District Court\n for the Central District of California\n Consuelo B. Marshall, District Judge, Presiding\n\n Argued and Submitted May 16, 2019\n Pasadena, California\n\n Filed August 23, 2019\n\n Before: Kermit V. Lipez, * Kim McLane Wardlaw,\n and Andrew D. Hurwitz, Circuit Judges.\n\n Opinion by Judge Lipez\n\n\n\n *\n The Honorable Kermit V. Lipez, United States Circuit Judge for\nthe First Circuit, sitting by designation.\n\f2 USSEC V. FENG\n\n SUMMARY **\n\n\n Securities & Exchange Commission\n\n The panel affirmed the district court’s summary\njudgment in favor of the U.S. Securities and Exchange\nCommission (“SEC”) in its civil complaint filed against Hui\nFeng and his law firm, alleging securities fraud.\n\n The U.S. Immigrant Investor Program, also known as the\nEB-5 program, provides legal permanent residency in the\nUnited States to foreign nationals who invest in U.S.-based\nprojects. Multiple foreign investors may pool their money\nin the same enterprise, and these pooled investments are\nmade through “regional centers” which are regulated by the\nU.S. Citizenship and Immigration Services.\n\n Feng legally represented clients through the EB-5\nprocess, and entered into marketing agreements with\nregional centers. The basis of the agreements between the\nregional centers and Feng’s investors were known as private\nplacement memoranda (“PPMs”).\n\n The panel agreed with the district court that the EB-5\ninvestments in this case constituted “securities” in the form\nof investment contracts. The panel rejected Feng’s argument\nthat the transactions were not “securities” because his clients\ndid not expect profits from their investments. Specifically,\nthe panel held that the PPMs’ identification of the\ninvestments as securities, the form of the investment entity\n\n **\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f USSEC V. FENG 3\n\nas a limited partnership, and the promise of a fixed rate of\nreturn all indicated that the EB-5 transactions were\nsecurities. The panel rejected Feng’s contention that the\nadministrative fees upended the expectation of profits. The\npanel also rejected Feng’s assertion that his clients lacked an\nexpectation of profit because they were motivated to\nparticipate in the EB-5 program by the promise of visas, not\nby profit.\n\n Concerning the cause of action that Feng failed to\nregister as a broker in violation of Section 15(a) of the\nSecurities and Exchange Act of 1934, the panel agreed with\nthe district court’s conclusion that Feng was acting as a\nbroker and violated the registration requirement. The panel\nheld that the district court properly made its broker\ndetermination by utilizing the totality-of-the-circumstances\napproach by relying on the so-called Hansen factors. The\npanel rejected Feng’s arguments that the broker registration\nrequirement should not apply to his circumstances.\n\n The panel affirmed the district court’s finding that Feng\nengaged in securities fraud in violation of Section 17(a) of\nthe Securities Act of 1933 and Section 10(b) of the Securities\nExchange Act of 1934 based on two theories of fraud\nliability: material omissions, and schemes to defraud.\nConcerning material omissions, the panel held that where\nFeng worked as both a broker and an immigration attorney,\nhe had fiduciary duties to his clients, including the obligation\nto disclose conflicts of interest. Here, there was a risk that\nFeng’s judgment would be swayed by the promise of a\ncommission from the regional center, and this presented a\nconflict with Feng’s representation of a client, which he\nfailed to disclose to his clients, and this failure was material.\nConcerning schemes to defraud, the panel affirmed the\ndistrict court’s findings that Feng defrauded the regional\n\f4 USSEC V. FENG\n\ncenters that refused to pay commissions to U.S.-based\nattorneys not registered as brokers, and defrauded clients\nwho sought a reduction in their administrative fees.\n\n Finally, the panel held that the district court did not abuse\nits discretion in entering a disgorgement order. The district\ncourt calculated that Feng received $1.268 million for\ncommissions in connection with the EB-5 program, and\nordered disgorgement of the entire amount.\n\n\n COUNSEL\n\nAndrew B. Holmes (argued) and Matthew D. Taylor,\nHolmes Taylor Scott & Jones LLP, Los Angeles, California,\nfor Defendant-Appellant Law Offices of Feng & Associates.\n\nHui Feng (argued), Law Offices of Feng & Associates,\nFlushing, New York, for Defendant-Appellant Hui Feng.\n\nKerry J. Dingle (argued), Senior Counsel; Jeffery A. Berger,\nSenior Litigation Counsel; Michael A. Conley, Solicitor;\nRobert B. Stebbins, General Counsel; Securities &\nExchange Commission, Washington, D.C.; for Plaintiff-\nAppellee.\n\f USSEC V. FENG 5\n\n OPINION\n\nLIPEZ, Circuit Judge:\n\n This appeal requires us to decide whether certain\ninvestments made by participants in the U.S. Immigrant\nInvestor Program are “securities” subject to regulation by the\nSecurities and Exchange Commission (“SEC”) and, if so,\nwhether appellant Hui Feng was a securities “broker”\nrequired to register with the SEC and whether he committed\nsecurities fraud in connection with the transactions. 1 The\ndistrict court granted summary judgment for the SEC. We\naffirm.\n\n I.\n\nA. The EB-5 Program\n\n The U.S. Immigrant Investor Program, colloquially\nreferred to as the EB-5 program, provides legal permanent\nresidency in the United States to foreign nationals who\ninvest in U.S.-based projects. See 8 U.S.C.\n§ 1153(b)(5)(A). 2 Generally, qualified immigrants may gain\nU.S. visas through direct investment of at least $1 million in\na new commercial enterprise that creates at least ten full-\ntime jobs for U.S. workers. Id. § 1153(b)(5)(A), (C).\nInvestment in a business in a “targeted employment area”\n\n 1\n Both Hui Feng and his law firm, the Law Offices of Feng &\nAssociates P.C., are appellants in this matter. The parties’ arguments do\nnot distinguish between the two, and so, as a matter of convenience, we\nrefer to the appellants collectively as “Feng.”\n 2\n By statute, the Immigrant Investor Program is the fifth preference\nin the employment-based visa category, which gave rise to the nickname\n“EB-5.” See 8 U.S.C. § 1153(b)(5)(A).\n\f6 USSEC V. FENG\n\nlowers the required capital investment amount to $500,000.\nId. § 1153(b)(5)(C)(i), (ii).\n\n Multiple foreign investors may pool their money in the\nsame enterprise, provided that each invests the required\namount and “each individual investment results in the\ncreation of at least ten full-time positions.” 8 C.F.R.\n§ 204.6(g). Pooled investments are made through “regional\ncenters,” which are regulated by the U.S. Citizenship and\nImmigration Services (“USCIS”). The regional centers offer\nspecific projects to investors and manage the pooled\ninvestments. See id. § 204.6(e), (m).\n\n A foreign national investing in an enterprise must file an\nI-526 application with the USCIS to prove that the\ninvestment will satisfy EB-5 program requirements. Id.\n§ 204.6(a), (j)(2). Approval results in conditional permanent\nresident status. Two years later, the investor may remove\nthe conditions on lawful permanent resident status by filing\nan I-829 petition, demonstrating that the investment satisfied\nthe EB-5 requirements and created, or will create within a\nreasonable period, ten qualifying jobs. Id. § 216.6.\n\nB. Factual Background\n\n Feng and the SEC largely agree on the facts. To the\nextent that they have differing views of the record, we draw\nall justifiable inferences in the light most favorable to Feng.\nBlankenhorn v. City of Orange, 485 F.3d 463, 470 (9th Cir.\n2007).\n\n Feng conducts an immigration law practice in New York\nCity. Between 2010 and 2016, he led approximately\n150 clients through the EB-5 process, substantially all of\nwhom were Chinese nationals who did not speak English.\nFeng estimated that 80 percent of his clients found him\n\f USSEC V. FENG 7\n\nthrough his website, which posted comparisons and\nrecommendations of available EB-5 projects. Feng holds a\nlaw degree from Columbia University and an MBA from the\nTuck School of Business at Dartmouth, and he touted his\nskill at investigating and determining the most reliable\ninvestment opportunities.\n\n Feng charged his clients a $10,000 to $15,000 upfront\nfee. His services included serving as a liaison between\nclients and regional centers, explaining the English-language\noffering materials to his clients, negotiating with regional\ncenters regarding administrative fees charged to the clients\nby the centers, and compiling and submitting his clients’\nsigned offering documents to regional centers. On three\noccasions, investors wired funds directly to Feng, who\ntransferred the funds to the regional centers.\n\n Feng also entered into marketing agreements with\nregional centers. Through these agreements, he reserved a\nnumber of the limited investor “spots” in certain EB-5\nprojects, which he then promoted on his website and tried to\nfill within a set time period. If one of Feng’s clients made\nthe required capital contribution to one of these regional\ncenters, and if the USCIS approved the investor’s I-526\npetition, the regional center agreed to pay Feng commissions\nof $15,000 to $70,000. Although hundreds of regional\ncenters participate in the EB-5 program, Feng procured\ninvestors only for ten centers, all of which agreed to pay\ncommissions. Feng did not disclose to his clients the\ncommissions he received from the regional centers unless\nthey specifically asked about them. 3\n\n\n 3\n When the SEC’s investigation was pending, Feng revised his\nretainer agreement to state that regional centers may pay fees to an\n\f8 USSEC V. FENG\n\n Payment of a commission is an EB-5 industry practice,\nbut many regional centers refuse to pay a commission to a\nU.S.-based attorney who is not registered as a broker, to\navoid causing or aiding and abetting a violation of the broker\nregistration requirement in Section 15(a) of the Securities\nExchange Act of 1934. See 15 U.S.C. § 78o(a). Several\nregional centers offer a discount or rebate of a portion of the\nadministrative fee charged to the investor if there is no\nreferral commission for the attorney. Feng chose neither to\nregister as a broker, nor to forego commissions.\n\n When told by some regional centers that they could only\npay commissions to overseas recruiters, Feng responded that\nhe had referral partners in China who were finding investors.\nHe entered into marketing agreements in which he promised\nto relay the commissions to those individuals. He also\narranged agreements that were directly between the regional\ncenters and the individuals. The referral partners included\nFeng’s wife, mother, and mother-in-law. Feng described\nthese individuals in his deposition testimony as his\n“nominees,” whose role was to be “just a surrogate to receive\nthe money.” They did not do the work of recruiting\ninvestors—Feng did that himself. After the regional centers\npaid commissions to the individuals, they transferred the\nfunds to Feng. Feng asserts that the regional centers only\nasked for a foreign contact to pay and did not “care” who it\nwas.\n\n In 2014, Feng also created Atlantic Business Consulting\nLimited (“ABCL”), based in Hong Kong. According to\nFeng, this decision was “driven in part by regional centers\ninforming me they needed an overseas entity to pay.” ABCL\n\noverseas company owned by Feng upon completion of the client’s visa\napproval.\n\f USSEC V. FENG 9\n\nentered into marketing agreements with regional centers, and\nits only source of revenue was EB-5 commissions. Feng did\nnot inform the regional centers that he solicited and referred\nABCL’s investors himself. All of Feng’s employees worked\nboth for his law office and for ABCL. Feng’s mother,\nXiuyuan Tan, signed agreements between ABCL and\nregional centers as ABCL’s “President,” but Feng admitted\nshe played no role in ABCL. Feng also did not inform the\nregional centers that he was ABCL’s beneficial owner, with\nsole control of ABCL’s bank account. Representatives from\nthe regional centers testified that they would have ceased\ntheir marketing agreements if they had known about Feng’s\nrelationship with the individual referral agents and with\nABCL.\n\n In total, regional centers paid Feng and his overseas\nentities $1.268 million in commissions for investments made\nby Feng’s clients. At the time of the district court judgment,\nthese entities were contractually entitled to an additional\n$3.45 million in commissions, pending the approval of his\nclients’ I-526 petitions.\n\n The basis of the agreements between the regional centers\nand Feng’s investors was set forth in the regional centers’\noffering materials, also known as private placement\nmemoranda (“PPMs”). Many of the PPMs referred to the\ninvestments as “securities” and asserted that the investments\nwere compliant with U.S. securities laws. The regional\ncenters structured the investments as limited partnerships, in\nwhich the investors became limited partners and the regional\ncenter was the general partner. The regional centers\npromised investors a fixed, annual return on investment,\nwhich ranged across projects from 0.5 to 5 percent of the\ncapital contribution, and investors received Schedule K-1 tax\nforms to report their investment income from the\n\f10 USSEC V. FENG\n\npartnership. The partnerships all made construction loans or\notherwise financed a specified construction project. At the\nend of the investment term, typically five to six years, the\nregional centers promised the investors a return of their\ncapital contribution, subject to market risks.\n\n The PPMs required that investors pay an administrative\nfee, which ranged across projects from $30,000 to $50,000,\nin addition to the capital contribution of $1 million or\n$500,000. The PPMs expressly stated that these\nadministrative fees were for operating and marketing costs,\nwere not part of the capital contribution, and did not earn\ninterest.\n\n Approximately 20 percent of Feng’s clients asked him to\nseek a reduced administrative fee from the regional centers.\nIn those instances, Feng negotiated with regional centers and\nfacilitated contracts between those regional centers and his\nclients for a rebate of a portion of the administrative fee.\nFeng did not disclose to these clients, however, that the\nadministrative fees helped to fund his commissions, nor that\nthe regional centers offset the clients’ administrative fee\nrebate with a reduction in the commissions to which he was\ncontractually entitled. Indeed, Feng asked the regional\ncenters not to inform the clients about either fact. Feng\nexplained in his deposition that he wanted to create the\nappearance to his clients that he was advocating on their\nbehalf with the regional center and also to “keep as much of\nthe marketing fee as possible”—that is, his commission—\nand to avoid further “haggl[ing]” with clients who might ask\nFeng to rebate more of the commission. 4\n\n\n 4\n For example, one client testified that Feng sent him a check for\n$15,000, which Feng explained as a return of charges from the regional\n\f USSEC V. FENG 11\n\nC. Procedural Background\n\n The SEC filed a civil complaint against Feng and his law\nfirm on December 7, 2015. The first and second causes of\naction allege fraud under Section 17(a) of the Securities Act\nof 1933, 15 U.S.C. § 77q(a), and Section 10(b) of the\nSecurities Exchange Act of 1934, 15 U.S.C. § 78j(b) (“the\nExchange Act”), respectively. The third cause of action\nalleges failure to register as a broker-dealer under Section\n15(a) of the Exchange Act, 15 U.S.C. § 78o.\n\n The defendants filed a motion for judgment on the\npleadings, which the district court denied in August 2016.\nThe parties subsequently filed cross-motions for summary\njudgment. Feng argued that the EB-5 investments were not\n“securities” because the investors had no expectation of\nprofit—only of obtaining a green card. The SEC argued that\nthe undisputed evidence showed that Feng acted as a\n“broker” of “securities” as a matter of law. The district court\nfound no genuine dispute of material fact and granted\nsummary judgment for the SEC on all three causes of action.\nIn this appeal, Feng argues that the transactions were not\n“securities,” and so the district court erred by denying his\nmotion for summary judgment and granting the SEC’s\nmotion.\n\n We review the district court’s decision on cross-motions\nfor summary judgment de novo. Avery v. First Resolution\nMgmt. Corp., 568 F.3d 1018, 1021 (9th Cir. 2009).\nSummary judgment is appropriate when there is no genuine\ndispute of material fact and the moving party is entitled to\njudgment as a matter of law. Fed. R. Civ. P. 56(a). We\n\ncenter without disclosing that Feng had received a $70,000 commission\non the client’s $1 million investment.\n\f12 USSEC V. FENG\n\nconsider cross-motions “separately, giving the nonmoving\nparty in each instance the benefit of all reasonable\ninferences.” Zabriskie v. Fed. Nat’l Mortg. Ass’n, 912 F.3d\n1192, 1196 (9th Cir. 2019) (quoting ACLU of Nev. v. City of\nLas Vegas, 466 F.3d 784, 790–91 (9th Cir. 2006)).\n\n II.\n\n Feng contends that the district court erred in its threshold\ndetermination that the EB-5 investments at issue were\n“securities.” Because this finding is essential to all the\ncauses of action, we begin by considering whether the\ndistrict court properly made that determination as a matter of\nlaw.\n\n The district court found that the EB-5 investments were\nsecurities because they were made through investment\ncontracts between the regional centers and the investors.\nSection 3(a)(10) of the Securities Act defines the term\n“security” broadly, enumerating a long list of financial\ninstruments, including “any . . . investment contract.”\n15 U.S.C. § 78c(a)(10). Although the Securities Act does\nnot define the term “investment contract,” the parties agree\nthat the applicable test, set forth in SEC v. W. J. Howey Co.,\nrequires “a contract, transaction or scheme whereby a person\ninvests his money in a common enterprise and is led to\nexpect profits solely from the efforts of the promoter or a\nthird party.” 328 U.S. 293, 298–99 (1946). Courts applying\nHowey “conduct an objective inquiry into the character of\nthe instrument or transaction offered based on what the\npurchasers were ‘led to expect,’” including an analysis of the\npromotional materials associated with the transaction.\nWarfield v. Alaniz, 569 F.3d 1015, 1021 (9th Cir. 2009)\n(quoting Howey, 328 U.S. at 298–99).\n\f USSEC V. FENG 13\n\n It is undisputed that Feng’s clients invested money in a\ncommon enterprise managed by a third party. Feng argues\nthat the transactions nonetheless were not “securities”\nbecause the clients did not expect profits from their\ninvestments. He asserts that his clients would have\nrecognized that the modest fixed return they were due from\nthe regional centers would not exceed the significant\nadministrative fees that they paid. He further maintains that\nthe lack of expected profit was not a concern for his clients\nbecause their motivation for making an EB-5 investment was\nto obtain U.S. visas. We find Feng’s characterization of the\nEB-5 investments unpersuasive.\n\nA. The PPMs\n\n The regional centers promoted and described the EB-5\ninvestments in the PPMs. The PPMs repeatedly referred to\nthe investments as “securities” and explained that the\nofferings were made pursuant to U.S. securities laws.\nAlthough “the name given to an instrument is not\ndispositive,” “most instruments bearing th[e] traditional\ntitles [associated with securities] are likely to be covered by\nthe statutes.” United Hous. Found., Inc. v. Forman, 421 U.S.\n837, 850 (1975). Further, “the use of a traditional name such\nas ‘stocks’ or ‘bonds’ [may] lead a purchaser justifiably to\nassume that the federal securities laws apply.” Id. That\ninference inevitably would be drawn “when the underlying\ntransaction embodies some of the significant characteristics\ntypically associated with the named instrument.” Id. at 851.\n\n Such characteristics are evident and undisputed here.\nThe EB-5 transactions were structured as investments in\nlimited partnerships, with the promise of a fixed annual\nreturn on the investment, ranging from .5 to 5 percent.\nInvestments in limited partnerships generally constitute\ninvestment contracts, see SEC v. Murphy, 626 F.2d 633, 640\n\f14 USSEC V. FENG\n\n(9th Cir. 1980), and a return based on a fixed rate falls within\nthe broad definition of “profits,” SEC v. Edwards, 540 U.S.\n389, 394 (2004); see also id. at 396 (“[T]he commonsense\nunderstanding of ‘profits’ in the Howey test [is] simply\n‘financial returns on . . . investments.’” (quoting Forman,\n421 U.S. at 853)); Warfield, 569 F.3d at 1022 (holding that\nthe investors had an expectation of profit where the\npromotional materials emphasized “long-term income\nproduction potential”). The Supreme Court has reasoned\nthat the promise of a fixed rate of return should be viewed as\ntriggering an expectation of profits because the securities\nlaws are intended “to regulate all of the ‘countless and\nvariable schemes devised by those who seek the use of the\nmoney of others on the promise of profits.’” Edwards,\n540 U.S. at 396 (quoting Howey, 328 U.S. at 299). The\nCourt was wary that “unscrupulous marketers of\ninvestments” might otherwise “evade the securities laws by\npicking a rate of return to promise.” Id. at 394–95.\n\n The PPMs’ identification of the investments as\nsecurities, the form of the investment entity as a limited\npartnership, and the promise of a fixed rate of return all\nindicate that the EB-5 transactions were securities.\n\nB. The Administrative Fees\n\n Feng insists, however, that the promised return is\neffectively nullified by the administrative fees. When an\ninvestor’s administrative fee and capital contribution are\nconsidered together, Feng contends, the investor could not\nhave expected to make any profit. The record, however,\ndoes not support treating the administrative fee as part of the\ninvestment. The PPMs expressly distinguish those two\ncategories of funds, describing the investment as the “capital\ncontribution,” which would be “put at risk,” or utilized by,\nthe regional center. By contrast, the PPMs explain that the\n\f USSEC V. FENG 15\n\nadministrative fee would be used to defray marketing and\noperating expenses.\n\n Importantly, the separation between the capital\ncontribution and the administrative fee is necessary to satisfy\nthe EB-5 regulatory requirements. The USCIS requires that\na qualifying EB-5 investment place capital “at risk for the\npurpose of generating a return on the capital placed at risk.”\n8 C.F.R. § 204.6(j)(2). In other words, to be eligible for the\nEB-5 program, the offering’s terms must include a possible\nreturn on the capital contribution. Hence, even if Feng is\ncorrect that, as a mathematical proposition, combining the\ncapital contribution and administrative fee would eliminate\na return on the investment, the PPMs make clear that the two\ncategories of funds are to be viewed separately.\n\n Feng himself confirmed these separate identities when\nthe USCIS questioned whether one of his clients met the EB-\n5 requirements. In a letter to the USCIS, Feng stated that the\nclient’s capital contribution was separate from the\nadministrative fee and was therefore put “at risk for the\npurpose of generating a return.” The client, Feng asserted,\n“made an at-risk capital investment of $500,000 . . . . No\nportion of an investor’s contribution has been or will be\napplied towards legal service fees or administrative fee[s],\nmarketing costs or interest payment of [the regional center],\nwhich are paid from the separate $50,000 administrative\nfee.”\n\n Feng thus attempts to characterize the administrative\nfees in two different ways to suit his own purposes. On the\none hand, he explained that the administrative fee should be\nexcluded from the “investment” to establish that the\ninvestment satisfies the EB-5 requirement of a possible\nreturn. On the other hand, he seeks to include the\nadministrative fee in the securities analysis to show there is\n\f16 USSEC V. FENG\n\nno possibility of generating the return as required by the\nHowey test. That inconsistent approach is plainly contrary\nto Congress’s intent in enacting the securities laws, as it\nwould empower “unscrupulous marketers of investments” to\nstructure deals to evade the designation of a security.\nEdwards, 540 U.S. at 394. Feng’s argument that the\nadministrative fee upends the expectation of profits has no\nmerit.\n\nC. Motivation\n\n We likewise find unsupportable Feng’s assertion that—\nwhatever the form of the investment and the rate of return—\nhis clients nonetheless lacked an expectation of profit\nbecause they were motivated to participate in the EB-5\nprogram by the promise of visas, not by profit. See Warfield,\n569 F.3d at 1021 (holding that the focus of Howey is the\n“objective inquiry” into “what the purchasers were offered\nor promised,” although the investors’ subjective intent “may\nhave some bearing on the issue of whether they entered into\ninvestment contracts”). We do not doubt that the investors’\nprimary reason to participate in the EB-5 program was to\ngain U.S. visas. But Feng oversimplifies his clients’\nmotivations.\n\n An EB-5 investor’s interest in a visa is inextricably tied\nto the financial success of the regional center’s project. The\nEB-5 program by design links the success of the investment\nto the investor’s success in obtaining a visa. An EB-5\napplicant must prove to the USCIS that an investment in fact\nled to the creation of at least ten full-time jobs. 8 U.S.C.\n§ 1153(b)(5)(A); 8 C.F.R. §§ 204.6, 216.6. Understanding\nthe importance of a successful investment, Feng marketed\nhis ability to evaluate the multitude of regional centers and\nrecommend projects that were most likely to be sustained\nlong enough to create the requisite jobs to qualify the\n\f USSEC V. FENG 17\n\ninvestor for the EB-5 visa. The record also contains\ntestimony from some of Feng’s client-investors who said\nthat the likelihood of the regional center returning their\ncapital contribution was important to them or that they\nsought EB-5 projects with higher rates of return. Feng’s\nargument thus relies on a false dichotomy between the visa\nand the success of the investment. 5\n\n We therefore agree with the district court that the\ninvestments in this case constitute securities in the form of\ninvestment contracts.\n\n III.\n\n Having resolved the threshold securities issue, we next\nconsider Feng’s challenges to the grant of summary\njudgment for the SEC on each of the three causes of action\nbrought against him.\n\nA. Failure to Register as a Broker\n\n Section 15(a)(1) of the Exchange Act makes it unlawful\nfor a “broker . . . to induce or attempt to induce the purchase\nor sale of[] any security” without registering with the SEC.\n15 U.S.C. § 78o(a)(1). The statute defines “broker” as “any\n\n\n\n\n 5\n In a recent memorandum disposition, we rejected the argument that\nEB-5 investments do not constitute securities where the investors’\ninterest was to obtain visas rather than profits. SEC v. Liu, 754 F. App’x\n505, 507 (9th Cir. 2018). The facts in Liu mirror those in this case, with\na series of $500,000 EB-5 investments and accompanying $45,000\nadministrative fees. Id. The court held that the promise of an annual\nreturn of only 0.25 percent was enough to establish an expectation of\nprofit, “[e]ven if it was not [the investors’] primary motivation.” Id.\n\f18 USSEC V. FENG\n\nperson engaged in the business of effecting transactions in\nsecurities for the account of others.” Id. § 78c(a)(4)(A).\n\n The district court concluded that the uncontroverted\nevidence establishes that Feng was acting as a broker and\nviolated the registration requirement. We agree.\n\n 1. The Broker Determination\n\n In making its broker determination, the district court\nutilized the totality-of-the-circumstances approach that other\ncourts have used under Section 15(a)(1), relying on a set of\nfactors first set forth in SEC v. Hansen, No. 83 Civ. 3692,\n1984 WL 2413, at *10 (S.D.N.Y. Apr. 6, 1984). See SEC v.\nCollyard, 861 F.3d 760, 766 (8th Cir. 2017); SEC v. George,\n426 F.3d 786, 797 (6th Cir. 2005); see also SEC v. Imperiali,\nInc., 594 F. App’x 957, 961 (11th Cir. 2014) (per curiam)\n(quoting George, 426 F.3d at 797). Courts emphasize that\nthe so-called Hansen factors are “nonexclusive.” Collyard,\n861 F.3d at 766.\n\n In performing its inquiry, the district court considered\nwhether Feng:\n\n (1) is an employee of the issuer of the\n security;\n\n (2) received transaction-based income such\n as commissions rather than a salary;\n\n (3) sells or sold securities from other issuers;\n\n (4) was involved in negotiations between\n issuers and investors;\n\n (5) advertis[ed] for clients;\n\f USSEC V. FENG 19\n\n (6) gave advice or made valuations regarding\n the investment;\n\n (7) was an active finder of investors; and\n\n (8) regularly participates in securities\n transactions.\n\nSEC v. Feng, No. 15-09420, 2017 WL 6551107, at *7–8\n(C.D. Cal. Aug. 10, 2017) (footnote omitted). 6 The district\ncourt found that, although Feng was not an employee of the\nissuer, his conduct satisfied all the other factors. Id. at *8.\nHe received commissions from regional centers, sought\ninvestors and clients through his website and other online\nforums, negotiated with regional centers about the terms of\nthe projects, and gave advice about EB-5 projects’ likelihood\nof success.\n\n Although Feng does not contest the facts underlying\nthese conclusions, he objects to the factor-based analysis\ngenerally, arguing that the factors are so vague that they\nerroneously encompass individuals who are merely\nproviding legal advice in the EB-5 context. 7 To the contrary,\n\n 6\n The district court’s eight factors essentially tracked the\nconsiderations identified in Hansen. Feng, 2017 WL 6551107, at *7–8;\nHansen, 1984 WL 2413, at *10.\n 7\n Feng also argues that the district court’s factors omit the\nrequirement that a broker must have authority or control over the\naccounts of others. But the caselaw cited by Feng does not impose such\na requirement as a prerequisite for finding that someone is a broker. See\nSEC v. Kramer, 778 F. Supp. 2d 1320, 1339–40 (M.D. Fla. 2011)\n(discussing authority over an account as one of several factors); SEC v.\nM & A West, Inc., No. C-01-3376 VRW, 2005 WL 1514101, at *9 (N.D.\nCal. June 20, 2005) (same); cf. SEC v. Mapp, 240 F. Supp. 3d 569, 592\n\f20 USSEC V. FENG\n\nthe application of the Hansen factors reveals that much of\nthe work Feng performed for his clients, rather than being\ntraditional legal work, aligns with the indicia of broker\nactivity identified by those factors.\n\n For example, Feng asserts that the factor “received\ntransaction-based income such as commissions rather than a\nsalary” is unclear and could improperly include an attorney’s\ncontingent fee. However, Feng charged his clients a\nnoncontingent, upfront fee. And it is undisputed that Feng\nand his nominees received $1.268 million in commissions\nfrom the regional centers—not his legal clients—for\nrecruiting investors to the regional centers’ EB-5 projects.\nOn these facts, the “commissions” factor does not threaten\nto mislabel traditional legal services as broker activity.\n\n Similarly, Feng suggests that “involve[ment] in\nnegotiations between issuers and investors” should not\ninclude his legal work explaining to clients aspects of legal\ncompliance and regional center communications. Feng\nignores, however, the evidence that he communicated\nextensively with the regional centers, posing numerous\nquestions about the terms of the projects and negotiating the\namount of the administrative fees.\n\n In addition, Feng contends that the factor “gave advice\nor made valuations regarding the investment” should not\ninclude legal advice about the merits of the various EB-5\nprojects. He claims that he was advising his clients about\nhow to obtain legal permanent residency status, not about\n\n\n(E.D. Tex. 2017) (discussing but not deciding whether control over the\naccount of others is an element or a nondispositive factor). Further,\nassuming that control is a pertinent factor, it is present here in the\nmultiple instances in which client funds passed through Feng’s accounts.\n\f USSEC V. FENG 21\n\nhow to make money. The record, however, indicates that\nFeng’s services included investment advice. He reviewed\nand promoted certain EB-5 projects based on their likelihood\nof success under the EB-5 requirements, with particular\nattention to the projects’ likely ability to create the requisite\nnumber of jobs over time to qualify the investor for an EB-5\nvisa, as well as the likelihood the projects could return the\ninvestors’ capital contributions. This work goes beyond the\ntraditional work of a lawyer.\n\n Finally, Feng suggests that the factor “active finder of\ninvestors” improperly conflates seeking legal clients with\nseeking investors. This argument, too, is meritless. Feng\nsought a particular kind of client—one who would become\nan investor with a regional center with which he had a\nmarketing agreement and a reserved investor slot. Feng’s\n150 clients invested more than $65 million. In this EB-5\ncontext, his legal clients and the regional center investors\nwere, in fact, one and the same.\n\n 2. Feng’s Additional Arguments\n\n Feng goes beyond the Hansen analysis to argue more\ngenerally that the broker registration requirement should not\napply to his circumstances. First, he asserts that the\nrequirement should apply only to individuals who trade\nsecurities on an exchange and not to those involved in\ntransactions between private parties—deals commonly\nreferred to as “over-the-counter.” It is well established,\nhowever, that “[t]he 1934 [Exchange] Act was intended\nprincipally to protect investors against manipulation of stock\nprices through regulation of transactions upon securities\nexchanges and in over-the-counter markets.” Ernst & Ernst\nv. Hochfelder, 425 U.S. 185, 195 (1976) (emphasis added)\n(citing S. Rep. No. 792, at 1–5 (1934)). Although the\nExchange Act exempts brokers “whose business is\n\f22 USSEC V. FENG\n\nexclusively intrastate and who do[] not make use of any\nfacility of a national securities exchange,” Feng’s business\nwas not “exclusively intrastate.” See 15 U.S.C. § 78o(a)(1).\n\n Second, Feng argues that the broker registration\nrequirement is unnecessary in his circumstances because\nregistered-broker duties duplicate the fiduciary duties that\napply to the attorney-client relationship. Feng analogizes to\nthe registration exemption for lawyers under the Investment\nAdvisers Act of 1940, which excludes from the definition of\nan investment adviser “any lawyer, accountant, engineer, or\nteacher whose performance of such services is solely\nincidental to the practice of his profession.” 15 U.S.C.\n§ 80b-2(a)(11)(B). Feng asserts—without citation—that the\nInvestment Advisers Act exempts attorneys because of their\nheightened fiduciary duty to clients, and he urges us to apply\n“this rationale” to create an attorney exception to the broker\nregistration requirement in the Exchange Act.\n\n This argument fails for multiple reasons. As we have\nalready explained, even if such an exemption existed, the\nrecord belies any claim that Feng’s activities were “solely\nincidental to the practice of [law].” Moreover, if an\nattorney’s fiduciary duty were the rationale for the\nexemption under the Investment Advisers Act, the incidental\nnature of the investment-related work would seem irrelevant\nbecause the fiduciary duty presumably would cover any\nwork on behalf of clients. Feng essentially asks us to\nperform a legislative act by adding an exemption to one\nstatute on the basis of a different statute for which Congress\nexpressly provided an exemption. Self-evidently, we cannot\ndo that.\n\n In a last-ditch attempt to justify his failure to register,\nFeng argues that imposing the broker registration\nrequirement on U.S.-based attorneys would “force” them\n\f USSEC V. FENG 23\n\n“not to help these people.” This argument poses a false\nchoice. Attorneys who provide only legal advice to clients\nabout the EB-5 program are not required to register.\nAttorneys who act as brokers in the EB-5 context may\nlegally do so only if they register with the SEC.\n\n 3. Conclusion\n\n Given the undisputed facts in the record, we agree with\nthe district court that Feng, as a matter of law, was “engaged\nin the business of effecting transactions in securities for the\naccount of others” and thus was required to register with the\nSEC as a broker. 15 U.S.C. §§ 78c(4)(A), 78o(a)(1). 8\n\nB. The Fraud Determinations\n\n Feng argues that the district court erroneously found that\nhe committed securities fraud in violation of Section 17 of\nthe Securities Act and Section 10(b) of the Exchange Act, as\nimplemented through Rule 10b-5. The district court’s\ndeterminations were based on two theories of fraud liability:\nmaterial omissions and schemes to defraud. We review each\nin turn.\n\n\n\n\n 8\n Feng also appeals the denial of his motion for judgment on the\npleadings, arguing that the terms “security” and “broker” in Section\n15(a) of the Exchange Act are unconstitutionally vague. A word is not\nvague when it has a “settled legal meaning.” United States v. Williams,\n553 U.S. 285, 306 (2008). Section 15(a) was enacted 80 years ago, and\nit has been applied countless times by the courts. There was no error in\nthe district court’s denial of the motion for judgment on the pleadings.\n\f24 USSEC V. FENG\n\n 1. Material Omissions (Section 17(a)(2) and Rule\n 10b-5(b))\n\n The antifraud provisions of the Securities Act, Section\n17(a)(2), and the Exchange Act, Section 10(b), developed\nthrough Rule 10b-5(b), “forbid making a material\nmisstatement or omission in connection with the offer or sale\nof a security by means of interstate commerce.” SEC v. Dain\nRauscher, Inc., 254 F.3d 852, 856 (9th Cir. 2001). Under\nthese provisions, it is unlawful for any person, in the offer or\nsale of securities, “to obtain money or property by means of\nany untrue statement of a material fact or any omission to\nstate a material fact necessary in order to make the\nstatements made, in light of the circumstances under which\nthey were made, not misleading.” 15 U.S.C. § 77q(a)(2).\nThe text of Rule 10b-5 largely mirrors the text of Section\n17(a). 9 Although there are differences in the state of mind\nrequirements for Rule 10b-5(b) and Section 17(a)(2), a\nshowing of intentional or knowing conduct satisfies both.\nSee Aaron v. SEC, 446 U.S. 680, 695–97 (1980) (holding\nthat scienter is an element of Section 10(b), Rule 10b-5, and\nSection 17(a)(1), but not Section 17(a)(2) or (3)).\n\n A fiduciary duty to disclose certain information renders\nan omission of that information misleading. See United\nStates v. Laurienti, 611 F.3d 530, 541 (9th Cir. 2010). In\nLaurienti, the court addressed a broker’s liability under Rule\n10b-5(b) for failure to disclose commissions to his clients.\nThe court noted that, while the broker/client relationship\n 9\n Section 17(a) of the Securities Act prohibits fraud in the sale of\nsecurities. Rule 10b-5, adopted by the SEC pursuant to its authority\nunder Section 10(b) of the Exchange Act, incorporates the operative\nlanguage of Section 17(a) explicitly and also extends the prohibition to\nfraud committed in the purchase of securities. See Ernst & Ernst,\n425 U.S. at 212 n.32.\n\f USSEC V. FENG 25\n\ndoes not itself impose a fiduciary duty, circumstances may\ncreate “a relationship of trust and confidence.” Id. at 540.\nOften, a broker has a fiduciary duty when he has\ndiscretionary authority over a client’s account, “but we have\nrecognized that particular factual circumstances may serve\nto create a fiduciary duty between a broker and his customer\neven in the absence of a discretionary account.” Id. (quoting\nUnited States v. Skelly, 442 F.3d 94, 98 (2d Cir. 2006)). The\ncourt held that, however the duty arises, “if a broker and a\nclient have a trust relationship, . . . then the broker has an\nobligation to disclose all facts material to that relationship.”\nId. at 541.\n\n Here, as we have determined, Feng worked as both a\nbroker and an immigration attorney. Indeed, Feng\nemphasizes his role as an attorney and acknowledges that, as\nsuch, he had fiduciary duties to his clients. Those duties\nincluded the obligation to disclose conflicts of interest. See\nN.Y. Rules of Prof’l Conduct (22 NYCRR § 1200.0) 1.4\n(duty to communicate any information needed by the client\nto give informed consent), 1.7 (duty to disclose a conflict),\n1.8 (duty to disclose business relationships).\n\n Feng’s argument that there is no conflict because his\nclients were not harmed applies the wrong standard. The\nconflict of interest rule addresses the risk of harm to clients,\nasking whether a “reasonable lawyer would conclude . . .\nthere is a significant risk that the lawyer’s professional\njudgment on behalf of a client will be adversely affected by\nthe lawyer’s own financial, business, property or other\npersonal interests.” N.Y. Rules of Prof’l Conduct 1.7(a). In\nthese transactions, there was a risk that Feng’s judgment\nwould be swayed by the promise of commissions—that, for\nexample, he would prefer a regional center that paid a higher\ncommission but had a slightly lower likelihood of success\n\f26 USSEC V. FENG\n\nover a regional center that offered a lower commission but\nhad a better opportunity for the client.\n\n Indeed, when Feng was asked in his deposition whether\nthe commission from the regional center and his\nrepresentation of a client presented a conflict, he\nacknowledged that it did:\n\n Yes. I see it as sort of like a conflict because\n we’re in between, between regional center\n and the client. I’m aware of getting legal fees\n from the client, but also, even though we are\n not issuer’s counsel, we are definitely not\n regional center’s counsel. But regional\n centers pay us after the 526 getting approved.\n Even though we realize there’s such a\n conflict, but our interest is first and foremost\n with the clients.\n\nHis assertion that he shared his clients’ goal of obtaining a\ngreen card does not eliminate the risk of the commission’s\ninfluence on his professional judgment. Further, the rules\nprovide that only the client, fully informed of both the\nconflict and the risks stemming from the conflict, may make\na judgment about whether to accept representation\nnotwithstanding the conflict. N.Y. Rules of Prof’l Conduct\n1.4, 1.7.\n\n “[E]ven in a trust relationship,” the duty is “to disclose\nonly material facts.” See Laurienti, 611 F.3d at 541.\nGenerally, an omission is material “if there is ‘a substantial\nlikelihood that the disclosure of the omitted fact would have\nbeen viewed by the reasonable investor as having\nsignificantly altered the “total mix” of information made\navailable.’” SEC v. Phan, 500 F.3d 895, 908 (9th Cir. 2007)\n(quoting Basic Inc. v. Levinson, 485 U.S. 224, 231–32\n\f USSEC V. FENG 27\n\n(1988)). Here, the record includes statements of individual\ninvestors who said that if they had known about Feng’s\ncommissions, they would have attempted to negotiate their\nadministrative fees. Feng also testified that he concealed his\ncommissions to avoid anticipated client requests for rebates.\nHe said that he thought disclosure of his commissions would\nencourage the clients to ask for more reductions in their\nadministrative fees, and he did not want to open the door to\nnegotiations that would lower his clients’ costs but decrease\nhis payout. We thus agree with the district court that Feng’s\nfailure to disclose his conflict of interest was material. See\nTSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 450 (1976)\n(holding that materiality may be resolved at summary\njudgment “if the established omissions are so obviously\nimportant to an investor[] that reasonable minds cannot\ndiffer on the question of materiality” (internal quotation\nmarks omitted)).\n\n Finally, Feng’s testimony establishes intent to deceive.\nFeng stated that he did not want to tell clients about his\ncommissions from the regional centers because it would be\ncostly. Feng’s brief similarly describes his choice not to\ndisclose as a “business decision.”\n\n 2. Schemes to Defraud (Section 17(a)(1) and (3) and\n Rule 10b-5(a) and (c))\n\n The securities fraud provisions also prohibit any person,\nin the offer or sale of securities, “to employ any device,\nscheme, or artifice to defraud” or “to engage in any\ntransaction, practice, or course of business which operates or\nwould operate as a fraud or deceit upon the purchaser.”\n15 U.S.C. § 77q(a)(1), (3); see also 17 C.F.R. § 240.10b-\n5(a), (c). The state of mind requirement varies among these\nprovisions, but a showing of intentional or knowing conduct\nclears all thresholds. See Aaron, 446 U.S. at 695–97.\n\f28 USSEC V. FENG\n\n The district court found that Feng defrauded the regional\ncenters that refused to pay commissions to U.S.-based\nattorneys not registered as brokers. Feng evaded this\nrestriction by informing the regional centers that he was\naffiliated with foreign referral agents and companies,\nholding himself out to the regional centers as a middleman.\nFeng furthered this scheme by setting up ABCL in China to\nreceive commission payments, while maintaining control of\nboth ABCL’s bank account and operations. Feng performed\nthe recruitment services himself and received the\ncommissions that regional centers believed they were paying\nto the affiliated recruiters. Feng recruited his mother to sign\ndocuments as ABCL’s president, even though she had no\nactual role in the company. Representatives from the\nregional centers testified they would have ceased this\narrangement if they knew Feng’s relationship with either the\nagents or ABCL.\n\n The district court also found that Feng engaged in a\nscheme to defraud clients who sought a reduction in their\nadministrative fees. Generally, as noted, regional centers\nwere willing to rebate a portion of the administrative fee that\nthey charged investors if the regional center could also\nreduce the commission it paid for the recruitment of that\ninvestor. When a client asked Feng to negotiate with the\nregional center to reduce his administrative fee, Feng\narranged with the regional center to lower the administrative\nfee by reducing the commission. He took measures to\nconceal this arrangement from his clients, however, because\nhe wanted to “keep as much of the marketing fee as possible”\nand avoid further negotiation with clients, while also\ncreating the appearance to his clients that he was prevailing\non their behalf with the regional centers. Feng asked the\nregional centers to rebate a portion of the administrative fee\n\f USSEC V. FENG 29\n\nbut not to disclose that the funds were reallocated from a\ncommission.\n\n Feng challenges on two grounds the district court’s\nfinding that his conduct violated the Securities and Exchange\nActs. First, Feng asserts that the scheme-to-defraud\nallegations under Rule 10b-5(a) and (c) improperly duplicate\nthe Rule 10b-5(b) material omission claim. We see no\nduplication. The district court addressed the fraud against\nthe regional centers only as a scheme to defraud, not as a\nmaterial omission. With regard to the fraud against the\nclients, the district court focused on Feng’s failure to\ndisclose his conflict of interest to his clients as material\nomissions and, separately, on his efforts to conceal the\nsource of the rebates as a scheme to defraud.\n\n Second, Feng argues that his establishment of overseas\nentities and occasional arrangement of rebates for clients\nwere not unlawful. But, deceptive conduct that is not\n“inherently unlawful” may form the basis of a scheme to\ndefraud. SEC v. Wey, 246 F. Supp. 3d 894, 918 (S.D.N.Y.\n2017). In addition, Feng’s attempt to analogize the rebate\nscheme to a store proprietor giving a customer a discount is\ninapt. The fraudulent conduct was his scheme to make the\nrebate appear to come from his advocacy on behalf of the\nclient, while concealing the relationship between the\nreduction and his undisclosed commission.\n\n We accordingly affirm the district court’s finding that\nFeng engaged in securities fraud in violation of Section 17(a)\nand Section 10(b).\n\n IV.\n\n Finally, Feng challenges the district court’s\ndisgorgement order. The district court calculated that Feng\n\f30 USSEC V. FENG\n\nreceived $1.268 million for commissions in connection with\nthe EB-5 program and ordered disgorgement of the entire\namount. Feng contests the disgorgement of fees paid to his\noverseas entities, suggesting that some of the fees were used\nto cover expenses. He does not dispute that he controlled\nthose overseas entities.\n\n We review the court’s imposition of disgorgement for\nabuse of discretion. SEC v. JT Wallenbrock & Assocs.,\n440 F.3d 1109, 1113 (9th Cir. 2006). A disgorgement\ncalculation “should include ‘all gains flowing from the\nillegal activities,’” id. at 1114 (quoting SEC v. Cross Fin.\nServs., 908 F. Supp. 718, 734 (C.D. Cal. 1995), and “requires\nonly a ‘reasonable approximation of profits causally\nconnected to the violation,’” id. at 1113–14 (quoting SEC v.\nFirst Pac. Bancorp, 142 F.3d 1186, 1192 n. 6 (9th Cir.\n1998)).\n\n Feng used his overseas entities—ABCL and his other\nnominees—to advance his scheme to defraud the regional\ncenters. The money paid by regional centers to Feng’s\noverseas entities is therefore exactly the focus of the fraud\nagainst the regional centers, such that the district court could\nfind a “causal[] connect[ion]” through which the funds\n“flow[] from the illegal activities.” Id. at 1114. Feng should\nnot have been collecting these commissions in the first place,\nand it would be unjust to permit him to retain some of the ill-\ngotten funds to cover his expenses. See id. (“[I]t would be\nunjust to permit the defendants to offset against the investor\ndollars they received the expenses of running the very\nbusiness they created to defraud those investors into giving\nthe defendants the money in the first place.”). We\naccordingly find no abuse of discretion in the disgorgement\norder.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Ussec v. Hui Feng\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-23\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n U.S. SECURITIES & EXCHANGE No. 17-56522\\n COMMISSION,\\n Plaintiff-Appellee, D.C. No.\\n 2:15-cv-09420-\\n v. CBM-SS\\n\\n HUI FENG; LAW OFFICES OF FENG\\n AND ASSOCIATES PC, OPINION\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the Central District of California\\n Consuelo B. Marshall, District Judge, Presiding\\n\\n Argued and Submitted May 16, 2019\\n Pasadena, California\\n\\n Filed August 23, 2019\\n\\n Before: Kermit V. Lipez, * Kim McLane Wardlaw,\\n and Andrew D. Hurwitz, Circuit Judges.\\n\\n Opinion by Judge Lipez\\n\\n\\n\\n *\\n The Honorable Kermit V. Lipez, United States Circuit Judge for\\nthe First Circuit, sitting by designation.\\n\\f2 USSEC V. FENG\\n\\n SUMMARY **\\n\\n\\n Securities & Exchange Commission\\n\\n The panel affirmed the district court’s summary\\njudgment in favor of the U.S. Securities and Exchange\\nCommission (“SEC”) in its civil complaint filed against Hui\\nFeng and his law firm, alleging securities fraud.\\n\\n The U.S. Immigrant Investor Program, also known as the\\nEB-5 program, provides legal permanent residency in the\\nUnited States to foreign nationals who invest in U.S.-based\\nprojects. Multiple foreign investors may pool their money\\nin the same enterprise, and these pooled investments are\\nmade through “regional centers” which are regulated by the\\nU.S. Citizenship and Immigration Services.\\n\\n Feng legally represented clients through the EB-5\\nprocess, and entered into marketing agreements with\\nregional centers. The basis of the agreements between the\\nregional centers and Feng’s investors were known as private\\nplacement memoranda (“PPMs”).\\n\\n The panel agreed with the district court that the EB-5\\ninvestments in this case constituted “securities” in the form\\nof investment contracts. The panel rejected Feng’s argument\\nthat the transactions were not “securities” because his clients\\ndid not expect profits from their investments. Specifically,\\nthe panel held that the PPMs’ identification of the\\ninvestments as securities, the form of the investment entity\\n\\n **\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f USSEC V. FENG 3\\n\\nas a limited partnership, and the promise of a fixed rate of\\nreturn all indicated that the EB-5 transactions were\\nsecurities. The panel rejected Feng’s contention that the\\nadministrative fees upended the expectation of profits. The\\npanel also rejected Feng’s assertion that his clients lacked an\\nexpectation of profit because they were motivated to\\nparticipate in the EB-5 program by the promise of visas, not\\nby profit.\\n\\n Concerning the cause of action that Feng failed to\\nregister as a broker in violation of Section 15(a) of the\\nSecurities and Exchange Act of 1934, the panel agreed with\\nthe district court’s conclusion that Feng was acting as a\\nbroker and violated the registration requirement. The panel\\nheld that the district court properly made its broker\\ndetermination by utilizing the totality-of-the-circumstances\\napproach by relying on the so-called Hansen factors. The\\npanel rejected Feng’s arguments that the broker registration\\nrequirement should not apply to his circumstances.\\n\\n The panel affirmed the district court’s finding that Feng\\nengaged in securities fraud in violation of Section 17(a) of\\nthe Securities Act of 1933 and Section 10(b) of the Securities\\nExchange Act of 1934 based on two theories of fraud\\nliability: material omissions, and schemes to defraud.\\nConcerning material omissions, the panel held that where\\nFeng worked as both a broker and an immigration attorney,\\nhe had fiduciary duties to his clients, including the obligation\\nto disclose conflicts of interest. Here, there was a risk that\\nFeng’s judgment would be swayed by the promise of a\\ncommission from the regional center, and this presented a\\nconflict with Feng’s representation of a client, which he\\nfailed to disclose to his clients, and this failure was material.\\nConcerning schemes to defraud, the panel affirmed the\\ndistrict court’s findings that Feng defrauded the regional\\n\\f4 USSEC V. FENG\\n\\ncenters that refused to pay commissions to U.S.-based\\nattorneys not registered as brokers, and defrauded clients\\nwho sought a reduction in their administrative fees.\\n\\n Finally, the panel held that the district court did not abuse\\nits discretion in entering a disgorgement order. The district\\ncourt calculated that Feng received $1.268 million for\\ncommissions in connection with the EB-5 program, and\\nordered disgorgement of the entire amount.\\n\\n\\n COUNSEL\\n\\nAndrew B. Holmes (argued) and Matthew D. Taylor,\\nHolmes Taylor Scott & Jones LLP, Los Angeles, California,\\nfor Defendant-Appellant Law Offices of Feng & Associates.\\n\\nHui Feng (argued), Law Offices of Feng & Associates,\\nFlushing, New York, for Defendant-Appellant Hui Feng.\\n\\nKerry J. Dingle (argued), Senior Counsel; Jeffery A. Berger,\\nSenior Litigation Counsel; Michael A. Conley, Solicitor;\\nRobert B. Stebbins, General Counsel; Securities &\\nExchange Commission, Washington, D.C.; for Plaintiff-\\nAppellee.\\n\\f USSEC V. FENG 5\\n\\n OPINION\\n\\nLIPEZ, Circuit Judge:\\n\\n This appeal requires us to decide whether certain\\ninvestments made by participants in the U.S. Immigrant\\nInvestor Program are “securities” subject to regulation by the\\nSecurities and Exchange Commission (“SEC”) and, if so,\\nwhether appellant Hui Feng was a securities “broker”\\nrequired to register with the SEC and whether he committed\\nsecurities fraud in connection with the transactions. 1 The\\ndistrict court granted summary judgment for the SEC. We\\naffirm.\\n\\n I.\\n\\nA. The EB-5 Program\\n\\n The U.S. Immigrant Investor Program, colloquially\\nreferred to as the EB-5 program, provides legal permanent\\nresidency in the United States to foreign nationals who\\ninvest in U.S.-based projects. See 8 U.S.C.\\n§ 1153(b)(5)(A). 2 Generally, qualified immigrants may gain\\nU.S. visas through direct investment of at least $1 million in\\na new commercial enterprise that creates at least ten full-\\ntime jobs for U.S. workers. Id. § 1153(b)(5)(A), (C).\\nInvestment in a business in a “targeted employment area”\\n\\n 1\\n Both Hui Feng and his law firm, the Law Offices of Feng &\\nAssociates P.C., are appellants in this matter. The parties’ arguments do\\nnot distinguish between the two, and so, as a matter of convenience, we\\nrefer to the appellants collectively as “Feng.”\\n 2\\n By statute, the Immigrant Investor Program is the fifth preference\\nin the employment-based visa category, which gave rise to the nickname\\n“EB-5.” See 8 U.S.C. § 1153(b)(5)(A).\\n\\f6 USSEC V. FENG\\n\\nlowers the required capital investment amount to $500,000.\\nId. § 1153(b)(5)(C)(i), (ii).\\n\\n Multiple foreign investors may pool their money in the\\nsame enterprise, provided that each invests the required\\namount and “each individual investment results in the\\ncreation of at least ten full-time positions.” 8 C.F.R.\\n§ 204.6(g). Pooled investments are made through “regional\\ncenters,” which are regulated by the U.S. Citizenship and\\nImmigration Services (“USCIS”). The regional centers offer\\nspecific projects to investors and manage the pooled\\ninvestments. See id. § 204.6(e), (m).\\n\\n A foreign national investing in an enterprise must file an\\nI-526 application with the USCIS to prove that the\\ninvestment will satisfy EB-5 program requirements. Id.\\n§ 204.6(a), (j)(2). Approval results in conditional permanent\\nresident status. Two years later, the investor may remove\\nthe conditions on lawful permanent resident status by filing\\nan I-829 petition, demonstrating that the investment satisfied\\nthe EB-5 requirements and created, or will create within a\\nreasonable period, ten qualifying jobs. Id. § 216.6.\\n\\nB. Factual Background\\n\\n Feng and the SEC largely agree on the facts. To the\\nextent that they have differing views of the record, we draw\\nall justifiable inferences in the light most favorable to Feng.\\nBlankenhorn v. City of Orange, 485 F.3d 463, 470 (9th Cir.\\n2007).\\n\\n Feng conducts an immigration law practice in New York\\nCity. Between 2010 and 2016, he led approximately\\n150 clients through the EB-5 process, substantially all of\\nwhom were Chinese nationals who did not speak English.\\nFeng estimated that 80 percent of his clients found him\\n\\f USSEC V. FENG 7\\n\\nthrough his website, which posted comparisons and\\nrecommendations of available EB-5 projects. Feng holds a\\nlaw degree from Columbia University and an MBA from the\\nTuck School of Business at Dartmouth, and he touted his\\nskill at investigating and determining the most reliable\\ninvestment opportunities.\\n\\n Feng charged his clients a $10,000 to $15,000 upfront\\nfee. His services included serving as a liaison between\\nclients and regional centers, explaining the English-language\\noffering materials to his clients, negotiating with regional\\ncenters regarding administrative fees charged to the clients\\nby the centers, and compiling and submitting his clients’\\nsigned offering documents to regional centers. On three\\noccasions, investors wired funds directly to Feng, who\\ntransferred the funds to the regional centers.\\n\\n Feng also entered into marketing agreements with\\nregional centers. Through these agreements, he reserved a\\nnumber of the limited investor “spots” in certain EB-5\\nprojects, which he then promoted on his website and tried to\\nfill within a set time period. If one of Feng’s clients made\\nthe required capital contribution to one of these regional\\ncenters, and if the USCIS approved the investor’s I-526\\npetition, the regional center agreed to pay Feng commissions\\nof $15,000 to $70,000. Although hundreds of regional\\ncenters participate in the EB-5 program, Feng procured\\ninvestors only for ten centers, all of which agreed to pay\\ncommissions. Feng did not disclose to his clients the\\ncommissions he received from the regional centers unless\\nthey specifically asked about them. 3\\n\\n\\n 3\\n When the SEC’s investigation was pending, Feng revised his\\nretainer agreement to state that regional centers may pay fees to an\\n\\f8 USSEC V. FENG\\n\\n Payment of a commission is an EB-5 industry practice,\\nbut many regional centers refuse to pay a commission to a\\nU.S.-based attorney who is not registered as a broker, to\\navoid causing or aiding and abetting a violation of the broker\\nregistration requirement in Section 15(a) of the Securities\\nExchange Act of 1934. See 15 U.S.C. § 78o(a). Several\\nregional centers offer a discount or rebate of a portion of the\\nadministrative fee charged to the investor if there is no\\nreferral commission for the attorney. Feng chose neither to\\nregister as a broker, nor to forego commissions.\\n\\n When told by some regional centers that they could only\\npay commissions to overseas recruiters, Feng responded that\\nhe had referral partners in China who were finding investors.\\nHe entered into marketing agreements in which he promised\\nto relay the commissions to those individuals. He also\\narranged agreements that were directly between the regional\\ncenters and the individuals. The referral partners included\\nFeng’s wife, mother, and mother-in-law. Feng described\\nthese individuals in his deposition testimony as his\\n“nominees,” whose role was to be “just a surrogate to receive\\nthe money.” They did not do the work of recruiting\\ninvestors—Feng did that himself. After the regional centers\\npaid commissions to the individuals, they transferred the\\nfunds to Feng. Feng asserts that the regional centers only\\nasked for a foreign contact to pay and did not “care” who it\\nwas.\\n\\n In 2014, Feng also created Atlantic Business Consulting\\nLimited (“ABCL”), based in Hong Kong. According to\\nFeng, this decision was “driven in part by regional centers\\ninforming me they needed an overseas entity to pay.” ABCL\\n\\noverseas company owned by Feng upon completion of the client’s visa\\napproval.\\n\\f USSEC V. FENG 9\\n\\nentered into marketing agreements with regional centers, and\\nits only source of revenue was EB-5 commissions. Feng did\\nnot inform the regional centers that he solicited and referred\\nABCL’s investors himself. All of Feng’s employees worked\\nboth for his law office and for ABCL. Feng’s mother,\\nXiuyuan Tan, signed agreements between ABCL and\\nregional centers as ABCL’s “President,” but Feng admitted\\nshe played no role in ABCL. Feng also did not inform the\\nregional centers that he was ABCL’s beneficial owner, with\\nsole control of ABCL’s bank account. Representatives from\\nthe regional centers testified that they would have ceased\\ntheir marketing agreements if they had known about Feng’s\\nrelationship with the individual referral agents and with\\nABCL.\\n\\n In total, regional centers paid Feng and his overseas\\nentities $1.268 million in commissions for investments made\\nby Feng’s clients. At the time of the district court judgment,\\nthese entities were contractually entitled to an additional\\n$3.45 million in commissions, pending the approval of his\\nclients’ I-526 petitions.\\n\\n The basis of the agreements between the regional centers\\nand Feng’s investors was set forth in the regional centers’\\noffering materials, also known as private placement\\nmemoranda (“PPMs”). Many of the PPMs referred to the\\ninvestments as “securities” and asserted that the investments\\nwere compliant with U.S. securities laws. The regional\\ncenters structured the investments as limited partnerships, in\\nwhich the investors became limited partners and the regional\\ncenter was the general partner. The regional centers\\npromised investors a fixed, annual return on investment,\\nwhich ranged across projects from 0.5 to 5 percent of the\\ncapital contribution, and investors received Schedule K-1 tax\\nforms to report their investment income from the\\n\\f10 USSEC V. FENG\\n\\npartnership. The partnerships all made construction loans or\\notherwise financed a specified construction project. At the\\nend of the investment term, typically five to six years, the\\nregional centers promised the investors a return of their\\ncapital contribution, subject to market risks.\\n\\n The PPMs required that investors pay an administrative\\nfee, which ranged across projects from $30,000 to $50,000,\\nin addition to the capital contribution of $1 million or\\n$500,000. The PPMs expressly stated that these\\nadministrative fees were for operating and marketing costs,\\nwere not part of the capital contribution, and did not earn\\ninterest.\\n\\n Approximately 20 percent of Feng’s clients asked him to\\nseek a reduced administrative fee from the regional centers.\\nIn those instances, Feng negotiated with regional centers and\\nfacilitated contracts between those regional centers and his\\nclients for a rebate of a portion of the administrative fee.\\nFeng did not disclose to these clients, however, that the\\nadministrative fees helped to fund his commissions, nor that\\nthe regional centers offset the clients’ administrative fee\\nrebate with a reduction in the commissions to which he was\\ncontractually entitled. Indeed, Feng asked the regional\\ncenters not to inform the clients about either fact. Feng\\nexplained in his deposition that he wanted to create the\\nappearance to his clients that he was advocating on their\\nbehalf with the regional center and also to “keep as much of\\nthe marketing fee as possible”—that is, his commission—\\nand to avoid further “haggl[ing]” with clients who might ask\\nFeng to rebate more of the commission. 4\\n\\n\\n 4\\n For example, one client testified that Feng sent him a check for\\n$15,000, which Feng explained as a return of charges from the regional\\n\\f USSEC V. FENG 11\\n\\nC. Procedural Background\\n\\n The SEC filed a civil complaint against Feng and his law\\nfirm on December 7, 2015. The first and second causes of\\naction allege fraud under Section 17(a) of the Securities Act\\nof 1933, 15 U.S.C. § 77q(a), and Section 10(b) of the\\nSecurities Exchange Act of 1934, 15 U.S.C. § 78j(b) (“the\\nExchange Act”), respectively. The third cause of action\\nalleges failure to register as a broker-dealer under Section\\n15(a) of the Exchange Act, 15 U.S.C. § 78o.\\n\\n The defendants filed a motion for judgment on the\\npleadings, which the district court denied in August 2016.\\nThe parties subsequently filed cross-motions for summary\\njudgment. Feng argued that the EB-5 investments were not\\n“securities” because the investors had no expectation of\\nprofit—only of obtaining a green card. The SEC argued that\\nthe undisputed evidence showed that Feng acted as a\\n“broker” of “securities” as a matter of law. The district court\\nfound no genuine dispute of material fact and granted\\nsummary judgment for the SEC on all three causes of action.\\nIn this appeal, Feng argues that the transactions were not\\n“securities,” and so the district court erred by denying his\\nmotion for summary judgment and granting the SEC’s\\nmotion.\\n\\n We review the district court’s decision on cross-motions\\nfor summary judgment de novo. Avery v. First Resolution\\nMgmt. Corp., 568 F.3d 1018, 1021 (9th Cir. 2009).\\nSummary judgment is appropriate when there is no genuine\\ndispute of material fact and the moving party is entitled to\\njudgment as a matter of law. Fed. R. Civ. P. 56(a). We\\n\\ncenter without disclosing that Feng had received a $70,000 commission\\non the client’s $1 million investment.\\n\\f12 USSEC V. FENG\\n\\nconsider cross-motions “separately, giving the nonmoving\\nparty in each instance the benefit of all reasonable\\ninferences.” Zabriskie v. Fed. Nat’l Mortg. Ass’n, 912 F.3d\\n1192, 1196 (9th Cir. 2019) (quoting ACLU of Nev. v. City of\\nLas Vegas, 466 F.3d 784, 790–91 (9th Cir. 2006)).\\n\\n II.\\n\\n Feng contends that the district court erred in its threshold\\ndetermination that the EB-5 investments at issue were\\n“securities.” Because this finding is essential to all the\\ncauses of action, we begin by considering whether the\\ndistrict court properly made that determination as a matter of\\nlaw.\\n\\n The district court found that the EB-5 investments were\\nsecurities because they were made through investment\\ncontracts between the regional centers and the investors.\\nSection 3(a)(10) of the Securities Act defines the term\\n“security” broadly, enumerating a long list of financial\\ninstruments, including “any . . . investment contract.”\\n15 U.S.C. § 78c(a)(10). Although the Securities Act does\\nnot define the term “investment contract,” the parties agree\\nthat the applicable test, set forth in SEC v. W. J. Howey Co.,\\nrequires “a contract, transaction or scheme whereby a person\\ninvests his money in a common enterprise and is led to\\nexpect profits solely from the efforts of the promoter or a\\nthird party.” 328 U.S. 293, 298–99 (1946). Courts applying\\nHowey “conduct an objective inquiry into the character of\\nthe instrument or transaction offered based on what the\\npurchasers were ‘led to expect,’” including an analysis of the\\npromotional materials associated with the transaction.\\nWarfield v. Alaniz, 569 F.3d 1015, 1021 (9th Cir. 2009)\\n(quoting Howey, 328 U.S. at 298–99).\\n\\f USSEC V. FENG 13\\n\\n It is undisputed that Feng’s clients invested money in a\\ncommon enterprise managed by a third party. Feng argues\\nthat the transactions nonetheless were not “securities”\\nbecause the clients did not expect profits from their\\ninvestments. He asserts that his clients would have\\nrecognized that the modest fixed return they were due from\\nthe regional centers would not exceed the significant\\nadministrative fees that they paid. He further maintains that\\nthe lack of expected profit was not a concern for his clients\\nbecause their motivation for making an EB-5 investment was\\nto obtain U.S. visas. We find Feng’s characterization of the\\nEB-5 investments unpersuasive.\\n\\nA. The PPMs\\n\\n The regional centers promoted and described the EB-5\\ninvestments in the PPMs. The PPMs repeatedly referred to\\nthe investments as “securities” and explained that the\\nofferings were made pursuant to U.S. securities laws.\\nAlthough “the name given to an instrument is not\\ndispositive,” “most instruments bearing th[e] traditional\\ntitles [associated with securities] are likely to be covered by\\nthe statutes.” United Hous. Found., Inc. v. Forman, 421 U.S.\\n837, 850 (1975). Further, “the use of a traditional name such\\nas ‘stocks’ or ‘bonds’ [may] lead a purchaser justifiably to\\nassume that the federal securities laws apply.” Id. That\\ninference inevitably would be drawn “when the underlying\\ntransaction embodies some of the significant characteristics\\ntypically associated with the named instrument.” Id. at 851.\\n\\n Such characteristics are evident and undisputed here.\\nThe EB-5 transactions were structured as investments in\\nlimited partnerships, with the promise of a fixed annual\\nreturn on the investment, ranging from .5 to 5 percent.\\nInvestments in limited partnerships generally constitute\\ninvestment contracts, see SEC v. Murphy, 626 F.2d 633, 640\\n\\f14 USSEC V. FENG\\n\\n(9th Cir. 1980), and a return based on a fixed rate falls within\\nthe broad definition of “profits,” SEC v. Edwards, 540 U.S.\\n389, 394 (2004); see also id. at 396 (“[T]he commonsense\\nunderstanding of ‘profits’ in the Howey test [is] simply\\n‘financial returns on . . . investments.’” (quoting Forman,\\n421 U.S. at 853)); Warfield, 569 F.3d at 1022 (holding that\\nthe investors had an expectation of profit where the\\npromotional materials emphasized “long-term income\\nproduction potential”). The Supreme Court has reasoned\\nthat the promise of a fixed rate of return should be viewed as\\ntriggering an expectation of profits because the securities\\nlaws are intended “to regulate all of the ‘countless and\\nvariable schemes devised by those who seek the use of the\\nmoney of others on the promise of profits.’” Edwards,\\n540 U.S. at 396 (quoting Howey, 328 U.S. at 299). The\\nCourt was wary that “unscrupulous marketers of\\ninvestments” might otherwise “evade the securities laws by\\npicking a rate of return to promise.” Id. at 394–95.\\n\\n The PPMs’ identification of the investments as\\nsecurities, the form of the investment entity as a limited\\npartnership, and the promise of a fixed rate of return all\\nindicate that the EB-5 transactions were securities.\\n\\nB. The Administrative Fees\\n\\n Feng insists, however, that the promised return is\\neffectively nullified by the administrative fees. When an\\ninvestor’s administrative fee and capital contribution are\\nconsidered together, Feng contends, the investor could not\\nhave expected to make any profit. The record, however,\\ndoes not support treating the administrative fee as part of the\\ninvestment. The PPMs expressly distinguish those two\\ncategories of funds, describing the investment as the “capital\\ncontribution,” which would be “put at risk,” or utilized by,\\nthe regional center. By contrast, the PPMs explain that the\\n\\f USSEC V. FENG 15\\n\\nadministrative fee would be used to defray marketing and\\noperating expenses.\\n\\n Importantly, the separation between the capital\\ncontribution and the administrative fee is necessary to satisfy\\nthe EB-5 regulatory requirements. The USCIS requires that\\na qualifying EB-5 investment place capital “at risk for the\\npurpose of generating a return on the capital placed at risk.”\\n8 C.F.R. § 204.6(j)(2). In other words, to be eligible for the\\nEB-5 program, the offering’s terms must include a possible\\nreturn on the capital contribution. Hence, even if Feng is\\ncorrect that, as a mathematical proposition, combining the\\ncapital contribution and administrative fee would eliminate\\na return on the investment, the PPMs make clear that the two\\ncategories of funds are to be viewed separately.\\n\\n Feng himself confirmed these separate identities when\\nthe USCIS questioned whether one of his clients met the EB-\\n5 requirements. In a letter to the USCIS, Feng stated that the\\nclient’s capital contribution was separate from the\\nadministrative fee and was therefore put “at risk for the\\npurpose of generating a return.” The client, Feng asserted,\\n“made an at-risk capital investment of $500,000 . . . . No\\nportion of an investor’s contribution has been or will be\\napplied towards legal service fees or administrative fee[s],\\nmarketing costs or interest payment of [the regional center],\\nwhich are paid from the separate $50,000 administrative\\nfee.”\\n\\n Feng thus attempts to characterize the administrative\\nfees in two different ways to suit his own purposes. On the\\none hand, he explained that the administrative fee should be\\nexcluded from the “investment” to establish that the\\ninvestment satisfies the EB-5 requirement of a possible\\nreturn. On the other hand, he seeks to include the\\nadministrative fee in the securities analysis to show there is\\n\\f16 USSEC V. FENG\\n\\nno possibility of generating the return as required by the\\nHowey test. That inconsistent approach is plainly contrary\\nto Congress’s intent in enacting the securities laws, as it\\nwould empower “unscrupulous marketers of investments” to\\nstructure deals to evade the designation of a security.\\nEdwards, 540 U.S. at 394. Feng’s argument that the\\nadministrative fee upends the expectation of profits has no\\nmerit.\\n\\nC. Motivation\\n\\n We likewise find unsupportable Feng’s assertion that—\\nwhatever the form of the investment and the rate of return—\\nhis clients nonetheless lacked an expectation of profit\\nbecause they were motivated to participate in the EB-5\\nprogram by the promise of visas, not by profit. See Warfield,\\n569 F.3d at 1021 (holding that the focus of Howey is the\\n“objective inquiry” into “what the purchasers were offered\\nor promised,” although the investors’ subjective intent “may\\nhave some bearing on the issue of whether they entered into\\ninvestment contracts”). We do not doubt that the investors’\\nprimary reason to participate in the EB-5 program was to\\ngain U.S. visas. But Feng oversimplifies his clients’\\nmotivations.\\n\\n An EB-5 investor’s interest in a visa is inextricably tied\\nto the financial success of the regional center’s project. The\\nEB-5 program by design links the success of the investment\\nto the investor’s success in obtaining a visa. An EB-5\\napplicant must prove to the USCIS that an investment in fact\\nled to the creation of at least ten full-time jobs. 8 U.S.C.\\n§ 1153(b)(5)(A); 8 C.F.R. §§ 204.6, 216.6. Understanding\\nthe importance of a successful investment, Feng marketed\\nhis ability to evaluate the multitude of regional centers and\\nrecommend projects that were most likely to be sustained\\nlong enough to create the requisite jobs to qualify the\\n\\f USSEC V. FENG 17\\n\\ninvestor for the EB-5 visa. The record also contains\\ntestimony from some of Feng’s client-investors who said\\nthat the likelihood of the regional center returning their\\ncapital contribution was important to them or that they\\nsought EB-5 projects with higher rates of return. Feng’s\\nargument thus relies on a false dichotomy between the visa\\nand the success of the investment. 5\\n\\n We therefore agree with the district court that the\\ninvestments in this case constitute securities in the form of\\ninvestment contracts.\\n\\n III.\\n\\n Having resolved the threshold securities issue, we next\\nconsider Feng’s challenges to the grant of summary\\njudgment for the SEC on each of the three causes of action\\nbrought against him.\\n\\nA. Failure to Register as a Broker\\n\\n Section 15(a)(1) of the Exchange Act makes it unlawful\\nfor a “broker . . . to induce or attempt to induce the purchase\\nor sale of[] any security” without registering with the SEC.\\n15 U.S.C. § 78o(a)(1). The statute defines “broker” as “any\\n\\n\\n\\n\\n 5\\n In a recent memorandum disposition, we rejected the argument that\\nEB-5 investments do not constitute securities where the investors’\\ninterest was to obtain visas rather than profits. SEC v. Liu, 754 F. App’x\\n505, 507 (9th Cir. 2018). The facts in Liu mirror those in this case, with\\na series of $500,000 EB-5 investments and accompanying $45,000\\nadministrative fees. Id. The court held that the promise of an annual\\nreturn of only 0.25 percent was enough to establish an expectation of\\nprofit, “[e]ven if it was not [the investors’] primary motivation.” Id.\\n\\f18 USSEC V. FENG\\n\\nperson engaged in the business of effecting transactions in\\nsecurities for the account of others.” Id. § 78c(a)(4)(A).\\n\\n The district court concluded that the uncontroverted\\nevidence establishes that Feng was acting as a broker and\\nviolated the registration requirement. We agree.\\n\\n 1. The Broker Determination\\n\\n In making its broker determination, the district court\\nutilized the totality-of-the-circumstances approach that other\\ncourts have used under Section 15(a)(1), relying on a set of\\nfactors first set forth in SEC v. Hansen, No. 83 Civ. 3692,\\n1984 WL 2413, at *10 (S.D.N.Y. Apr. 6, 1984). See SEC v.\\nCollyard, 861 F.3d 760, 766 (8th Cir. 2017); SEC v. George,\\n426 F.3d 786, 797 (6th Cir. 2005); see also SEC v. Imperiali,\\nInc., 594 F. App’x 957, 961 (11th Cir. 2014) (per curiam)\\n(quoting George, 426 F.3d at 797). Courts emphasize that\\nthe so-called Hansen factors are “nonexclusive.” Collyard,\\n861 F.3d at 766.\\n\\n In performing its inquiry, the district court considered\\nwhether Feng:\\n\\n (1) is an employee of the issuer of the\\n security;\\n\\n (2) received transaction-based income such\\n as commissions rather than a salary;\\n\\n (3) sells or sold securities from other issuers;\\n\\n (4) was involved in negotiations between\\n issuers and investors;\\n\\n (5) advertis[ed] for clients;\\n\\f USSEC V. FENG 19\\n\\n (6) gave advice or made valuations regarding\\n the investment;\\n\\n (7) was an active finder of investors; and\\n\\n (8) regularly participates in securities\\n transactions.\\n\\nSEC v. Feng, No. 15-09420, 2017 WL 6551107, at *7–8\\n(C.D. Cal. Aug. 10, 2017) (footnote omitted). 6 The district\\ncourt found that, although Feng was not an employee of the\\nissuer, his conduct satisfied all the other factors. Id. at *8.\\nHe received commissions from regional centers, sought\\ninvestors and clients through his website and other online\\nforums, negotiated with regional centers about the terms of\\nthe projects, and gave advice about EB-5 projects’ likelihood\\nof success.\\n\\n Although Feng does not contest the facts underlying\\nthese conclusions, he objects to the factor-based analysis\\ngenerally, arguing that the factors are so vague that they\\nerroneously encompass individuals who are merely\\nproviding legal advice in the EB-5 context. 7 To the contrary,\\n\\n 6\\n The district court’s eight factors essentially tracked the\\nconsiderations identified in Hansen. Feng, 2017 WL 6551107, at *7–8;\\nHansen, 1984 WL 2413, at *10.\\n 7\\n Feng also argues that the district court’s factors omit the\\nrequirement that a broker must have authority or control over the\\naccounts of others. But the caselaw cited by Feng does not impose such\\na requirement as a prerequisite for finding that someone is a broker. See\\nSEC v. Kramer, 778 F. Supp. 2d 1320, 1339–40 (M.D. Fla. 2011)\\n(discussing authority over an account as one of several factors); SEC v.\\nM & A West, Inc., No. C-01-3376 VRW, 2005 WL 1514101, at *9 (N.D.\\nCal. June 20, 2005) (same); cf. SEC v. Mapp, 240 F. Supp. 3d 569, 592\\n\\f20 USSEC V. FENG\\n\\nthe application of the Hansen factors reveals that much of\\nthe work Feng performed for his clients, rather than being\\ntraditional legal work, aligns with the indicia of broker\\nactivity identified by those factors.\\n\\n For example, Feng asserts that the factor “received\\ntransaction-based income such as commissions rather than a\\nsalary” is unclear and could improperly include an attorney’s\\ncontingent fee. However, Feng charged his clients a\\nnoncontingent, upfront fee. And it is undisputed that Feng\\nand his nominees received $1.268 million in commissions\\nfrom the regional centers—not his legal clients—for\\nrecruiting investors to the regional centers’ EB-5 projects.\\nOn these facts, the “commissions” factor does not threaten\\nto mislabel traditional legal services as broker activity.\\n\\n Similarly, Feng suggests that “involve[ment] in\\nnegotiations between issuers and investors” should not\\ninclude his legal work explaining to clients aspects of legal\\ncompliance and regional center communications. Feng\\nignores, however, the evidence that he communicated\\nextensively with the regional centers, posing numerous\\nquestions about the terms of the projects and negotiating the\\namount of the administrative fees.\\n\\n In addition, Feng contends that the factor “gave advice\\nor made valuations regarding the investment” should not\\ninclude legal advice about the merits of the various EB-5\\nprojects. He claims that he was advising his clients about\\nhow to obtain legal permanent residency status, not about\\n\\n\\n(E.D. Tex. 2017) (discussing but not deciding whether control over the\\naccount of others is an element or a nondispositive factor). Further,\\nassuming that control is a pertinent factor, it is present here in the\\nmultiple instances in which client funds passed through Feng’s accounts.\\n\\f USSEC V. FENG 21\\n\\nhow to make money. The record, however, indicates that\\nFeng’s services included investment advice. He reviewed\\nand promoted certain EB-5 projects based on their likelihood\\nof success under the EB-5 requirements, with particular\\nattention to the projects’ likely ability to create the requisite\\nnumber of jobs over time to qualify the investor for an EB-5\\nvisa, as well as the likelihood the projects could return the\\ninvestors’ capital contributions. This work goes beyond the\\ntraditional work of a lawyer.\\n\\n Finally, Feng suggests that the factor “active finder of\\ninvestors” improperly conflates seeking legal clients with\\nseeking investors. This argument, too, is meritless. Feng\\nsought a particular kind of client—one who would become\\nan investor with a regional center with which he had a\\nmarketing agreement and a reserved investor slot. Feng’s\\n150 clients invested more than $65 million. In this EB-5\\ncontext, his legal clients and the regional center investors\\nwere, in fact, one and the same.\\n\\n 2. Feng’s Additional Arguments\\n\\n Feng goes beyond the Hansen analysis to argue more\\ngenerally that the broker registration requirement should not\\napply to his circumstances. First, he asserts that the\\nrequirement should apply only to individuals who trade\\nsecurities on an exchange and not to those involved in\\ntransactions between private parties—deals commonly\\nreferred to as “over-the-counter.” It is well established,\\nhowever, that “[t]he 1934 [Exchange] Act was intended\\nprincipally to protect investors against manipulation of stock\\nprices through regulation of transactions upon securities\\nexchanges and in over-the-counter markets.” Ernst & Ernst\\nv. Hochfelder, 425 U.S. 185, 195 (1976) (emphasis added)\\n(citing S. Rep. No. 792, at 1–5 (1934)). Although the\\nExchange Act exempts brokers “whose business is\\n\\f22 USSEC V. FENG\\n\\nexclusively intrastate and who do[] not make use of any\\nfacility of a national securities exchange,” Feng’s business\\nwas not “exclusively intrastate.” See 15 U.S.C. § 78o(a)(1).\\n\\n Second, Feng argues that the broker registration\\nrequirement is unnecessary in his circumstances because\\nregistered-broker duties duplicate the fiduciary duties that\\napply to the attorney-client relationship. Feng analogizes to\\nthe registration exemption for lawyers under the Investment\\nAdvisers Act of 1940, which excludes from the definition of\\nan investment adviser “any lawyer, accountant, engineer, or\\nteacher whose performance of such services is solely\\nincidental to the practice of his profession.” 15 U.S.C.\\n§ 80b-2(a)(11)(B). Feng asserts—without citation—that the\\nInvestment Advisers Act exempts attorneys because of their\\nheightened fiduciary duty to clients, and he urges us to apply\\n“this rationale” to create an attorney exception to the broker\\nregistration requirement in the Exchange Act.\\n\\n This argument fails for multiple reasons. As we have\\nalready explained, even if such an exemption existed, the\\nrecord belies any claim that Feng’s activities were “solely\\nincidental to the practice of [law].” Moreover, if an\\nattorney’s fiduciary duty were the rationale for the\\nexemption under the Investment Advisers Act, the incidental\\nnature of the investment-related work would seem irrelevant\\nbecause the fiduciary duty presumably would cover any\\nwork on behalf of clients. Feng essentially asks us to\\nperform a legislative act by adding an exemption to one\\nstatute on the basis of a different statute for which Congress\\nexpressly provided an exemption. Self-evidently, we cannot\\ndo that.\\n\\n In a last-ditch attempt to justify his failure to register,\\nFeng argues that imposing the broker registration\\nrequirement on U.S.-based attorneys would “force” them\\n\\f USSEC V. FENG 23\\n\\n“not to help these people.” This argument poses a false\\nchoice. Attorneys who provide only legal advice to clients\\nabout the EB-5 program are not required to register.\\nAttorneys who act as brokers in the EB-5 context may\\nlegally do so only if they register with the SEC.\\n\\n 3. Conclusion\\n\\n Given the undisputed facts in the record, we agree with\\nthe district court that Feng, as a matter of law, was “engaged\\nin the business of effecting transactions in securities for the\\naccount of others” and thus was required to register with the\\nSEC as a broker. 15 U.S.C. §§ 78c(4)(A), 78o(a)(1). 8\\n\\nB. The Fraud Determinations\\n\\n Feng argues that the district court erroneously found that\\nhe committed securities fraud in violation of Section 17 of\\nthe Securities Act and Section 10(b) of the Exchange Act, as\\nimplemented through Rule 10b-5. The district court’s\\ndeterminations were based on two theories of fraud liability:\\nmaterial omissions and schemes to defraud. We review each\\nin turn.\\n\\n\\n\\n\\n 8\\n Feng also appeals the denial of his motion for judgment on the\\npleadings, arguing that the terms “security” and “broker” in Section\\n15(a) of the Exchange Act are unconstitutionally vague. A word is not\\nvague when it has a “settled legal meaning.” United States v. Williams,\\n553 U.S. 285, 306 (2008). Section 15(a) was enacted 80 years ago, and\\nit has been applied countless times by the courts. There was no error in\\nthe district court’s denial of the motion for judgment on the pleadings.\\n\\f24 USSEC V. FENG\\n\\n 1. Material Omissions (Section 17(a)(2) and Rule\\n 10b-5(b))\\n\\n The antifraud provisions of the Securities Act, Section\\n17(a)(2), and the Exchange Act, Section 10(b), developed\\nthrough Rule 10b-5(b), “forbid making a material\\nmisstatement or omission in connection with the offer or sale\\nof a security by means of interstate commerce.” SEC v. Dain\\nRauscher, Inc., 254 F.3d 852, 856 (9th Cir. 2001). Under\\nthese provisions, it is unlawful for any person, in the offer or\\nsale of securities, “to obtain money or property by means of\\nany untrue statement of a material fact or any omission to\\nstate a material fact necessary in order to make the\\nstatements made, in light of the circumstances under which\\nthey were made, not misleading.” 15 U.S.C. § 77q(a)(2).\\nThe text of Rule 10b-5 largely mirrors the text of Section\\n17(a). 9 Although there are differences in the state of mind\\nrequirements for Rule 10b-5(b) and Section 17(a)(2), a\\nshowing of intentional or knowing conduct satisfies both.\\nSee Aaron v. SEC, 446 U.S. 680, 695–97 (1980) (holding\\nthat scienter is an element of Section 10(b), Rule 10b-5, and\\nSection 17(a)(1), but not Section 17(a)(2) or (3)).\\n\\n A fiduciary duty to disclose certain information renders\\nan omission of that information misleading. See United\\nStates v. Laurienti, 611 F.3d 530, 541 (9th Cir. 2010). In\\nLaurienti, the court addressed a broker’s liability under Rule\\n10b-5(b) for failure to disclose commissions to his clients.\\nThe court noted that, while the broker/client relationship\\n 9\\n Section 17(a) of the Securities Act prohibits fraud in the sale of\\nsecurities. Rule 10b-5, adopted by the SEC pursuant to its authority\\nunder Section 10(b) of the Exchange Act, incorporates the operative\\nlanguage of Section 17(a) explicitly and also extends the prohibition to\\nfraud committed in the purchase of securities. See Ernst & Ernst,\\n425 U.S. at 212 n.32.\\n\\f USSEC V. FENG 25\\n\\ndoes not itself impose a fiduciary duty, circumstances may\\ncreate “a relationship of trust and confidence.” Id. at 540.\\nOften, a broker has a fiduciary duty when he has\\ndiscretionary authority over a client’s account, “but we have\\nrecognized that particular factual circumstances may serve\\nto create a fiduciary duty between a broker and his customer\\neven in the absence of a discretionary account.” Id. (quoting\\nUnited States v. Skelly, 442 F.3d 94, 98 (2d Cir. 2006)). The\\ncourt held that, however the duty arises, “if a broker and a\\nclient have a trust relationship, . . . then the broker has an\\nobligation to disclose all facts material to that relationship.”\\nId. at 541.\\n\\n Here, as we have determined, Feng worked as both a\\nbroker and an immigration attorney. Indeed, Feng\\nemphasizes his role as an attorney and acknowledges that, as\\nsuch, he had fiduciary duties to his clients. Those duties\\nincluded the obligation to disclose conflicts of interest. See\\nN.Y. Rules of Prof’l Conduct (22 NYCRR § 1200.0) 1.4\\n(duty to communicate any information needed by the client\\nto give informed consent), 1.7 (duty to disclose a conflict),\\n1.8 (duty to disclose business relationships).\\n\\n Feng’s argument that there is no conflict because his\\nclients were not harmed applies the wrong standard. The\\nconflict of interest rule addresses the risk of harm to clients,\\nasking whether a “reasonable lawyer would conclude . . .\\nthere is a significant risk that the lawyer’s professional\\njudgment on behalf of a client will be adversely affected by\\nthe lawyer’s own financial, business, property or other\\npersonal interests.” N.Y. Rules of Prof’l Conduct 1.7(a). In\\nthese transactions, there was a risk that Feng’s judgment\\nwould be swayed by the promise of commissions—that, for\\nexample, he would prefer a regional center that paid a higher\\ncommission but had a slightly lower likelihood of success\\n\\f26 USSEC V. FENG\\n\\nover a regional center that offered a lower commission but\\nhad a better opportunity for the client.\\n\\n Indeed, when Feng was asked in his deposition whether\\nthe commission from the regional center and his\\nrepresentation of a client presented a conflict, he\\nacknowledged that it did:\\n\\n Yes. I see it as sort of like a conflict because\\n we’re in between, between regional center\\n and the client. I’m aware of getting legal fees\\n from the client, but also, even though we are\\n not issuer’s counsel, we are definitely not\\n regional center’s counsel. But regional\\n centers pay us after the 526 getting approved.\\n Even though we realize there’s such a\\n conflict, but our interest is first and foremost\\n with the clients.\\n\\nHis assertion that he shared his clients’ goal of obtaining a\\ngreen card does not eliminate the risk of the commission’s\\ninfluence on his professional judgment. Further, the rules\\nprovide that only the client, fully informed of both the\\nconflict and the risks stemming from the conflict, may make\\na judgment about whether to accept representation\\nnotwithstanding the conflict. N.Y. Rules of Prof’l Conduct\\n1.4, 1.7.\\n\\n “[E]ven in a trust relationship,” the duty is “to disclose\\nonly material facts.” See Laurienti, 611 F.3d at 541.\\nGenerally, an omission is material “if there is ‘a substantial\\nlikelihood that the disclosure of the omitted fact would have\\nbeen viewed by the reasonable investor as having\\nsignificantly altered the “total mix” of information made\\navailable.’” SEC v. Phan, 500 F.3d 895, 908 (9th Cir. 2007)\\n(quoting Basic Inc. v. Levinson, 485 U.S. 224, 231–32\\n\\f USSEC V. FENG 27\\n\\n(1988)). Here, the record includes statements of individual\\ninvestors who said that if they had known about Feng’s\\ncommissions, they would have attempted to negotiate their\\nadministrative fees. Feng also testified that he concealed his\\ncommissions to avoid anticipated client requests for rebates.\\nHe said that he thought disclosure of his commissions would\\nencourage the clients to ask for more reductions in their\\nadministrative fees, and he did not want to open the door to\\nnegotiations that would lower his clients’ costs but decrease\\nhis payout. We thus agree with the district court that Feng’s\\nfailure to disclose his conflict of interest was material. See\\nTSC Indus., Inc. v. Northway, Inc., 426 U.S. 438, 450 (1976)\\n(holding that materiality may be resolved at summary\\njudgment “if the established omissions are so obviously\\nimportant to an investor[] that reasonable minds cannot\\ndiffer on the question of materiality” (internal quotation\\nmarks omitted)).\\n\\n Finally, Feng’s testimony establishes intent to deceive.\\nFeng stated that he did not want to tell clients about his\\ncommissions from the regional centers because it would be\\ncostly. Feng’s brief similarly describes his choice not to\\ndisclose as a “business decision.”\\n\\n 2. Schemes to Defraud (Section 17(a)(1) and (3) and\\n Rule 10b-5(a) and (c))\\n\\n The securities fraud provisions also prohibit any person,\\nin the offer or sale of securities, “to employ any device,\\nscheme, or artifice to defraud” or “to engage in any\\ntransaction, practice, or course of business which operates or\\nwould operate as a fraud or deceit upon the purchaser.”\\n15 U.S.C. § 77q(a)(1), (3); see also 17 C.F.R. § 240.10b-\\n5(a), (c). The state of mind requirement varies among these\\nprovisions, but a showing of intentional or knowing conduct\\nclears all thresholds. See Aaron, 446 U.S. at 695–97.\\n\\f28 USSEC V. FENG\\n\\n The district court found that Feng defrauded the regional\\ncenters that refused to pay commissions to U.S.-based\\nattorneys not registered as brokers. Feng evaded this\\nrestriction by informing the regional centers that he was\\naffiliated with foreign referral agents and companies,\\nholding himself out to the regional centers as a middleman.\\nFeng furthered this scheme by setting up ABCL in China to\\nreceive commission payments, while maintaining control of\\nboth ABCL’s bank account and operations. Feng performed\\nthe recruitment services himself and received the\\ncommissions that regional centers believed they were paying\\nto the affiliated recruiters. Feng recruited his mother to sign\\ndocuments as ABCL’s president, even though she had no\\nactual role in the company. Representatives from the\\nregional centers testified they would have ceased this\\narrangement if they knew Feng’s relationship with either the\\nagents or ABCL.\\n\\n The district court also found that Feng engaged in a\\nscheme to defraud clients who sought a reduction in their\\nadministrative fees. Generally, as noted, regional centers\\nwere willing to rebate a portion of the administrative fee that\\nthey charged investors if the regional center could also\\nreduce the commission it paid for the recruitment of that\\ninvestor. When a client asked Feng to negotiate with the\\nregional center to reduce his administrative fee, Feng\\narranged with the regional center to lower the administrative\\nfee by reducing the commission. He took measures to\\nconceal this arrangement from his clients, however, because\\nhe wanted to “keep as much of the marketing fee as possible”\\nand avoid further negotiation with clients, while also\\ncreating the appearance to his clients that he was prevailing\\non their behalf with the regional centers. Feng asked the\\nregional centers to rebate a portion of the administrative fee\\n\\f USSEC V. FENG 29\\n\\nbut not to disclose that the funds were reallocated from a\\ncommission.\\n\\n Feng challenges on two grounds the district court’s\\nfinding that his conduct violated the Securities and Exchange\\nActs. First, Feng asserts that the scheme-to-defraud\\nallegations under Rule 10b-5(a) and (c) improperly duplicate\\nthe Rule 10b-5(b) material omission claim. We see no\\nduplication. The district court addressed the fraud against\\nthe regional centers only as a scheme to defraud, not as a\\nmaterial omission. With regard to the fraud against the\\nclients, the district court focused on Feng’s failure to\\ndisclose his conflict of interest to his clients as material\\nomissions and, separately, on his efforts to conceal the\\nsource of the rebates as a scheme to defraud.\\n\\n Second, Feng argues that his establishment of overseas\\nentities and occasional arrangement of rebates for clients\\nwere not unlawful. But, deceptive conduct that is not\\n“inherently unlawful” may form the basis of a scheme to\\ndefraud. SEC v. Wey, 246 F. Supp. 3d 894, 918 (S.D.N.Y.\\n2017). In addition, Feng’s attempt to analogize the rebate\\nscheme to a store proprietor giving a customer a discount is\\ninapt. The fraudulent conduct was his scheme to make the\\nrebate appear to come from his advocacy on behalf of the\\nclient, while concealing the relationship between the\\nreduction and his undisclosed commission.\\n\\n We accordingly affirm the district court’s finding that\\nFeng engaged in securities fraud in violation of Section 17(a)\\nand Section 10(b).\\n\\n IV.\\n\\n Finally, Feng challenges the district court’s\\ndisgorgement order. The district court calculated that Feng\\n\\f30 USSEC V. FENG\\n\\nreceived $1.268 million for commissions in connection with\\nthe EB-5 program and ordered disgorgement of the entire\\namount. Feng contests the disgorgement of fees paid to his\\noverseas entities, suggesting that some of the fees were used\\nto cover expenses. He does not dispute that he controlled\\nthose overseas entities.\\n\\n We review the court’s imposition of disgorgement for\\nabuse of discretion. SEC v. JT Wallenbrock & Assocs.,\\n440 F.3d 1109, 1113 (9th Cir. 2006). A disgorgement\\ncalculation “should include ‘all gains flowing from the\\nillegal activities,’” id. at 1114 (quoting SEC v. Cross Fin.\\nServs., 908 F. Supp. 718, 734 (C.D. Cal. 1995), and “requires\\nonly a ‘reasonable approximation of profits causally\\nconnected to the violation,’” id. at 1113–14 (quoting SEC v.\\nFirst Pac. Bancorp, 142 F.3d 1186, 1192 n. 6 (9th Cir.\\n1998)).\\n\\n Feng used his overseas entities—ABCL and his other\\nnominees—to advance his scheme to defraud the regional\\ncenters. The money paid by regional centers to Feng’s\\noverseas entities is therefore exactly the focus of the fraud\\nagainst the regional centers, such that the district court could\\nfind a “causal[] connect[ion]” through which the funds\\n“flow[] from the illegal activities.” Id. at 1114. Feng should\\nnot have been collecting these commissions in the first place,\\nand it would be unjust to permit him to retain some of the ill-\\ngotten funds to cover his expenses. See id. (“[I]t would be\\nunjust to permit the defendants to offset against the investor\\ndollars they received the expenses of running the very\\nbusiness they created to defraud those investors into giving\\nthe defendants the money in the first place.”). We\\naccordingly find no abuse of discretion in the disgorgement\\norder.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4094,"cells":{"id":{"kind":"number","value":4655516,"string":"4,655,516"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Luis Pleitez-Lopez v. William Barr"},"decision_date":{"kind":"string","value":"2019-08-23"},"docket_number":{"kind":"string","value":"16-73656"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Mary M. Schroeder and Susan P. Graber, Circuit Judges, and Michael H. Watson,* District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"GRABER, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n LUIS URIEL PLEITEZ-LOPEZ, aka Luis No. 16-73656\\n Raul Pleitez-Rios,\\n Petitioner, Agency No.\\n A205-319-371\\n v.\\n\\n WILLIAM P. BARR, Attorney OPINION\\n General,\\n Respondent.\\n\\n\\n On Petition for Review of an Order of the\\n Board of Immigration Appeals\\n\\n Argued and Submitted August 12, 2019\\n Pasadena, California\\n\\n Filed August 23, 2019\\n\\n Before: Mary M. Schroeder and Susan P. Graber, Circuit\\n Judges, and Michael H. Watson,* District Judge.\\n\\n Opinion by Judge Graber\\n\\n\\n\\n\\n *\\n The Honorable Michael H. Watson, United States District Judge for\\nthe Southern District of Ohio, sitting by designation.\\n\\f2 PLEITEZ-LOPEZ V. BARR\\n\\n SUMMARY**\\n\\n\\n Immigration\\n\\n Granting Luis Pleitez-Lopez’s petition for review of a\\ndecision of the Board of Immigration Appeals affirming an\\nimmigration judge’s denial of a continuance for Petitioner to\\nupdate his fingerprints with the Department of Homeland\\nSecurity (DHS), the panel held that Petitioner’s reliance on\\nhis lawyer’s erroneous advice that he was not required to\\nupdate his fingerprints was reasonable and constituted “good\\ncause” to grant a continuance, and remanded.\\n\\n Petitioner provided fingerprints to DHS before his first\\nmerits hearing, but was granted a continuance to file for\\nrelief. At a later hearing, the IJ reset Petitioner’s case for a\\nlater date and instructed Petitioner that he must have his\\nfingerprints retaken 60 days before that hearing or the IJ\\nwould find that he had abandoned his applications for relief.\\nHowever, when his hearing date arrived, Petitioner had not\\nprovided the required fingerprints to DHS because his\\nattorney’s office had erroneously advised him that he was not\\nrequired to do so. The IJ denied the continuance, and the BIA\\naffirmed, concluding that Petitioner lacked good cause for\\nfailing to update his fingerprints because the IJ properly\\nadvised him of his obligation to update his fingerprints and\\nthe consequence for failing to do so.\\n\\n The panel held that the BIA’s decision to deny a\\ncontinuance was an abuse of discretion in two ways. First,\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f PLEITEZ-LOPEZ V. BARR 3\\n\\nthe BIA failed to analyze all the factors set out by Cui v.\\nMukasey, 538 F.3d 1289 (9th Cir. 2008), which instructs\\nthe court to consider: (1) the importance of the evidence,\\n(2) the unreasonableness of the immigrant’s conduct,\\n(3) the inconvenience to the court, and (4) the number of\\ncontinuances previously granted. Second, the panel held that\\nthe BIA analyzed the unreasonableness of Petitioner’s\\nconduct in an arbitrary and irrational manner. Contrary to the\\nBIA, the panel concluded that Petitioner acted reasonably in\\nrelying on his lawyer’s advice, explaining that Petitioner’s\\nreliance was especially reasonable because the lawyer’s bad\\nadvice post-dated the IJ’s instructions.\\n\\n\\n COUNSEL\\n\\nShawn J. Beam (argued), Los Angeles, California, for\\nPetitioner.\\n\\nJane T. Schaffner (argued), Trial Attorney; Douglas E.\\nGinsburg, Assistant Director; Office of Immigration\\nLitigation, Civil Division, United States Department of\\nJustice, Washington, D.C.; for Respondent.\\n\\f4 PLEITEZ-LOPEZ V. BARR\\n\\n OPINION\\n\\nGRABER, Circuit Judge:\\n\\n Petitioner Luis Pleitez-Lopez seeks review of the Board\\nof Immigration Appeals’ decision affirming the immigration\\njudge’s (“IJ”) denial of a continuance for Petitioner to update\\nhis fingerprints with the Department of Homeland Security\\n(“DHS”). Petitioner did not update his fingerprints because\\nhis lawyer advised him, incorrectly, that he was not required\\nto do so. He contends that his reliance on his lawyer’s advice\\nwas reasonable and constituted “good cause” to grant a\\ncontinuance under 8 C.F.R. § 1003.29. We agree and grant\\nthe petition.\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\n Petitioner, a native and citizen of Guatemala, entered the\\nUnited States in 2002 without being admitted or paroled. In\\nAugust 2012, the government placed Petitioner in removal\\nproceedings. Petitioner conceded removability and sought\\nvarious forms of relief.\\n\\n Petitioner originally provided the required fingerprints to\\nDHS in December 2013, before his first merits hearing. At\\nthat hearing, Petitioner’s lawyer sought a continuance to file\\nfor cancellation of removal; he had learned only recently that\\nPetitioner might meet the residency requirement for that form\\nof relief. The IJ granted the continuance.\\n\\n At a later hearing, the IJ accepted Petitioner’s application\\nfor cancellation of removal and reset the case for a merits\\nhearing in September 2014, when Petitioner’s fingerprints\\n\\f PLEITEZ-LOPEZ V. BARR 5\\n\\nstill would have been valid.1 But, at the September hearing,\\nthe IJ reset the merits hearing to May 2015 because he had\\n“to do another case.” After resetting the hearing date, the IJ\\n(through an interpreter) instructed Petitioner that he must\\nhave his fingerprints retaken 60 days before the May hearing\\nor the IJ would find that Petitioner had abandoned his\\napplications for relief. The government’s lawyer gave\\nPetitioner a written set of fingerprint instructions, and\\nPetitioner confirmed to the IJ that he understood the\\nfingerprint requirement.\\n\\n When the May 2015 hearing arrived, Petitioner had not\\nprovided updated fingerprints to DHS. His lawyer’s office,\\noperating under the erroneous belief that DHS fingerprints\\nremained valid for 18 months, told Petitioner that he need not\\nsubmit updated fingerprints to DHS before the May hearing.\\nPetitioner had, however, recently provided updated\\nfingerprints to the California Department of Justice. Thus,\\nPetitioner mistakenly believed that he had complied with the\\nfingerprint requirement, telling the IJ: “I have to repeat this\\nagain, that I talked to my attorney and he said to me, ‘These\\nare the ones you need to take.’”\\n\\n In light of the erroneous legal advice, Petitioner’s lawyer\\nrequested a “brief continuance” for Petitioner to provide\\nupdated fingerprints, but the IJ denied the continuance. The\\nIJ repeatedly expressed disbelief that Petitioner had relied on\\nhis lawyer’s advice over the IJ’s contrary instructions at the\\nearlier hearing. “A misunderstanding! I told him, very\\nclearly gave him—have [sic] the trial attorney give him\\n\\n\\n 1\\n DHS fingerprints expire after 15 months, Cui v. Mukasey, 538 F.3d\\n1289, 1292 (9th Cir. 2008), so Petitioner’s December 2013 fingerprints\\nremained valid through at least part of March 2015.\\n\\f6 PLEITEZ-LOPEZ V. BARR\\n\\nwritten instructions and told him to follow those\\ninstructions.” “Who does he believe, the judge or the lawyer?\\n. . . You shouldn’t have to even think about this.”\\n\\n The IJ held that Petitioner lacked good cause for failing\\nto update his fingerprints, because Petitioner was “not\\nreasonable in disregarding what the court instructed him to do\\neven if his counsel gave him other instructions.” The IJ also\\nfound that Petitioner’s failure to obtain updated fingerprints\\nburdened the government because, if DHS could not perform\\nthe corresponding background investigation, the government\\ncould not prepare adequately for the merits hearing. (But the\\ngovernment did not object when Petitioner requested a\\ncontinuance.) The IJ deemed Petitioner’s applications\\nabandoned and granted him voluntary departure to\\nGuatemala.\\n\\n The BIA affirmed the IJ’s decision that Petitioner lacked\\ngood cause for failing to update his fingerprints. According\\nto the BIA, Petitioner lacked good cause because the IJ\\n“properly advised [Petitioner] of his obligations to comply\\nwith the biometric processing requirements and the\\nconsequences for failing to do so.” Petitioner timely sought\\nour review.\\n\\n STANDARD OF REVIEW\\n\\n “Where, as here, the BIA reviewed the IJ’s factual\\nfindings for clear error, and reviewed de novo all other\\nissues,” we limit our review to the BIA’s decision, except to\\nthe extent that it expressly adopted the IJ’s opinion. Singh v.\\nWhitaker, 914 F.3d 654, 658 (9th Cir. 2019). We review for\\nabuse of discretion an agency’s denial of a continuance.\\nMalilia v. Holder, 632 F.3d 598, 602 (9th Cir. 2011).\\n\\f PLEITEZ-LOPEZ V. BARR 7\\n\\n DISCUSSION\\n\\n Under Cui v. Mukasey, 538 F.3d 1289, 1295–96 (9th Cir.\\n2008), “it is clearly an abuse of discretion for an IJ to deny a\\nrequest from an otherwise diligent applicant for a short\\ncontinuance to submit fingerprints.” That holding alone\\nmight resolve this case (despite Cui’s statement that “there\\nare no bright-line rules for deciding when an IJ’s denial of a\\ncontinuance warrants reversal”). Id. at 1295. But Cui\\ninstructs us to consider four factors to assess whether the\\ndenial of a continuance constituted an abuse of discretion:\\n“(1) the importance of the evidence, (2) the unreasonableness\\nof the immigrant’s conduct, (3) the inconvenience to the\\ncourt, and (4) the number of continuances previously\\ngranted.” Id. at 1292.\\n\\n The BIA considered only the unreasonableness of\\nPetitioner’s conduct and did not address the other factors or\\nadopt the IJ’s finding of a burden to the government. We\\nhold that the BIA’s decision to deny a continuance was an\\nabuse of discretion in two ways. The BIA failed to analyze\\nall the Cui factors and, in addition, analyzed the\\nunreasonableness of Petitioner’s conduct in an arbitrary and\\nirrational manner. See Hernandez-Velasquez v. Holder,\\n611 F.3d 1073, 1077 (9th Cir. 2010) (“An error of law is an\\nabuse of discretion.”); Avagyan v. Holder, 646 F.3d 672, 678\\n(9th Cir. 2011) (“The BIA abuses its discretion when its\\ndecision is arbitrary, irrational, or contrary to law.” (internal\\nquotation marks omitted)). We turn next to the Cui factors.\\n\\n First, the fingerprint evidence was vital to Petitioner’s\\napplications because he could not obtain relief without an\\nupdated fingerprint check. See Cui, 538 F.3d at 1292–93\\n(“Thus, when the IJ denied her request for a continuance to\\n\\f8 PLEITEZ-LOPEZ V. BARR\\n\\nresubmit her fingerprints, he effectively pretermitted any\\nhope Cui had of obtaining relief. Her case was over.”).\\n\\n Second, Petitioner acted reasonably in relying on his\\nlawyer’s advice that he did not need to update his fingerprints\\nwith DHS.\\n\\n For the alien unfamiliar with the laws of our\\n country, an attorney serves a special role in\\n helping the alien through a complex and\\n completely foreign process. It is therefore\\n reasonable for an alien to trust and rely upon\\n an attorney’s advice to such an extent that if\\n an alien fails to show up to a hearing because\\n of an attorney, we can say that this is an\\n exceptional circumstance “beyond the control\\n of the alien.” 8 U.S.C. § 1229a(e)(1).\\n\\nMonjaraz-Munoz v. INS, 327 F.3d 892, 897 (9th Cir. 2003).\\nLikewise, if an alien fails to provide updated fingerprints to\\nDHS “because of an attorney,” id., the alien’s reliance on the\\nlawyer’s advice may constitute the requisite “good cause” for\\na continuance under 8 C.F.R. § 1003.29. Here, because the\\nlawyer’s bad advice post-dated the IJ’s instructions,\\nPetitioner’s reliance on the later advice was especially\\nreasonable. For example, the fingerprint requirements might\\nhave changed since his last hearing, including by lengthening\\nthe period during which fingerprints were valid or by\\nallowing the California Department of Justice to send\\nfingerprints to DHS.2 If the lawyer’s bad advice had\\n\\n\\n 2\\n See Mendoza-Ortiz v. Holder, 464 F. App’x 615, 617 (9th Cir. 2011)\\n(unpublished) (holding that the IJ abused his discretion by denying a\\ncontinuance to complete a “fingerprinting and background check” where\\n\\f PLEITEZ-LOPEZ V. BARR 9\\n\\npreceded the IJ’s instructions, we might reach a different\\nresult.\\n\\n Of course, the IJ warned Petitioner that he must submit\\nupdated fingerprints and that his applications would be\\ndeemed abandoned if he did not do so. The government cited\\nseveral unpublished cases for the proposition that Petitioner\\nbehaved unreasonably because the IJ warned him of his\\nobligations and the consequences of failing to comply. But\\nnone of those unprecedential decisions involved petitioners\\nwho failed to submit updated fingerprints because their\\nlawyers told them not to do so after the IJ had given the\\nfingerprint instructions. In other words, none of those\\npetitioners demonstrated good cause for a continuance. By\\ncontrast, here Petitioner did show good cause; he relied on his\\nlawyer’s advice because of the special relationship between\\nan alien who is navigating the complex immigration process\\nand the lawyer who is helping him. Monjaraz-Munoz,\\n327 F.3d at 897. The BIA’s emphasis on the IJ’s instructions\\nto Petitioner failed to account for that special relationship or\\nfor the fact that the lawyer’s advice not to provide updated\\nfingerprints post-dated the IJ’s instructions to Petitioner.\\n\\n Third, the inconvenience to the court of granting the\\ncontinuance would have been minimal. Cui explained that\\ngranting a continuance in these circumstances “would not\\nhave resulted in any cognizable inconvenience to the court or\\n\\n\\nthe petitioner’s lawyer “told the IJ that his office had mistakenly advised\\n[the petitioner] that he had completed all required background checks”).\\nIn concurrence, Judge Silverman wrote: “The lawyer’s conduct here was\\ninexcusable, but the client’s reliance on his lawyer was not.” Id. at 618\\n(Silverman, J., concurring). We are not bound by Mendoza-Ortiz, but we\\nfind its reasoning persuasive.\\n\\f10 PLEITEZ-LOPEZ V. BARR\\n\\nthe government.” 538 F.3d at 1295. Because the IJ already\\nhad reserved time for the May 2015 merits hearing, the IJ\\ncould have allowed Petitioner to present testimony without\\nsuffering any inconvenience.3 Id. Then the IJ could have\\neither “issued a tentative grant of relief conditioned on the\\nsubmission and analysis of [Petitioner’s] fingerprints” or\\n“denied relief immediately.” Id. The former option would\\nhave required merely “a brief hearing to hear the security\\ncheck results” at a later date. Id. In the alternative, the\\nrecord discloses no reason why a continuance without taking\\nevidence would have inconvenienced the court.\\n\\n Fourth, and finally, before his request for the fingerprint\\ncontinuance, Petitioner had sought just one continuance (to\\nfile his application for cancellation of removal). In Cui, we\\nfound that this factor weighed in the petitioner’s favor even\\nwhere she had “requested few continuances” before the\\ncontinuance to submit fingerprints. 538 F.3d at 1295. Thus,\\nthis factor weighs in Petitioner’s favor because he also\\nrequested few continuances.\\n\\n Petition GRANTED and REMANDED for further\\nproceedings consistent with this opinion.\\n\\n\\n\\n\\n 3\\n The government argued that this case differs from Cui because\\nPetitioner initially requested a continuance during the May 2015 hearing\\nfor a different reason—his witness was unavailable because she had to\\nhave her fingerprints taken that day. That is true, but the BIA’s decision\\nrelied solely on the absence of updated fingerprints. And, even without\\nhis other witness, Petitioner still could have presented part of his case\\nthrough his own testimony and the documentary evidence that he had\\npreviously submitted.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432769/\", \"author_raw\": \"GRABER, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MARY M SCHROEDER"},"judge2":{"kind":"string","value":"SUSAN P GRABER"},"judge3":{"kind":"string","value":"MICHAEL H WATSON"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4655516/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Luis Pleitez-Lopez v. William Barr\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-23\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n LUIS URIEL PLEITEZ-LOPEZ, aka Luis No. 16-73656\n Raul Pleitez-Rios,\n Petitioner, Agency No.\n A205-319-371\n v.\n\n WILLIAM P. BARR, Attorney OPINION\n General,\n Respondent.\n\n\n On Petition for Review of an Order of the\n Board of Immigration Appeals\n\n Argued and Submitted August 12, 2019\n Pasadena, California\n\n Filed August 23, 2019\n\n Before: Mary M. Schroeder and Susan P. Graber, Circuit\n Judges, and Michael H. Watson,* District Judge.\n\n Opinion by Judge Graber\n\n\n\n\n *\n The Honorable Michael H. Watson, United States District Judge for\nthe Southern District of Ohio, sitting by designation.\n\f2 PLEITEZ-LOPEZ V. BARR\n\n SUMMARY**\n\n\n Immigration\n\n Granting Luis Pleitez-Lopez’s petition for review of a\ndecision of the Board of Immigration Appeals affirming an\nimmigration judge’s denial of a continuance for Petitioner to\nupdate his fingerprints with the Department of Homeland\nSecurity (DHS), the panel held that Petitioner’s reliance on\nhis lawyer’s erroneous advice that he was not required to\nupdate his fingerprints was reasonable and constituted “good\ncause” to grant a continuance, and remanded.\n\n Petitioner provided fingerprints to DHS before his first\nmerits hearing, but was granted a continuance to file for\nrelief. At a later hearing, the IJ reset Petitioner’s case for a\nlater date and instructed Petitioner that he must have his\nfingerprints retaken 60 days before that hearing or the IJ\nwould find that he had abandoned his applications for relief.\nHowever, when his hearing date arrived, Petitioner had not\nprovided the required fingerprints to DHS because his\nattorney’s office had erroneously advised him that he was not\nrequired to do so. The IJ denied the continuance, and the BIA\naffirmed, concluding that Petitioner lacked good cause for\nfailing to update his fingerprints because the IJ properly\nadvised him of his obligation to update his fingerprints and\nthe consequence for failing to do so.\n\n The panel held that the BIA’s decision to deny a\ncontinuance was an abuse of discretion in two ways. First,\n\n **\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\f PLEITEZ-LOPEZ V. BARR 3\n\nthe BIA failed to analyze all the factors set out by Cui v.\nMukasey, 538 F.3d 1289 (9th Cir. 2008), which instructs\nthe court to consider: (1) the importance of the evidence,\n(2) the unreasonableness of the immigrant’s conduct,\n(3) the inconvenience to the court, and (4) the number of\ncontinuances previously granted. Second, the panel held that\nthe BIA analyzed the unreasonableness of Petitioner’s\nconduct in an arbitrary and irrational manner. Contrary to the\nBIA, the panel concluded that Petitioner acted reasonably in\nrelying on his lawyer’s advice, explaining that Petitioner’s\nreliance was especially reasonable because the lawyer’s bad\nadvice post-dated the IJ’s instructions.\n\n\n COUNSEL\n\nShawn J. Beam (argued), Los Angeles, California, for\nPetitioner.\n\nJane T. Schaffner (argued), Trial Attorney; Douglas E.\nGinsburg, Assistant Director; Office of Immigration\nLitigation, Civil Division, United States Department of\nJustice, Washington, D.C.; for Respondent.\n\f4 PLEITEZ-LOPEZ V. BARR\n\n OPINION\n\nGRABER, Circuit Judge:\n\n Petitioner Luis Pleitez-Lopez seeks review of the Board\nof Immigration Appeals’ decision affirming the immigration\njudge’s (“IJ”) denial of a continuance for Petitioner to update\nhis fingerprints with the Department of Homeland Security\n(“DHS”). Petitioner did not update his fingerprints because\nhis lawyer advised him, incorrectly, that he was not required\nto do so. He contends that his reliance on his lawyer’s advice\nwas reasonable and constituted “good cause” to grant a\ncontinuance under 8 C.F.R. § 1003.29. We agree and grant\nthe petition.\n\n FACTUAL AND PROCEDURAL BACKGROUND\n\n Petitioner, a native and citizen of Guatemala, entered the\nUnited States in 2002 without being admitted or paroled. In\nAugust 2012, the government placed Petitioner in removal\nproceedings. Petitioner conceded removability and sought\nvarious forms of relief.\n\n Petitioner originally provided the required fingerprints to\nDHS in December 2013, before his first merits hearing. At\nthat hearing, Petitioner’s lawyer sought a continuance to file\nfor cancellation of removal; he had learned only recently that\nPetitioner might meet the residency requirement for that form\nof relief. The IJ granted the continuance.\n\n At a later hearing, the IJ accepted Petitioner’s application\nfor cancellation of removal and reset the case for a merits\nhearing in September 2014, when Petitioner’s fingerprints\n\f PLEITEZ-LOPEZ V. BARR 5\n\nstill would have been valid.1 But, at the September hearing,\nthe IJ reset the merits hearing to May 2015 because he had\n“to do another case.” After resetting the hearing date, the IJ\n(through an interpreter) instructed Petitioner that he must\nhave his fingerprints retaken 60 days before the May hearing\nor the IJ would find that Petitioner had abandoned his\napplications for relief. The government’s lawyer gave\nPetitioner a written set of fingerprint instructions, and\nPetitioner confirmed to the IJ that he understood the\nfingerprint requirement.\n\n When the May 2015 hearing arrived, Petitioner had not\nprovided updated fingerprints to DHS. His lawyer’s office,\noperating under the erroneous belief that DHS fingerprints\nremained valid for 18 months, told Petitioner that he need not\nsubmit updated fingerprints to DHS before the May hearing.\nPetitioner had, however, recently provided updated\nfingerprints to the California Department of Justice. Thus,\nPetitioner mistakenly believed that he had complied with the\nfingerprint requirement, telling the IJ: “I have to repeat this\nagain, that I talked to my attorney and he said to me, ‘These\nare the ones you need to take.’”\n\n In light of the erroneous legal advice, Petitioner’s lawyer\nrequested a “brief continuance” for Petitioner to provide\nupdated fingerprints, but the IJ denied the continuance. The\nIJ repeatedly expressed disbelief that Petitioner had relied on\nhis lawyer’s advice over the IJ’s contrary instructions at the\nearlier hearing. “A misunderstanding! I told him, very\nclearly gave him—have [sic] the trial attorney give him\n\n\n 1\n DHS fingerprints expire after 15 months, Cui v. Mukasey, 538 F.3d\n1289, 1292 (9th Cir. 2008), so Petitioner’s December 2013 fingerprints\nremained valid through at least part of March 2015.\n\f6 PLEITEZ-LOPEZ V. BARR\n\nwritten instructions and told him to follow those\ninstructions.” “Who does he believe, the judge or the lawyer?\n. . . You shouldn’t have to even think about this.”\n\n The IJ held that Petitioner lacked good cause for failing\nto update his fingerprints, because Petitioner was “not\nreasonable in disregarding what the court instructed him to do\neven if his counsel gave him other instructions.” The IJ also\nfound that Petitioner’s failure to obtain updated fingerprints\nburdened the government because, if DHS could not perform\nthe corresponding background investigation, the government\ncould not prepare adequately for the merits hearing. (But the\ngovernment did not object when Petitioner requested a\ncontinuance.) The IJ deemed Petitioner’s applications\nabandoned and granted him voluntary departure to\nGuatemala.\n\n The BIA affirmed the IJ’s decision that Petitioner lacked\ngood cause for failing to update his fingerprints. According\nto the BIA, Petitioner lacked good cause because the IJ\n“properly advised [Petitioner] of his obligations to comply\nwith the biometric processing requirements and the\nconsequences for failing to do so.” Petitioner timely sought\nour review.\n\n STANDARD OF REVIEW\n\n “Where, as here, the BIA reviewed the IJ’s factual\nfindings for clear error, and reviewed de novo all other\nissues,” we limit our review to the BIA’s decision, except to\nthe extent that it expressly adopted the IJ’s opinion. Singh v.\nWhitaker, 914 F.3d 654, 658 (9th Cir. 2019). We review for\nabuse of discretion an agency’s denial of a continuance.\nMalilia v. Holder, 632 F.3d 598, 602 (9th Cir. 2011).\n\f PLEITEZ-LOPEZ V. BARR 7\n\n DISCUSSION\n\n Under Cui v. Mukasey, 538 F.3d 1289, 1295–96 (9th Cir.\n2008), “it is clearly an abuse of discretion for an IJ to deny a\nrequest from an otherwise diligent applicant for a short\ncontinuance to submit fingerprints.” That holding alone\nmight resolve this case (despite Cui’s statement that “there\nare no bright-line rules for deciding when an IJ’s denial of a\ncontinuance warrants reversal”). Id. at 1295. But Cui\ninstructs us to consider four factors to assess whether the\ndenial of a continuance constituted an abuse of discretion:\n“(1) the importance of the evidence, (2) the unreasonableness\nof the immigrant’s conduct, (3) the inconvenience to the\ncourt, and (4) the number of continuances previously\ngranted.” Id. at 1292.\n\n The BIA considered only the unreasonableness of\nPetitioner’s conduct and did not address the other factors or\nadopt the IJ’s finding of a burden to the government. We\nhold that the BIA’s decision to deny a continuance was an\nabuse of discretion in two ways. The BIA failed to analyze\nall the Cui factors and, in addition, analyzed the\nunreasonableness of Petitioner’s conduct in an arbitrary and\nirrational manner. See Hernandez-Velasquez v. Holder,\n611 F.3d 1073, 1077 (9th Cir. 2010) (“An error of law is an\nabuse of discretion.”); Avagyan v. Holder, 646 F.3d 672, 678\n(9th Cir. 2011) (“The BIA abuses its discretion when its\ndecision is arbitrary, irrational, or contrary to law.” (internal\nquotation marks omitted)). We turn next to the Cui factors.\n\n First, the fingerprint evidence was vital to Petitioner’s\napplications because he could not obtain relief without an\nupdated fingerprint check. See Cui, 538 F.3d at 1292–93\n(“Thus, when the IJ denied her request for a continuance to\n\f8 PLEITEZ-LOPEZ V. BARR\n\nresubmit her fingerprints, he effectively pretermitted any\nhope Cui had of obtaining relief. Her case was over.”).\n\n Second, Petitioner acted reasonably in relying on his\nlawyer’s advice that he did not need to update his fingerprints\nwith DHS.\n\n For the alien unfamiliar with the laws of our\n country, an attorney serves a special role in\n helping the alien through a complex and\n completely foreign process. It is therefore\n reasonable for an alien to trust and rely upon\n an attorney’s advice to such an extent that if\n an alien fails to show up to a hearing because\n of an attorney, we can say that this is an\n exceptional circumstance “beyond the control\n of the alien.” 8 U.S.C. § 1229a(e)(1).\n\nMonjaraz-Munoz v. INS, 327 F.3d 892, 897 (9th Cir. 2003).\nLikewise, if an alien fails to provide updated fingerprints to\nDHS “because of an attorney,” id., the alien’s reliance on the\nlawyer’s advice may constitute the requisite “good cause” for\na continuance under 8 C.F.R. § 1003.29. Here, because the\nlawyer’s bad advice post-dated the IJ’s instructions,\nPetitioner’s reliance on the later advice was especially\nreasonable. For example, the fingerprint requirements might\nhave changed since his last hearing, including by lengthening\nthe period during which fingerprints were valid or by\nallowing the California Department of Justice to send\nfingerprints to DHS.2 If the lawyer’s bad advice had\n\n\n 2\n See Mendoza-Ortiz v. Holder, 464 F. App’x 615, 617 (9th Cir. 2011)\n(unpublished) (holding that the IJ abused his discretion by denying a\ncontinuance to complete a “fingerprinting and background check” where\n\f PLEITEZ-LOPEZ V. BARR 9\n\npreceded the IJ’s instructions, we might reach a different\nresult.\n\n Of course, the IJ warned Petitioner that he must submit\nupdated fingerprints and that his applications would be\ndeemed abandoned if he did not do so. The government cited\nseveral unpublished cases for the proposition that Petitioner\nbehaved unreasonably because the IJ warned him of his\nobligations and the consequences of failing to comply. But\nnone of those unprecedential decisions involved petitioners\nwho failed to submit updated fingerprints because their\nlawyers told them not to do so after the IJ had given the\nfingerprint instructions. In other words, none of those\npetitioners demonstrated good cause for a continuance. By\ncontrast, here Petitioner did show good cause; he relied on his\nlawyer’s advice because of the special relationship between\nan alien who is navigating the complex immigration process\nand the lawyer who is helping him. Monjaraz-Munoz,\n327 F.3d at 897. The BIA’s emphasis on the IJ’s instructions\nto Petitioner failed to account for that special relationship or\nfor the fact that the lawyer’s advice not to provide updated\nfingerprints post-dated the IJ’s instructions to Petitioner.\n\n Third, the inconvenience to the court of granting the\ncontinuance would have been minimal. Cui explained that\ngranting a continuance in these circumstances “would not\nhave resulted in any cognizable inconvenience to the court or\n\n\nthe petitioner’s lawyer “told the IJ that his office had mistakenly advised\n[the petitioner] that he had completed all required background checks”).\nIn concurrence, Judge Silverman wrote: “The lawyer’s conduct here was\ninexcusable, but the client’s reliance on his lawyer was not.” Id. at 618\n(Silverman, J., concurring). We are not bound by Mendoza-Ortiz, but we\nfind its reasoning persuasive.\n\f10 PLEITEZ-LOPEZ V. BARR\n\nthe government.” 538 F.3d at 1295. Because the IJ already\nhad reserved time for the May 2015 merits hearing, the IJ\ncould have allowed Petitioner to present testimony without\nsuffering any inconvenience.3 Id. Then the IJ could have\neither “issued a tentative grant of relief conditioned on the\nsubmission and analysis of [Petitioner’s] fingerprints” or\n“denied relief immediately.” Id. The former option would\nhave required merely “a brief hearing to hear the security\ncheck results” at a later date. Id. In the alternative, the\nrecord discloses no reason why a continuance without taking\nevidence would have inconvenienced the court.\n\n Fourth, and finally, before his request for the fingerprint\ncontinuance, Petitioner had sought just one continuance (to\nfile his application for cancellation of removal). In Cui, we\nfound that this factor weighed in the petitioner’s favor even\nwhere she had “requested few continuances” before the\ncontinuance to submit fingerprints. 538 F.3d at 1295. Thus,\nthis factor weighs in Petitioner’s favor because he also\nrequested few continuances.\n\n Petition GRANTED and REMANDED for further\nproceedings consistent with this opinion.\n\n\n\n\n 3\n The government argued that this case differs from Cui because\nPetitioner initially requested a continuance during the May 2015 hearing\nfor a different reason—his witness was unavailable because she had to\nhave her fingerprints taken that day. That is true, but the BIA’s decision\nrelied solely on the absence of updated fingerprints. And, even without\nhis other witness, Petitioner still could have presented part of his case\nthrough his own testimony and the documentary evidence that he had\npreviously submitted.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Luis Pleitez-Lopez v. William Barr\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-23\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n LUIS URIEL PLEITEZ-LOPEZ, aka Luis No. 16-73656\\n Raul Pleitez-Rios,\\n Petitioner, Agency No.\\n A205-319-371\\n v.\\n\\n WILLIAM P. BARR, Attorney OPINION\\n General,\\n Respondent.\\n\\n\\n On Petition for Review of an Order of the\\n Board of Immigration Appeals\\n\\n Argued and Submitted August 12, 2019\\n Pasadena, California\\n\\n Filed August 23, 2019\\n\\n Before: Mary M. Schroeder and Susan P. Graber, Circuit\\n Judges, and Michael H. Watson,* District Judge.\\n\\n Opinion by Judge Graber\\n\\n\\n\\n\\n *\\n The Honorable Michael H. Watson, United States District Judge for\\nthe Southern District of Ohio, sitting by designation.\\n\\f2 PLEITEZ-LOPEZ V. BARR\\n\\n SUMMARY**\\n\\n\\n Immigration\\n\\n Granting Luis Pleitez-Lopez’s petition for review of a\\ndecision of the Board of Immigration Appeals affirming an\\nimmigration judge’s denial of a continuance for Petitioner to\\nupdate his fingerprints with the Department of Homeland\\nSecurity (DHS), the panel held that Petitioner’s reliance on\\nhis lawyer’s erroneous advice that he was not required to\\nupdate his fingerprints was reasonable and constituted “good\\ncause” to grant a continuance, and remanded.\\n\\n Petitioner provided fingerprints to DHS before his first\\nmerits hearing, but was granted a continuance to file for\\nrelief. At a later hearing, the IJ reset Petitioner’s case for a\\nlater date and instructed Petitioner that he must have his\\nfingerprints retaken 60 days before that hearing or the IJ\\nwould find that he had abandoned his applications for relief.\\nHowever, when his hearing date arrived, Petitioner had not\\nprovided the required fingerprints to DHS because his\\nattorney’s office had erroneously advised him that he was not\\nrequired to do so. The IJ denied the continuance, and the BIA\\naffirmed, concluding that Petitioner lacked good cause for\\nfailing to update his fingerprints because the IJ properly\\nadvised him of his obligation to update his fingerprints and\\nthe consequence for failing to do so.\\n\\n The panel held that the BIA’s decision to deny a\\ncontinuance was an abuse of discretion in two ways. First,\\n\\n **\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\f PLEITEZ-LOPEZ V. BARR 3\\n\\nthe BIA failed to analyze all the factors set out by Cui v.\\nMukasey, 538 F.3d 1289 (9th Cir. 2008), which instructs\\nthe court to consider: (1) the importance of the evidence,\\n(2) the unreasonableness of the immigrant’s conduct,\\n(3) the inconvenience to the court, and (4) the number of\\ncontinuances previously granted. Second, the panel held that\\nthe BIA analyzed the unreasonableness of Petitioner’s\\nconduct in an arbitrary and irrational manner. Contrary to the\\nBIA, the panel concluded that Petitioner acted reasonably in\\nrelying on his lawyer’s advice, explaining that Petitioner’s\\nreliance was especially reasonable because the lawyer’s bad\\nadvice post-dated the IJ’s instructions.\\n\\n\\n COUNSEL\\n\\nShawn J. Beam (argued), Los Angeles, California, for\\nPetitioner.\\n\\nJane T. Schaffner (argued), Trial Attorney; Douglas E.\\nGinsburg, Assistant Director; Office of Immigration\\nLitigation, Civil Division, United States Department of\\nJustice, Washington, D.C.; for Respondent.\\n\\f4 PLEITEZ-LOPEZ V. BARR\\n\\n OPINION\\n\\nGRABER, Circuit Judge:\\n\\n Petitioner Luis Pleitez-Lopez seeks review of the Board\\nof Immigration Appeals’ decision affirming the immigration\\njudge’s (“IJ”) denial of a continuance for Petitioner to update\\nhis fingerprints with the Department of Homeland Security\\n(“DHS”). Petitioner did not update his fingerprints because\\nhis lawyer advised him, incorrectly, that he was not required\\nto do so. He contends that his reliance on his lawyer’s advice\\nwas reasonable and constituted “good cause” to grant a\\ncontinuance under 8 C.F.R. § 1003.29. We agree and grant\\nthe petition.\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\n Petitioner, a native and citizen of Guatemala, entered the\\nUnited States in 2002 without being admitted or paroled. In\\nAugust 2012, the government placed Petitioner in removal\\nproceedings. Petitioner conceded removability and sought\\nvarious forms of relief.\\n\\n Petitioner originally provided the required fingerprints to\\nDHS in December 2013, before his first merits hearing. At\\nthat hearing, Petitioner’s lawyer sought a continuance to file\\nfor cancellation of removal; he had learned only recently that\\nPetitioner might meet the residency requirement for that form\\nof relief. The IJ granted the continuance.\\n\\n At a later hearing, the IJ accepted Petitioner’s application\\nfor cancellation of removal and reset the case for a merits\\nhearing in September 2014, when Petitioner’s fingerprints\\n\\f PLEITEZ-LOPEZ V. BARR 5\\n\\nstill would have been valid.1 But, at the September hearing,\\nthe IJ reset the merits hearing to May 2015 because he had\\n“to do another case.” After resetting the hearing date, the IJ\\n(through an interpreter) instructed Petitioner that he must\\nhave his fingerprints retaken 60 days before the May hearing\\nor the IJ would find that Petitioner had abandoned his\\napplications for relief. The government’s lawyer gave\\nPetitioner a written set of fingerprint instructions, and\\nPetitioner confirmed to the IJ that he understood the\\nfingerprint requirement.\\n\\n When the May 2015 hearing arrived, Petitioner had not\\nprovided updated fingerprints to DHS. His lawyer’s office,\\noperating under the erroneous belief that DHS fingerprints\\nremained valid for 18 months, told Petitioner that he need not\\nsubmit updated fingerprints to DHS before the May hearing.\\nPetitioner had, however, recently provided updated\\nfingerprints to the California Department of Justice. Thus,\\nPetitioner mistakenly believed that he had complied with the\\nfingerprint requirement, telling the IJ: “I have to repeat this\\nagain, that I talked to my attorney and he said to me, ‘These\\nare the ones you need to take.’”\\n\\n In light of the erroneous legal advice, Petitioner’s lawyer\\nrequested a “brief continuance” for Petitioner to provide\\nupdated fingerprints, but the IJ denied the continuance. The\\nIJ repeatedly expressed disbelief that Petitioner had relied on\\nhis lawyer’s advice over the IJ’s contrary instructions at the\\nearlier hearing. “A misunderstanding! I told him, very\\nclearly gave him—have [sic] the trial attorney give him\\n\\n\\n 1\\n DHS fingerprints expire after 15 months, Cui v. Mukasey, 538 F.3d\\n1289, 1292 (9th Cir. 2008), so Petitioner’s December 2013 fingerprints\\nremained valid through at least part of March 2015.\\n\\f6 PLEITEZ-LOPEZ V. BARR\\n\\nwritten instructions and told him to follow those\\ninstructions.” “Who does he believe, the judge or the lawyer?\\n. . . You shouldn’t have to even think about this.”\\n\\n The IJ held that Petitioner lacked good cause for failing\\nto update his fingerprints, because Petitioner was “not\\nreasonable in disregarding what the court instructed him to do\\neven if his counsel gave him other instructions.” The IJ also\\nfound that Petitioner’s failure to obtain updated fingerprints\\nburdened the government because, if DHS could not perform\\nthe corresponding background investigation, the government\\ncould not prepare adequately for the merits hearing. (But the\\ngovernment did not object when Petitioner requested a\\ncontinuance.) The IJ deemed Petitioner’s applications\\nabandoned and granted him voluntary departure to\\nGuatemala.\\n\\n The BIA affirmed the IJ’s decision that Petitioner lacked\\ngood cause for failing to update his fingerprints. According\\nto the BIA, Petitioner lacked good cause because the IJ\\n“properly advised [Petitioner] of his obligations to comply\\nwith the biometric processing requirements and the\\nconsequences for failing to do so.” Petitioner timely sought\\nour review.\\n\\n STANDARD OF REVIEW\\n\\n “Where, as here, the BIA reviewed the IJ’s factual\\nfindings for clear error, and reviewed de novo all other\\nissues,” we limit our review to the BIA’s decision, except to\\nthe extent that it expressly adopted the IJ’s opinion. Singh v.\\nWhitaker, 914 F.3d 654, 658 (9th Cir. 2019). We review for\\nabuse of discretion an agency’s denial of a continuance.\\nMalilia v. Holder, 632 F.3d 598, 602 (9th Cir. 2011).\\n\\f PLEITEZ-LOPEZ V. BARR 7\\n\\n DISCUSSION\\n\\n Under Cui v. Mukasey, 538 F.3d 1289, 1295–96 (9th Cir.\\n2008), “it is clearly an abuse of discretion for an IJ to deny a\\nrequest from an otherwise diligent applicant for a short\\ncontinuance to submit fingerprints.” That holding alone\\nmight resolve this case (despite Cui’s statement that “there\\nare no bright-line rules for deciding when an IJ’s denial of a\\ncontinuance warrants reversal”). Id. at 1295. But Cui\\ninstructs us to consider four factors to assess whether the\\ndenial of a continuance constituted an abuse of discretion:\\n“(1) the importance of the evidence, (2) the unreasonableness\\nof the immigrant’s conduct, (3) the inconvenience to the\\ncourt, and (4) the number of continuances previously\\ngranted.” Id. at 1292.\\n\\n The BIA considered only the unreasonableness of\\nPetitioner’s conduct and did not address the other factors or\\nadopt the IJ’s finding of a burden to the government. We\\nhold that the BIA’s decision to deny a continuance was an\\nabuse of discretion in two ways. The BIA failed to analyze\\nall the Cui factors and, in addition, analyzed the\\nunreasonableness of Petitioner’s conduct in an arbitrary and\\nirrational manner. See Hernandez-Velasquez v. Holder,\\n611 F.3d 1073, 1077 (9th Cir. 2010) (“An error of law is an\\nabuse of discretion.”); Avagyan v. Holder, 646 F.3d 672, 678\\n(9th Cir. 2011) (“The BIA abuses its discretion when its\\ndecision is arbitrary, irrational, or contrary to law.” (internal\\nquotation marks omitted)). We turn next to the Cui factors.\\n\\n First, the fingerprint evidence was vital to Petitioner’s\\napplications because he could not obtain relief without an\\nupdated fingerprint check. See Cui, 538 F.3d at 1292–93\\n(“Thus, when the IJ denied her request for a continuance to\\n\\f8 PLEITEZ-LOPEZ V. BARR\\n\\nresubmit her fingerprints, he effectively pretermitted any\\nhope Cui had of obtaining relief. Her case was over.”).\\n\\n Second, Petitioner acted reasonably in relying on his\\nlawyer’s advice that he did not need to update his fingerprints\\nwith DHS.\\n\\n For the alien unfamiliar with the laws of our\\n country, an attorney serves a special role in\\n helping the alien through a complex and\\n completely foreign process. It is therefore\\n reasonable for an alien to trust and rely upon\\n an attorney’s advice to such an extent that if\\n an alien fails to show up to a hearing because\\n of an attorney, we can say that this is an\\n exceptional circumstance “beyond the control\\n of the alien.” 8 U.S.C. § 1229a(e)(1).\\n\\nMonjaraz-Munoz v. INS, 327 F.3d 892, 897 (9th Cir. 2003).\\nLikewise, if an alien fails to provide updated fingerprints to\\nDHS “because of an attorney,” id., the alien’s reliance on the\\nlawyer’s advice may constitute the requisite “good cause” for\\na continuance under 8 C.F.R. § 1003.29. Here, because the\\nlawyer’s bad advice post-dated the IJ’s instructions,\\nPetitioner’s reliance on the later advice was especially\\nreasonable. For example, the fingerprint requirements might\\nhave changed since his last hearing, including by lengthening\\nthe period during which fingerprints were valid or by\\nallowing the California Department of Justice to send\\nfingerprints to DHS.2 If the lawyer’s bad advice had\\n\\n\\n 2\\n See Mendoza-Ortiz v. Holder, 464 F. App’x 615, 617 (9th Cir. 2011)\\n(unpublished) (holding that the IJ abused his discretion by denying a\\ncontinuance to complete a “fingerprinting and background check” where\\n\\f PLEITEZ-LOPEZ V. BARR 9\\n\\npreceded the IJ’s instructions, we might reach a different\\nresult.\\n\\n Of course, the IJ warned Petitioner that he must submit\\nupdated fingerprints and that his applications would be\\ndeemed abandoned if he did not do so. The government cited\\nseveral unpublished cases for the proposition that Petitioner\\nbehaved unreasonably because the IJ warned him of his\\nobligations and the consequences of failing to comply. But\\nnone of those unprecedential decisions involved petitioners\\nwho failed to submit updated fingerprints because their\\nlawyers told them not to do so after the IJ had given the\\nfingerprint instructions. In other words, none of those\\npetitioners demonstrated good cause for a continuance. By\\ncontrast, here Petitioner did show good cause; he relied on his\\nlawyer’s advice because of the special relationship between\\nan alien who is navigating the complex immigration process\\nand the lawyer who is helping him. Monjaraz-Munoz,\\n327 F.3d at 897. The BIA’s emphasis on the IJ’s instructions\\nto Petitioner failed to account for that special relationship or\\nfor the fact that the lawyer’s advice not to provide updated\\nfingerprints post-dated the IJ’s instructions to Petitioner.\\n\\n Third, the inconvenience to the court of granting the\\ncontinuance would have been minimal. Cui explained that\\ngranting a continuance in these circumstances “would not\\nhave resulted in any cognizable inconvenience to the court or\\n\\n\\nthe petitioner’s lawyer “told the IJ that his office had mistakenly advised\\n[the petitioner] that he had completed all required background checks”).\\nIn concurrence, Judge Silverman wrote: “The lawyer’s conduct here was\\ninexcusable, but the client’s reliance on his lawyer was not.” Id. at 618\\n(Silverman, J., concurring). We are not bound by Mendoza-Ortiz, but we\\nfind its reasoning persuasive.\\n\\f10 PLEITEZ-LOPEZ V. BARR\\n\\nthe government.” 538 F.3d at 1295. Because the IJ already\\nhad reserved time for the May 2015 merits hearing, the IJ\\ncould have allowed Petitioner to present testimony without\\nsuffering any inconvenience.3 Id. Then the IJ could have\\neither “issued a tentative grant of relief conditioned on the\\nsubmission and analysis of [Petitioner’s] fingerprints” or\\n“denied relief immediately.” Id. The former option would\\nhave required merely “a brief hearing to hear the security\\ncheck results” at a later date. Id. In the alternative, the\\nrecord discloses no reason why a continuance without taking\\nevidence would have inconvenienced the court.\\n\\n Fourth, and finally, before his request for the fingerprint\\ncontinuance, Petitioner had sought just one continuance (to\\nfile his application for cancellation of removal). In Cui, we\\nfound that this factor weighed in the petitioner’s favor even\\nwhere she had “requested few continuances” before the\\ncontinuance to submit fingerprints. 538 F.3d at 1295. Thus,\\nthis factor weighs in Petitioner’s favor because he also\\nrequested few continuances.\\n\\n Petition GRANTED and REMANDED for further\\nproceedings consistent with this opinion.\\n\\n\\n\\n\\n 3\\n The government argued that this case differs from Cui because\\nPetitioner initially requested a continuance during the May 2015 hearing\\nfor a different reason—his witness was unavailable because she had to\\nhave her fingerprints taken that day. That is true, but the BIA’s decision\\nrelied solely on the absence of updated fingerprints. And, even without\\nhis other witness, Petitioner still could have presented part of his case\\nthrough his own testimony and the documentary evidence that he had\\npreviously submitted.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4095,"cells":{"id":{"kind":"number","value":4655517,"string":"4,655,517"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"FTC v. Qualcomm Incorporated"},"decision_date":{"kind":"string","value":"2019-08-23"},"docket_number":{"kind":"string","value":"19-16122"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PER CURIAM\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION FILED\\n UNITED STATES COURT OF APPEALS AUG 23 2019\\n MOLLY C. DWYER, CLERK\\n U.S. COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\nFEDERAL TRADE COMMISSION, No. 19-16122\\n\\n Plaintiff-Appellee, D.C. No. 5:17-cv-00220-LHK\\n Northern District of California,\\n v. San Jose\\n\\nQUALCOMM INCORPORATED, a ORDER\\nDelaware corporation,\\n\\n Defendant-Appellant,\\n\\nSAMSUNG ELECTRONICS COMPANY,\\nLTD.; et al.,\\n\\n Intervenors,\\n\\n\\n\\nBefore: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\\n\\nPER CURIAM:\\n\\n Appellant Qualcomm Incorporated (“Qualcomm”) moves for a partial stay\\n\\npending appeal of the district court’s May 21, 2019 permanent injunction, which it\\n\\nentered following a trial on antitrust claims brought by the Federal Trade\\n\\nCommission (“FTC”). We grant Qualcomm’s motion.\\n\\n The FTC alleged that Qualcomm, a leader in cellular standard technology,\\n\\nviolated Sections 1 and 2 of the Sherman Act and Section 5 of the FTC Act in\\n\\fconnection with the licensing of its standard essential patents (“SEPs”) and sale of\\n\\nits code division multiple access (“CDMA”) and premium long-term evolution\\n\\n(“LTE”) modem chips. Specifically, Qualcomm refused to license SEPs to rival\\n\\nchip suppliers, allegedly in contravention of commitments Qualcomm made to\\n\\ncertain standard setting organizations in the industry; refused to sell modem chips\\n\\nto any original equipment manufacturers (“OEMs”) that lacked patent licensing\\n\\nagreements with Qualcomm; and imposed in its OEM licensing agreements\\n\\nexcessive royalty rates on a per-handset basis, irrespective of whether the handset\\n\\ncontained a Qualcomm chip or a chip from one of Qualcomm’s competitors. The\\n\\ncomplaint alleged that the upshot of this conduct was to maintain Qualcomm’s\\n\\nmonopoly in the CDMA and premium LTE chip markets and impose an\\n\\nanticompetitive surcharge on its competitors’ chips.\\n\\n After a ten-day trial, the district court issued extensive findings of fact and\\n\\ndetermined that Qualcomm’s practices violate the antitrust laws. The district court\\n\\nconcluded that Qualcomm (1) has an antitrust duty to license its SEPs to rival chip\\n\\nsuppliers, and (2) engaged in anticompetitive conduct by using its royalty rates to\\n\\neffectively impose a surcharge on its competitors’ chips. The district court entered\\n\\na multipart permanent injunction.\\n\\n Qualcomm seeks a stay of the injunction’s provisions requiring that\\n\\nQualcomm make exhaustive SEP licenses available to its competitors, prohibiting\\n\\n\\n 2 19-16122\\n\\fQualcomm from conditioning chip sales on the purchase of patent licenses, and\\n\\nrequiring Qualcomm to negotiate or renegotiate its license agreements in that\\n\\nrespect.\\n\\n To determine whether to issue a stay pending appeal, we consider “(1)\\n\\nwhether the stay applicant has made a strong showing that he is likely to succeed\\n\\non the merits; (2) whether the applicant will be irreparably injured absent a stay;\\n\\n(3) whether issuance of the stay will substantially injure the other parties interested\\n\\nin the proceeding; and (4) where the public interest lies.” Nken v. Holder, 556 U.S.\\n\\n418, 426 (2009) (quoting Hilton v. Braunskill, 481 U.S. 770, 776 (1987)). An\\n\\napplicant for a stay “need not demonstrate that it is more likely than not they will\\n\\nwin on the merits,” but rather must show “a reasonable probability” or “fair\\n\\nprospect” of success. Leiva-Perez v. Holder, 640 F.3d 962, 966–67 (9th Cir. 2011)\\n\\n(quoting Hollingsworth v. Perry, 558 U.S. 183, 190 (2010)). Applying those\\n\\nfactors here, we grant Qualcomm’s motion for a partial stay of the injunction\\n\\npending appeal.\\n\\n It is well-settled that, “as a general matter, the Sherman Act ‘does not restrict\\n\\nthe . . . right of [a] trader or manufacturer engaged in an entirely private business,\\n\\nfreely to exercise his own independent discretion as to parties with whom he will\\n\\ndeal.’” Verizon Commc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP (“Trinko”),\\n\\n540 U.S. 398, 408 (2004) (second alteration in original) (quoting United States v.\\n\\n\\n 3 19-16122\\n\\fColgate & Co., 250 U.S. 300, 307 (1919)). The Supreme Court recognized a very\\n\\nlimited exception to that general rule when a monopolist terminated a voluntary\\n\\nand profitable course of dealing with a competitor and sacrificed short-term\\n\\nbenefits to exclude competition in the long run. See generally Aspen Skiing Co. v.\\n\\nAspen Highlands Skiing Corp., 472 U.S. 585 (1985). That exception, however, is\\n\\n“at or near the outer boundary of [Sherman Act] liability.” Trinko, 540 U.S. at\\n\\n409. And, here, even the two government agencies charged with the enforcement\\n\\nof antitrust laws—the FTC and the Antitrust Division of the Department of Justice\\n\\n(“DOJ”), see FTC v. AT&T Mobility LLC, 883 F.3d 848, 862 (9th Cir. 2018) (en\\n\\nbanc)—disagree as to whether Qualcomm’s conduct implicates the duty to deal.\\n\\nIndeed, while the FTC prosecuted this antitrust enforcement action, the DOJ filed a\\n\\nstatement of interest expressing its stark disagreement that Qualcomm has any\\n\\nantitrust duty to deal with rival chip suppliers.\\n\\n We are satisfied that Qualcomm has shown, at minimum, the presence of\\n\\nserious questions on the merits of the district court’s determination that Qualcomm\\n\\nhas an antitrust duty to license its SEPs to rival chip suppliers. See Lair v. Bullock,\\n\\n697 F.3d 1200, 1204 (9th Cir. 2012). Qualcomm likewise has made the requisite\\n\\nshowing that its practice of charging OEMs royalties for its patents on a per-\\n\\n\\n\\n\\n 4 19-16122\\n\\fhandset basis does not violate the antitrust laws.1 See Doe v. Abbott Labs., 571\\n\\nF.3d 930, 931 (9th Cir. 2009) (holding that “allegations of monopoly leveraging\\n\\nthrough pricing conduct in two markets” do not “state a claim under § 2 of the\\n\\nSherman Act absent an antitrust refusal to deal (or some other exclusionary\\n\\npractice) in the monopoly market or below-cost pricing in the second market”\\n\\n(citation omitted)).\\n\\n Turning to the second Nken factor, we conclude that Qualcomm has\\n\\ndemonstrated a probability of irreparable harm. The injunction requires\\n\\nQualcomm to enter new contractual relationships and renegotiate existing ones on\\n\\na large scale. The fundamental business changes that the injunction imposes\\n\\ncannot be easily undone should Qualcomm prevail on appeal. See NCAA v. Bd. of\\n\\nRegents of Univ. of Okla., 463 U.S. 1311, 1313–14 (1983) (White, Circuit Justice)\\n\\n(equities favored stay where, absent a stay, appellant’s contracts to broadcast\\n\\ncollegiate football games would be void and could not be enforced, putting at risk\\n\\nbusiness for entire season); Am. Trucking Ass’ns, Inc. v. City of Los Angeles, 559\\n\\nF.3d 1046, 1057–59 (9th Cir. 2009) (irreparable harm likely where order subjected\\n\\n\\n\\n1\\n Breaking from her standard practice, then-FTC Commissioner Maureen K.\\nOhlhausen issued a written dissenting statement to express her disagreement with\\nthe theory urged in the complaint and adopted by the district court that\\nQualcomm’s royalty rates operate as an exclusionary tax or surcharge on\\ncompetitor products. See Dissenting Statement of Commissioner Maureen K.\\nOhlhausen In the Matter of Qualcomm, Inc., No. 141-0199, January 17, 2017.\\n\\n 5 19-16122\\n\\fplaintiff to immediate “Hobson’s choice” of either (1) signing agreements that\\n\\nwould cause it to “incur large costs” and “disrupt and change the whole nature of\\n\\nits business” or (2) refusing to sign agreements, causing “a loss of customer\\n\\ngoodwill” and potentially entire loss of business).\\n\\n Finally, the balance of equites also weighs in favor of a stay. See Lair, 697\\n\\nF.3d at 1215. Although the hardship to the party opposing the stay and the public\\n\\ninterest usually merge when the government is the opposing party, see Nken, 556\\n\\nU.S. at 435, this case is unique, as the government itself is divided about the\\n\\npropriety of the judgment and its impact on the public interest. Indeed, the\\n\\nDepartment of Defense and Department of Energy aver that the injunction\\n\\nthreatens national security, and the DOJ posits that the injunction has the effect of\\n\\nharming rather than benefiting consumers.\\n\\n Whether the district court’s order and injunction represent a trailblazing\\n\\napplication of the antitrust laws, or instead an improper excursion beyond the outer\\n\\nlimits of the Sherman Act, is a matter for another day. For now, weighing all\\n\\nrelevant factors, we conclude that the requested stay is warranted. Therefore,\\n\\npending the resolution of this appeal or until further order of this court, we stay the\\n\\nportions of the district court’s injunction requiring that (1) “Qualcomm must make\\n\\nexhaustive SEP licenses available to modem-chip suppliers,” and (2) “Qualcomm\\n\\nmust not condition the supply of modem chips on a customer’s patent license\\n\\n\\n 6 19-16122\\n\\fstatus” and “must negotiate or renegotiate license terms” with its customers in that\\n\\nrespect. This stay has the effect of maintaining the status quo ante during this\\n\\nexpedited appeal. See id. at 429 (“A stay ‘simply suspend[s] judicial alteration of\\n\\nthe status quo[.]’” (first alteration in original) (quoting Ohio Citizens for\\n\\nResponsible Energy, Inc. v. NRC, 479 U.S. 1213, 1313 (1986) (Scalia, Circuit\\n\\nJustice))).\\n\\n The current briefing schedule shall remain in effect, and the clerk shall place\\n\\nthis appeal on the calendar for January 2020. See 9th Cir. Gen. Order 3.3(g).\\n\\n So ordered.\\n\\n\\n\\n\\n 7 19-16122\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432770/\", \"author_raw\": \"PER CURIAM\"}]}"},"judge1":{"kind":"string","value":"TASHIMA"},"judge2":{"kind":"string","value":"M SMITH"},"judge3":{"kind":"string","value":"BENNETT"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{\"TASHIMA\": \", Circuit\", \"M SMITH\": \", Circuit\", \"BENNETT\": \", Circuit\"}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4655517/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: FTC v. Qualcomm Incorporated\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-23\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION FILED\n UNITED STATES COURT OF APPEALS AUG 23 2019\n MOLLY C. DWYER, CLERK\n U.S. COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\nFEDERAL TRADE COMMISSION, No. 19-16122\n\n Plaintiff-Appellee, D.C. No. 5:17-cv-00220-LHK\n Northern District of California,\n v. San Jose\n\nQUALCOMM INCORPORATED, a ORDER\nDelaware corporation,\n\n Defendant-Appellant,\n\nSAMSUNG ELECTRONICS COMPANY,\nLTD.; et al.,\n\n Intervenors,\n\n\n\nBefore: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\n\nPER CURIAM:\n\n Appellant Qualcomm Incorporated (“Qualcomm”) moves for a partial stay\n\npending appeal of the district court’s May 21, 2019 permanent injunction, which it\n\nentered following a trial on antitrust claims brought by the Federal Trade\n\nCommission (“FTC”). We grant Qualcomm’s motion.\n\n The FTC alleged that Qualcomm, a leader in cellular standard technology,\n\nviolated Sections 1 and 2 of the Sherman Act and Section 5 of the FTC Act in\n\fconnection with the licensing of its standard essential patents (“SEPs”) and sale of\n\nits code division multiple access (“CDMA”) and premium long-term evolution\n\n(“LTE”) modem chips. Specifically, Qualcomm refused to license SEPs to rival\n\nchip suppliers, allegedly in contravention of commitments Qualcomm made to\n\ncertain standard setting organizations in the industry; refused to sell modem chips\n\nto any original equipment manufacturers (“OEMs”) that lacked patent licensing\n\nagreements with Qualcomm; and imposed in its OEM licensing agreements\n\nexcessive royalty rates on a per-handset basis, irrespective of whether the handset\n\ncontained a Qualcomm chip or a chip from one of Qualcomm’s competitors. The\n\ncomplaint alleged that the upshot of this conduct was to maintain Qualcomm’s\n\nmonopoly in the CDMA and premium LTE chip markets and impose an\n\nanticompetitive surcharge on its competitors’ chips.\n\n After a ten-day trial, the district court issued extensive findings of fact and\n\ndetermined that Qualcomm’s practices violate the antitrust laws. The district court\n\nconcluded that Qualcomm (1) has an antitrust duty to license its SEPs to rival chip\n\nsuppliers, and (2) engaged in anticompetitive conduct by using its royalty rates to\n\neffectively impose a surcharge on its competitors’ chips. The district court entered\n\na multipart permanent injunction.\n\n Qualcomm seeks a stay of the injunction’s provisions requiring that\n\nQualcomm make exhaustive SEP licenses available to its competitors, prohibiting\n\n\n 2 19-16122\n\fQualcomm from conditioning chip sales on the purchase of patent licenses, and\n\nrequiring Qualcomm to negotiate or renegotiate its license agreements in that\n\nrespect.\n\n To determine whether to issue a stay pending appeal, we consider “(1)\n\nwhether the stay applicant has made a strong showing that he is likely to succeed\n\non the merits; (2) whether the applicant will be irreparably injured absent a stay;\n\n(3) whether issuance of the stay will substantially injure the other parties interested\n\nin the proceeding; and (4) where the public interest lies.” Nken v. Holder, 556 U.S.\n\n418, 426 (2009) (quoting Hilton v. Braunskill, 481 U.S. 770, 776 (1987)). An\n\napplicant for a stay “need not demonstrate that it is more likely than not they will\n\nwin on the merits,” but rather must show “a reasonable probability” or “fair\n\nprospect” of success. Leiva-Perez v. Holder, 640 F.3d 962, 966–67 (9th Cir. 2011)\n\n(quoting Hollingsworth v. Perry, 558 U.S. 183, 190 (2010)). Applying those\n\nfactors here, we grant Qualcomm’s motion for a partial stay of the injunction\n\npending appeal.\n\n It is well-settled that, “as a general matter, the Sherman Act ‘does not restrict\n\nthe . . . right of [a] trader or manufacturer engaged in an entirely private business,\n\nfreely to exercise his own independent discretion as to parties with whom he will\n\ndeal.’” Verizon Commc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP (“Trinko”),\n\n540 U.S. 398, 408 (2004) (second alteration in original) (quoting United States v.\n\n\n 3 19-16122\n\fColgate & Co., 250 U.S. 300, 307 (1919)). The Supreme Court recognized a very\n\nlimited exception to that general rule when a monopolist terminated a voluntary\n\nand profitable course of dealing with a competitor and sacrificed short-term\n\nbenefits to exclude competition in the long run. See generally Aspen Skiing Co. v.\n\nAspen Highlands Skiing Corp., 472 U.S. 585 (1985). That exception, however, is\n\n“at or near the outer boundary of [Sherman Act] liability.” Trinko, 540 U.S. at\n\n409. And, here, even the two government agencies charged with the enforcement\n\nof antitrust laws—the FTC and the Antitrust Division of the Department of Justice\n\n(“DOJ”), see FTC v. AT&T Mobility LLC, 883 F.3d 848, 862 (9th Cir. 2018) (en\n\nbanc)—disagree as to whether Qualcomm’s conduct implicates the duty to deal.\n\nIndeed, while the FTC prosecuted this antitrust enforcement action, the DOJ filed a\n\nstatement of interest expressing its stark disagreement that Qualcomm has any\n\nantitrust duty to deal with rival chip suppliers.\n\n We are satisfied that Qualcomm has shown, at minimum, the presence of\n\nserious questions on the merits of the district court’s determination that Qualcomm\n\nhas an antitrust duty to license its SEPs to rival chip suppliers. See Lair v. Bullock,\n\n697 F.3d 1200, 1204 (9th Cir. 2012). Qualcomm likewise has made the requisite\n\nshowing that its practice of charging OEMs royalties for its patents on a per-\n\n\n\n\n 4 19-16122\n\fhandset basis does not violate the antitrust laws.1 See Doe v. Abbott Labs., 571\n\nF.3d 930, 931 (9th Cir. 2009) (holding that “allegations of monopoly leveraging\n\nthrough pricing conduct in two markets” do not “state a claim under § 2 of the\n\nSherman Act absent an antitrust refusal to deal (or some other exclusionary\n\npractice) in the monopoly market or below-cost pricing in the second market”\n\n(citation omitted)).\n\n Turning to the second Nken factor, we conclude that Qualcomm has\n\ndemonstrated a probability of irreparable harm. The injunction requires\n\nQualcomm to enter new contractual relationships and renegotiate existing ones on\n\na large scale. The fundamental business changes that the injunction imposes\n\ncannot be easily undone should Qualcomm prevail on appeal. See NCAA v. Bd. of\n\nRegents of Univ. of Okla., 463 U.S. 1311, 1313–14 (1983) (White, Circuit Justice)\n\n(equities favored stay where, absent a stay, appellant’s contracts to broadcast\n\ncollegiate football games would be void and could not be enforced, putting at risk\n\nbusiness for entire season); Am. Trucking Ass’ns, Inc. v. City of Los Angeles, 559\n\nF.3d 1046, 1057–59 (9th Cir. 2009) (irreparable harm likely where order subjected\n\n\n\n1\n Breaking from her standard practice, then-FTC Commissioner Maureen K.\nOhlhausen issued a written dissenting statement to express her disagreement with\nthe theory urged in the complaint and adopted by the district court that\nQualcomm’s royalty rates operate as an exclusionary tax or surcharge on\ncompetitor products. See Dissenting Statement of Commissioner Maureen K.\nOhlhausen In the Matter of Qualcomm, Inc., No. 141-0199, January 17, 2017.\n\n 5 19-16122\n\fplaintiff to immediate “Hobson’s choice” of either (1) signing agreements that\n\nwould cause it to “incur large costs” and “disrupt and change the whole nature of\n\nits business” or (2) refusing to sign agreements, causing “a loss of customer\n\ngoodwill” and potentially entire loss of business).\n\n Finally, the balance of equites also weighs in favor of a stay. See Lair, 697\n\nF.3d at 1215. Although the hardship to the party opposing the stay and the public\n\ninterest usually merge when the government is the opposing party, see Nken, 556\n\nU.S. at 435, this case is unique, as the government itself is divided about the\n\npropriety of the judgment and its impact on the public interest. Indeed, the\n\nDepartment of Defense and Department of Energy aver that the injunction\n\nthreatens national security, and the DOJ posits that the injunction has the effect of\n\nharming rather than benefiting consumers.\n\n Whether the district court’s order and injunction represent a trailblazing\n\napplication of the antitrust laws, or instead an improper excursion beyond the outer\n\nlimits of the Sherman Act, is a matter for another day. For now, weighing all\n\nrelevant factors, we conclude that the requested stay is warranted. Therefore,\n\npending the resolution of this appeal or until further order of this court, we stay the\n\nportions of the district court’s injunction requiring that (1) “Qualcomm must make\n\nexhaustive SEP licenses available to modem-chip suppliers,” and (2) “Qualcomm\n\nmust not condition the supply of modem chips on a customer’s patent license\n\n\n 6 19-16122\n\fstatus” and “must negotiate or renegotiate license terms” with its customers in that\n\nrespect. This stay has the effect of maintaining the status quo ante during this\n\nexpedited appeal. See id. at 429 (“A stay ‘simply suspend[s] judicial alteration of\n\nthe status quo[.]’” (first alteration in original) (quoting Ohio Citizens for\n\nResponsible Energy, Inc. v. NRC, 479 U.S. 1213, 1313 (1986) (Scalia, Circuit\n\nJustice))).\n\n The current briefing schedule shall remain in effect, and the clerk shall place\n\nthis appeal on the calendar for January 2020. See 9th Cir. Gen. Order 3.3(g).\n\n So ordered.\n\n\n\n\n 7 19-16122\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: FTC v. Qualcomm Incorporated\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-23\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION FILED\\n UNITED STATES COURT OF APPEALS AUG 23 2019\\n MOLLY C. DWYER, CLERK\\n U.S. COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\nFEDERAL TRADE COMMISSION, No. 19-16122\\n\\n Plaintiff-Appellee, D.C. No. 5:17-cv-00220-LHK\\n Northern District of California,\\n v. San Jose\\n\\nQUALCOMM INCORPORATED, a ORDER\\nDelaware corporation,\\n\\n Defendant-Appellant,\\n\\nSAMSUNG ELECTRONICS COMPANY,\\nLTD.; et al.,\\n\\n Intervenors,\\n\\n\\n\\nBefore: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.\\n\\nPER CURIAM:\\n\\n Appellant Qualcomm Incorporated (“Qualcomm”) moves for a partial stay\\n\\npending appeal of the district court’s May 21, 2019 permanent injunction, which it\\n\\nentered following a trial on antitrust claims brought by the Federal Trade\\n\\nCommission (“FTC”). We grant Qualcomm’s motion.\\n\\n The FTC alleged that Qualcomm, a leader in cellular standard technology,\\n\\nviolated Sections 1 and 2 of the Sherman Act and Section 5 of the FTC Act in\\n\\fconnection with the licensing of its standard essential patents (“SEPs”) and sale of\\n\\nits code division multiple access (“CDMA”) and premium long-term evolution\\n\\n(“LTE”) modem chips. Specifically, Qualcomm refused to license SEPs to rival\\n\\nchip suppliers, allegedly in contravention of commitments Qualcomm made to\\n\\ncertain standard setting organizations in the industry; refused to sell modem chips\\n\\nto any original equipment manufacturers (“OEMs”) that lacked patent licensing\\n\\nagreements with Qualcomm; and imposed in its OEM licensing agreements\\n\\nexcessive royalty rates on a per-handset basis, irrespective of whether the handset\\n\\ncontained a Qualcomm chip or a chip from one of Qualcomm’s competitors. The\\n\\ncomplaint alleged that the upshot of this conduct was to maintain Qualcomm’s\\n\\nmonopoly in the CDMA and premium LTE chip markets and impose an\\n\\nanticompetitive surcharge on its competitors’ chips.\\n\\n After a ten-day trial, the district court issued extensive findings of fact and\\n\\ndetermined that Qualcomm’s practices violate the antitrust laws. The district court\\n\\nconcluded that Qualcomm (1) has an antitrust duty to license its SEPs to rival chip\\n\\nsuppliers, and (2) engaged in anticompetitive conduct by using its royalty rates to\\n\\neffectively impose a surcharge on its competitors’ chips. The district court entered\\n\\na multipart permanent injunction.\\n\\n Qualcomm seeks a stay of the injunction’s provisions requiring that\\n\\nQualcomm make exhaustive SEP licenses available to its competitors, prohibiting\\n\\n\\n 2 19-16122\\n\\fQualcomm from conditioning chip sales on the purchase of patent licenses, and\\n\\nrequiring Qualcomm to negotiate or renegotiate its license agreements in that\\n\\nrespect.\\n\\n To determine whether to issue a stay pending appeal, we consider “(1)\\n\\nwhether the stay applicant has made a strong showing that he is likely to succeed\\n\\non the merits; (2) whether the applicant will be irreparably injured absent a stay;\\n\\n(3) whether issuance of the stay will substantially injure the other parties interested\\n\\nin the proceeding; and (4) where the public interest lies.” Nken v. Holder, 556 U.S.\\n\\n418, 426 (2009) (quoting Hilton v. Braunskill, 481 U.S. 770, 776 (1987)). An\\n\\napplicant for a stay “need not demonstrate that it is more likely than not they will\\n\\nwin on the merits,” but rather must show “a reasonable probability” or “fair\\n\\nprospect” of success. Leiva-Perez v. Holder, 640 F.3d 962, 966–67 (9th Cir. 2011)\\n\\n(quoting Hollingsworth v. Perry, 558 U.S. 183, 190 (2010)). Applying those\\n\\nfactors here, we grant Qualcomm’s motion for a partial stay of the injunction\\n\\npending appeal.\\n\\n It is well-settled that, “as a general matter, the Sherman Act ‘does not restrict\\n\\nthe . . . right of [a] trader or manufacturer engaged in an entirely private business,\\n\\nfreely to exercise his own independent discretion as to parties with whom he will\\n\\ndeal.’” Verizon Commc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP (“Trinko”),\\n\\n540 U.S. 398, 408 (2004) (second alteration in original) (quoting United States v.\\n\\n\\n 3 19-16122\\n\\fColgate & Co., 250 U.S. 300, 307 (1919)). The Supreme Court recognized a very\\n\\nlimited exception to that general rule when a monopolist terminated a voluntary\\n\\nand profitable course of dealing with a competitor and sacrificed short-term\\n\\nbenefits to exclude competition in the long run. See generally Aspen Skiing Co. v.\\n\\nAspen Highlands Skiing Corp., 472 U.S. 585 (1985). That exception, however, is\\n\\n“at or near the outer boundary of [Sherman Act] liability.” Trinko, 540 U.S. at\\n\\n409. And, here, even the two government agencies charged with the enforcement\\n\\nof antitrust laws—the FTC and the Antitrust Division of the Department of Justice\\n\\n(“DOJ”), see FTC v. AT&T Mobility LLC, 883 F.3d 848, 862 (9th Cir. 2018) (en\\n\\nbanc)—disagree as to whether Qualcomm’s conduct implicates the duty to deal.\\n\\nIndeed, while the FTC prosecuted this antitrust enforcement action, the DOJ filed a\\n\\nstatement of interest expressing its stark disagreement that Qualcomm has any\\n\\nantitrust duty to deal with rival chip suppliers.\\n\\n We are satisfied that Qualcomm has shown, at minimum, the presence of\\n\\nserious questions on the merits of the district court’s determination that Qualcomm\\n\\nhas an antitrust duty to license its SEPs to rival chip suppliers. See Lair v. Bullock,\\n\\n697 F.3d 1200, 1204 (9th Cir. 2012). Qualcomm likewise has made the requisite\\n\\nshowing that its practice of charging OEMs royalties for its patents on a per-\\n\\n\\n\\n\\n 4 19-16122\\n\\fhandset basis does not violate the antitrust laws.1 See Doe v. Abbott Labs., 571\\n\\nF.3d 930, 931 (9th Cir. 2009) (holding that “allegations of monopoly leveraging\\n\\nthrough pricing conduct in two markets” do not “state a claim under § 2 of the\\n\\nSherman Act absent an antitrust refusal to deal (or some other exclusionary\\n\\npractice) in the monopoly market or below-cost pricing in the second market”\\n\\n(citation omitted)).\\n\\n Turning to the second Nken factor, we conclude that Qualcomm has\\n\\ndemonstrated a probability of irreparable harm. The injunction requires\\n\\nQualcomm to enter new contractual relationships and renegotiate existing ones on\\n\\na large scale. The fundamental business changes that the injunction imposes\\n\\ncannot be easily undone should Qualcomm prevail on appeal. See NCAA v. Bd. of\\n\\nRegents of Univ. of Okla., 463 U.S. 1311, 1313–14 (1983) (White, Circuit Justice)\\n\\n(equities favored stay where, absent a stay, appellant’s contracts to broadcast\\n\\ncollegiate football games would be void and could not be enforced, putting at risk\\n\\nbusiness for entire season); Am. Trucking Ass’ns, Inc. v. City of Los Angeles, 559\\n\\nF.3d 1046, 1057–59 (9th Cir. 2009) (irreparable harm likely where order subjected\\n\\n\\n\\n1\\n Breaking from her standard practice, then-FTC Commissioner Maureen K.\\nOhlhausen issued a written dissenting statement to express her disagreement with\\nthe theory urged in the complaint and adopted by the district court that\\nQualcomm’s royalty rates operate as an exclusionary tax or surcharge on\\ncompetitor products. See Dissenting Statement of Commissioner Maureen K.\\nOhlhausen In the Matter of Qualcomm, Inc., No. 141-0199, January 17, 2017.\\n\\n 5 19-16122\\n\\fplaintiff to immediate “Hobson’s choice” of either (1) signing agreements that\\n\\nwould cause it to “incur large costs” and “disrupt and change the whole nature of\\n\\nits business” or (2) refusing to sign agreements, causing “a loss of customer\\n\\ngoodwill” and potentially entire loss of business).\\n\\n Finally, the balance of equites also weighs in favor of a stay. See Lair, 697\\n\\nF.3d at 1215. Although the hardship to the party opposing the stay and the public\\n\\ninterest usually merge when the government is the opposing party, see Nken, 556\\n\\nU.S. at 435, this case is unique, as the government itself is divided about the\\n\\npropriety of the judgment and its impact on the public interest. Indeed, the\\n\\nDepartment of Defense and Department of Energy aver that the injunction\\n\\nthreatens national security, and the DOJ posits that the injunction has the effect of\\n\\nharming rather than benefiting consumers.\\n\\n Whether the district court’s order and injunction represent a trailblazing\\n\\napplication of the antitrust laws, or instead an improper excursion beyond the outer\\n\\nlimits of the Sherman Act, is a matter for another day. For now, weighing all\\n\\nrelevant factors, we conclude that the requested stay is warranted. Therefore,\\n\\npending the resolution of this appeal or until further order of this court, we stay the\\n\\nportions of the district court’s injunction requiring that (1) “Qualcomm must make\\n\\nexhaustive SEP licenses available to modem-chip suppliers,” and (2) “Qualcomm\\n\\nmust not condition the supply of modem chips on a customer’s patent license\\n\\n\\n 6 19-16122\\n\\fstatus” and “must negotiate or renegotiate license terms” with its customers in that\\n\\nrespect. This stay has the effect of maintaining the status quo ante during this\\n\\nexpedited appeal. See id. at 429 (“A stay ‘simply suspend[s] judicial alteration of\\n\\nthe status quo[.]’” (first alteration in original) (quoting Ohio Citizens for\\n\\nResponsible Energy, Inc. v. NRC, 479 U.S. 1213, 1313 (1986) (Scalia, Circuit\\n\\nJustice))).\\n\\n The current briefing schedule shall remain in effect, and the clerk shall place\\n\\nthis appeal on the calendar for January 2020. See 9th Cir. Gen. Order 3.3(g).\\n\\n So ordered.\\n\\n\\n\\n\\n 7 19-16122\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4096,"cells":{"id":{"kind":"number","value":4655518,"string":"4,655,518"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Courtney Bird v. State of Hawaii"},"decision_date":{"kind":"string","value":"2019-08-23"},"docket_number":{"kind":"string","value":"17-16076"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: Richard C. Tallman, Jay S. Bybee, and N. Randy Smith, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PER CURIAM\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCOURTNEY BIRD, No. 17-16076\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 1:15-cv-00304-\\n DKW-KJM\\nSTATE OF HAWAI‘I; DEPARTMENT OF\\nHUMAN SERVICES; DHS, SOCIAL\\nSERVICES DIVISION, CHILD OPINION\\nWELFARE SERVICES BRANCH;\\nPANKAJ BHANOT, Department of\\nHuman Services Director; JEFFREY\\nR. WOODLAND; DOES,\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the District of Hawai‘i\\n Derrick Kahala Watson, District Judge, Presiding\\n\\n Argued and Submitted February 12, 2019\\n Submission Vacated February 26, 2019\\n Resubmitted August 23, 2019\\n University of Hawai‘i Manoa\\n\\n Filed August 23, 2019\\n\\n Before: Richard C. Tallman, Jay S. Bybee,\\n and N. Randy Smith, Circuit Judges.\\n\\n2 BIRD V. STATE OF HAWAI‘I\\n\\n Per Curiam Opinion;\\n Concurrence by Judge Bybee\\n\\n\\n SUMMARY*\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s dismissal, on\\nstatute of limitations grounds, of a complaint brought\\npursuant to 42 U.S.C. § 1983 alleging that the Hawai‘i\\nDepartment of Human Services violated plaintiff’s right to\\ndue process when it listed her, without notice, on the State’s\\nChild Protective Services Central Registry.\\n\\n Plaintiff and her then-husband were placed on the State’s\\nRegistry after their 7-week-old baby died of cardiac arrest in\\n2007. Bird’s husband later confessed to killing the infant and\\na criminal investigation concluded that plaintiff was not a\\nsuspect. The Department of Health Services did not,\\nhowever, remove plaintiff’s name from the Registry, nor did\\nit notify her that her name was listed. In 2013, plaintiff\\ndiscovered, during a background check, that her name was on\\nthe Registry. Plaintiff’s attorney communicated with the\\nDepartment and threatened to sue when plaintiff’s name was\\nnot removed, but did not file suit until more than two years\\nlater.\\n\\n Applying Hawai‘i’s two-year statute of limitations, the\\npanel held that plaintiff’s complaint was not subject to any\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\n BIRD V. STATE OF HAWAI‘I 3\\n\\nexceptions from the normal discovery rule of accrual. The\\npanel held that plaintiff had knowledge of the injury giving\\nrise to her claim by May 2013 when she threatened to sue the\\nDepartment if she was not removed from the Registry.\\nAccordingly, her suit, filed in July 2015, was time-barred.\\n\\n The panel rejected plaintiff’s assertion that a claim\\nseeking injunctive relief to invalidate an ongoing\\nunconstitutional statutory and regulatory scheme does not\\naccrue until the statute is repealed. The panel held that the\\ntraditional interests of protecting defendants and facilitating\\nthe administration of claims was applicable to a due process\\nclaim and justified enforcing a limitations period through the\\ndiscovery rule of accrual. The panel further rejected\\nplaintiff’s claim that her complaint alleged a continuing\\nviolation. The panel held that because the violation plaintiff\\nalleged was the placement of her name on the Registry\\nwithout constitutionally required due process, she had\\nbrought only an individualized claim. As such, the\\nsystematic branch of the continuing violations doctrine was\\ninapplicable, and the discovery rule of accrual applied.\\n\\n Concurring in the per curiam opinion, Judge Bybee stated\\nthat plaintiff had assuredly stated a plausible due process\\nclaim and that the Department of Health Services should not\\ninterpret the panel’s decision, which was based on statute of\\nlimitations grounds, as in any way condoning the blatantly\\ninsufficient procedures by which the Department maintains\\nplaintiff’s name in its Central Child Abuse Registry.\\n\\n4 BIRD V. STATE OF HAWAI‘I\\n\\n COUNSEL\\n\\nMargery S. Bronster (argued), Robert M. Hatch, and Kelly A.\\nHiga, Bronster Fujichaku Robbins, Honolulu, Hawai‘i; John\\nY. U. Choi, Honolulu, Hawai‘i; for Plaintiff-Appellant.\\n\\nRyan M. Akamine (argued) and Caron M. Inagaki, Deputy\\nAttorneys General; Department of the Attorney General,\\nHonolulu, Hawai‘i; for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n On March 28, 2007, plaintiff-appellant Courtney Bird\\nreturned home from a dentist appointment to find her then-\\nhusband administering CPR to their seven-week-old baby,\\nwho later died at the hospital of cardiac arrest. As a result of\\nthe infant’s death, the Hawai‘i Department of Human\\nServices (“DHS”) listed both Bird and her husband on the\\nstate’s Central Child Abuse Registry (“CCAR”). Bird’s\\nhusband later confessed to killing their infant baby and the\\ncriminal investigation concluded that Bird was not a suspect.\\nDHS, however, did not remove Bird’s name from the CCAR.\\nThroughout this time, DHS never told Bird that she was listed\\non the registry.\\n\\n Bird did not learn that she had been listed on the CCAR\\nuntil June 2012—more than five years later—when the report\\nturned up on a background check. When a request for a\\nhearing and various communications with DHS failed to\\nresult in Bird’s removal from the CCAR, Bird’s attorney\\nthreatened DHS in May 2013 with suit in federal court. Bird\\n\\n BIRD V. STATE OF HAWAI‘I 5\\n\\ndid not file suit, however, until July 20, 2015—more than two\\nyears after her extensive communications with DHS and\\nprevious threat to sue. The district court determined that\\nBird’s claims brought under 42 U.S.C. § 1983 were untimely\\nand granted summary judgment in favor of the State.\\n\\n Because Bird’s complaint is not subject to any exception\\nfrom the normal discovery rule of accrual, we agree that\\nBird’s claim accrued no later than May 2013, when Bird\\nthreatened to sue DHS if she was not removed from the\\nCCAR. We affirm.\\n\\n I. FACTS AND PROCEEDINGS BELOW\\n\\nA. Statutory Background\\n\\n Like most states, Hawai‘i law requires DHS to maintain\\na central registry for reports of child abuse or neglect. See\\nHaw. Rev. Stat. § 350-2(d) (“The department shall maintain\\na central registry of reported child abuse or neglect cases\\n. . . .”). “Central registry reports are typically used to aid\\nagencies in the investigation, treatment, and prevention of\\nchild abuse cases . . . . Central registry records also are used\\nto screen persons who will be entrusted with the care of\\nchildren,” such as “individuals [who] apply[] to be foster or\\nadoptive parents or child or youth care providers.”\\nEstablishment and Maintenance of Central Registrations for\\nChild Abuse or Neglect Reports at 2, Child Welfare\\nInformation Gateway: U.S. Dep’t of Health & Human Serv.,\\nChildren’s Bureau (2018).\\n\\n At times relevant to this suit, § 350-2 further provided\\nthat reports maintained in the central registry “shall [be]\\npromptly expunge[d] . . . if: (1) The department has found the\\n\\n6 BIRD V. STATE OF HAWAI‘I\\n\\nreports to be unsubstantiated; or (2) The petition arising from\\nthe report has been dismissed by order of the family court\\nafter an adjudicatory hearing on the merits.” Act of May 18,\\n2017, sec. 3, § 350-2(d), 2017 Haw. Sess. Laws 60, 61–62.1\\nThe statute stated that “a report is unsubstantiated only when\\nthe department has found the allegations to be frivolous or to\\nhave been made in bad faith.” Id. The statute provided, and\\ncontinues to provide, however, no procedures by which DHS\\nshould investigate, record, or expunge reports in the CCAR.\\nSuch procedures are left to “the department’s rules.” Haw.\\nRev. Stat. § 350-2(a).\\n\\n The Hawai‘i Administrative Rules require DHS to record\\naccused individuals in the CCAR whenever it receives a\\nreport of serious child abuse. Haw. Admin. R. § 17-1610-18.\\nDHS is simultaneously required to commence an\\ninvestigation. Id. At the conclusion of its investigation, DHS\\nwill declare the report “confirmed,” “not confirmed,” or\\n“unsubstantiated.” Id. § 17-1610-19(a)(1)–(3). Only a report\\ndeclared “unsubstantiated” will be expunged from the CCAR.\\nId. § 17-1610-19(a)(1). In cases where the report is declared\\n“confirmed” or “not confirmed,” the individuals subject to the\\nallegations remain on the CCAR. See id. § 17-1610-19(a)(2),\\n(4).2 The regulations specify that DHS “shall comply with\\n\\n\\n 1\\n Haw. Rev. Stat. § 350-2(d) was amended effective May 29, 2017,\\nso that “not confirmed” reports of child abuse are also expunged from the\\nregistry. Act of May 18, 2017, sec. 3, § 350-2(d), 2017 Haw. Sess. Laws\\n60, 61–62.\\n 2\\n We note that while the Hawai‘i statute governing the CCAR has\\nbeen amended to expunge “not confirmed” reports of child abuse, the\\nregulations have not kept pace and still require that “not confirmed”\\nreports of child abuse remain on the registry. See Haw. Admin. R. § 17-\\n1610–19.\\n\\n BIRD V. STATE OF HAWAI‘I 7\\n\\nrequests from other states to check its central registry for the\\npurpose of conducting background checks in foster or\\nadoptive placement cases.” Id. § 17-1610-19(a)(5).\\n\\n Once DHS has completed its investigation and\\ndetermined that an accused individual should remain on the\\nCCAR, state regulations require DHS to give written notice\\nto the “identified perpetrator or maltreater.” Id. § 17-1610-\\n11(c). Anyone “dissatisfied with the disposition of the\\ndepartment’s assessment or action taken by the department”\\nhas two routes for challenging his or her inclusion in the\\nCCAR. Id. § 17-1610-12(b). First, if DHS has not yet\\ncommenced any proceeding against the listed individual in\\nfamily court, he or she may file an administrative appeal. Id.\\nAt the appeal, however, the listed individual remains subject\\nto the requirements for expungement laid out in Hawai‘i\\nRevised Statutes § 350-2, which at all relevant times in this\\ncase required the listed individual to prove that the report was\\n“unsubstantiated,” in other words, that it was “frivolous or . . .\\nmade in bad faith.” Act of May 18, 2017, sec. 3, § 350-2(d),\\n2017 Haw. Sess. Laws 60, 61–62. If the listed individual was\\nunable to meet this exacting standard, he or she remained on\\nthe registry.\\n\\n Alternatively, if DHS has petitioned the family court for\\ncustody of the listed individual’s child(ren), the listed\\nindividual is not permitted to seek administrative review.\\nHaw. Admin. R. § 17-1610-12(c). Instead, the listed\\nindividual’s exclusive means by which to expunge his or her\\nlisting in the CCAR is to prevail at an adjudicatory hearing\\nbefore the family court. Haw. Rev. Stat. § 350-2(d)(2). At\\nthe adjudicatory hearing, the family court will determine\\n“[w]hether the child’s physical or psychological health or\\nwelfare has been harmed or is subject to threatened harm by\\n\\n8 BIRD V. STATE OF HAWAI‘I\\n\\nthe acts or omissions of the [listed individual].” Haw. Rev.\\nStat. § 587A-28(d)(1). Significantly, if DHS returns custody\\nof the child and settles the family court proceeding before it\\nreaches the adjudicatory phase, the statutory scheme leaves\\nlisted individuals who are wrongfully listed in the CCAR\\nwith no recourse for expunging their names. See id.\\n\\nB. Factual Background\\n\\n In 2003, Bird married Frank Fontana and two years later\\ngave birth to a daughter, T.F. Fontana, who worked for the\\nNavy, was subsequently transferred to Hawai‘i. Bird and\\ntheir daughter accompanied him to Hawai‘i to live in base\\nhousing.\\n\\n Shortly after their arrival, in February 2007, Bird gave\\nbirth to a second daughter, C.F. Over the next six weeks,\\nbetween early February and March 20, C.F. was examined by\\nover thirty healthcare personnel in multiple well-child checks,\\nand as she was treated for jaundice, wheezing, apnea induced\\nby acid reflux, and a fever. Over the course of this treatment,\\nC.F. underwent two chest x-rays, neither of which showed\\nany injuries. None of the doctors and nurses that treated C.F.\\never noted any signs of abuse or neglect.\\n\\n On March 28, Bird returned home from a dentist\\nappointment to find Fontana administering CPR to C.F. C.F.\\nwas transported by ambulance to Tripler Army Medical\\nCenter, where she died of cardiac arrest. The doctors and\\nnurses who examined C.F. noted multiple bruises on her\\nbody, evidence of a fractured rib, and evidence of possible\\nfractures in her femurs. As a result, an emergency\\ndepartment nurse at the hospital called the Child Welfare\\n\\n BIRD V. STATE OF HAWAI‘I 9\\n\\nServices branch (“CWS”) of DHS to report a case of possible\\nchild abuse.\\n\\n In accordance with its regulations, DHS listed both Bird\\nand Fontana on the CCAR and initiated an investigation. On\\nApril 2, an investigative multidisciplinary team concluded\\nthat C.F.’s death was likely due to child maltreatment, and\\nrecommended that DHS seek custody of Bird and Fontana’s\\nremaining daughter, T.F. On the basis of this report, CWS\\nconcluded on April 5 that the alleged physical abuse of C.F.\\nby her parents, Fontana and Bird, was “confirmed.” DHS\\nthen filed a Petition for Foster Custody of their remaining\\ndaughter, T.F., with the family court. DHS, however, failed\\nto give Bird notice that she was now listed on the CCAR as\\na “perpetrator or maltreater.”\\n\\n At the same time, Naval Criminal Investigative Services\\n(“NCIS”) initiated an investigation into the death of C.F. In\\nJuly 2007, Fontana “confessed [to Navy investigators] to\\nharming [C.F.] with actions that caused [her] death.” Fontana\\nwas charged with first degree murder and subsequently pled\\nguilty to the crime. In November 2007, NCIS reported to\\nDHS that “[Bird was] not a suspect” in their on-going\\ninvestigation. Nevertheless, because DHS did not conclude\\nthat its initial assessment was “frivolous or in bad faith,”\\nDHS took no action and Bird remained on the CCAR as a\\n“confirmed” child abuser.\\n\\n Meanwhile, Bird was still engaged with the family court\\nin an attempt to regain custody of T.F. “After months of\\nhome checks, supervised visits, and evaluations by various\\nmedical and social work professionals,” Bird was granted\\nphysical custody of T.F. in November 2007. In December,\\nthe family court authorized Bird to leave Hawai‘i with T.F.\\n\\n10 BIRD V. STATE OF HAWAI‘I\\n\\nand return to Tennessee where she could live with her father\\non the condition “that Tennessee DHS approves the . . .\\nplacement, and Tennessee DHS puts in writing that services\\nfor [Bird] and [T.F.] will be in place when mother and\\ndaughter arrive in Tennessee, and both are received by the\\nCourt.” In June 2008, the Hawai‘i family court terminated its\\njurisdiction and revoked the prior order of family supervision.\\nAlthough the family court entered no findings with respect to\\nBird’s alleged mistreatment of C.F., it found that “[T.F.’s]\\nfamily can provide a safe family home without the assistance\\nof a service plan.”\\n\\n Bird alleges that throughout these events, she was never\\ngiven notice, written or otherwise, of her listing on the CCAR\\nand that she “did not have any knowledge that a child abuse\\nregistry even existed.” Without that notice, Bird believed that\\nan adjudicatory trial before the family court concerned only\\nher custody of her surviving daughter, T.F. She did not know\\nthat obtaining a finding or ruling from the family court would\\nbe the only way for her to get her name removed from a\\nregistry she did not know existed. Bird had rejected an\\nadjudicatory trial before the family court, opting instead to\\nwork with DHS to drop the case. Bird contends that without\\nnotice, she did not know “that failure to adjudicate [her]\\nclaim through the Family Court would result in [her] being on\\na list of confirmed child abusers for the rest of [her] life.”\\n\\n After returning to Tennessee, Bird eventually remarried.\\nBird and her new husband decided that they wanted to adopt\\na child from Africa, “[p]referably . . . an older child who was\\nHIV positive so that [they] could provide him or her with the\\nkind of medical and family care that [he or she was] lacking.”\\nDuring the adoption process, the adoption agency conducted\\nbackground checks with each of the states where Bird had\\n\\n BIRD V. STATE OF HAWAI‘I 11\\n\\npreviously lived, and discovered in June 2012 that she was\\nlisted on Hawai‘i’s Central Child Abuse Registry. This report\\nrendered Bird ineligible to adopt.\\n\\n Bird contacted DHS in June 2012 and requested an\\nadministrative hearing before the agency.3 DHS initially\\nrefused to give Bird a hearing, explaining that jurisdiction had\\nbeen turned over to the family court five years prior in\\nconnection with the custody dispute over T.F. After further\\ncorrespondence with Bird and her attorney, DHS offered Bird\\nan administrative hearing limited to determining “whether\\nDHS properly confirmed (i) abuse of a minor and (ii) threat\\nof abuse of a minor.” Bird rejected this hearing, deeming the\\nevidentiary standard impossible to meet. In a letter dated\\nMay 14, 2013, Bird’s attorney indicated to DHS that Bird\\nwould “proceed with litigation” if she was not removed from\\nthe registry.\\n\\nC. Procedural Background\\n\\n In July 2015, Bird filed a complaint in the First Circuit\\nCourt, State of Hawai‘i, seeking both injunctive and\\nmonetary relief against the State of Hawai’i, DHS and various\\nstate officials, in both their official and individual capacities.\\nShe alleged violations of her state constitutional right to due\\nprocess and her federal constitutional rights, a claim brought\\n\\n\\n\\n\\n 3\\n Bird was given official notice that she had been placed on the\\nregistry for the first time in September 2012 when DHS mailed and\\nemailed her a “Notice to Perpetrator.”\\n\\n12 BIRD V. STATE OF HAWAI‘I\\n\\nunder 42 U.S.C. § 1983.4 The defendants timely removed the\\naction to federal court, and sought summary judgment on the\\nground that Bird’s claims fail as a matter of law. At the\\nhearing on the motion, the district court raised sua sponte the\\nissue of whether Bird’s § 1983 claim is barred by the\\napplicable statute of limitations. In response, Bird’s attorney\\nrequested leave to amend the complaint.\\n\\n After supplemental briefing, the district court granted the\\ndefendants’ motion for summary judgment, finding that\\nBird’s § 1983 claim was untimely. The court concluded that\\na two-year statute of limitations applies to Bird’s § 1983\\nclaim, and that her claim had accrued “at the very latest—by\\nMay 2013, more than two years before Bird filed her July 20,\\n2015 Complaint.” The court rejected Bird’s argument that\\nher complaint was protected by the continuing violation\\ndoctrine and denied Bird leave to amend. Having dismissed\\nBird’s § 1983 claim, the court declined to exercise\\nsupplemental jurisdiction over Bird’s remaining state law\\nclaim, and remanded the case to state court. This appeal\\nfollowed.5\\n\\n\\n\\n 4\\n Bird’s complaint does not actually specify which constitutional\\nrights the State violated. Bird’s parallel state constitutional claim,\\nhowever, focuses on a violation of due process, and the injunctive relief\\nthat Bird seeks under her § 1983 claim would require the State to “provide\\nimmediate notice to all individuals who have been reported to the\\nRegistry” and “provide procedures for previously reported individuals to\\nseek to correct false reports.” Thus, Bird’s claim is most aptly\\ncharacterized as a procedural due process claim.\\n 5\\n We review de novo the district court’s conclusion that a claim is\\nbarred by a statute of limitations. Avila v. Spokane Sch. Dist. 81, 852 F.3d\\n936, 939 (9th Cir. 2017).\\n\\n BIRD V. STATE OF HAWAI‘I 13\\n\\n II. DISCUSSION\\n\\n Because 42 U.S.C. § 1983 does not contain its own statute\\nof limitations, “[a]ctions brought pursuant to 42 U.S.C.\\n§ 1983 are governed by the forum state’s statute of\\nlimitations for personal injury actions.” Knox v. Davis, 260\\nF.3d 1009, 1012–13 (9th Cir. 2001) (citing Wilson v. Garcia,\\n471 U.S. 261, 276 (1985)). In Hawai‘i, the statute of\\nlimitations for personal injury actions is two years. See Haw.\\nRev. Stat. § 657-7.\\n\\n Here, the parties disagree over when the statute of\\nlimitations commenced to run. “Although state law\\ndetermines the length of the limitations period, federal law\\ndetermines when a civil rights claim accrues.” Knox, 260\\nF.3d at 1013 (quoting Morales v. City of Los Angeles, 214\\nF.3d 1151, 1153–54 (9th Cir. 2000)). Under federal law, the\\n“discovery rule” typically governs the accrual of § 1983\\nclaims so that “a claim accrues when the plaintiff knows or\\nhas reason to know of the injury which is the basis of the\\naction.” Id. (quoting TwoRivers v. Lewis, 174 F.3d 987, 992\\n(9th Cir. 1999)).\\n\\n Because Bird learned in June 2012 that she had been\\nplaced on the CCAR, requested and was denied a hearing that\\nfall, rejected a proffered hearing in February 2013, and then\\nsent a letter in May 2013 to DHS threatening to “proceed\\nwith litigation,” the district court concluded that Bird was\\naware that she had been placed on the registry, felt this\\nplacement was unjustified, and was contemplating her legal\\noptions “at the very latest” by May 2013.\\n\\n Bird does not seriously dispute that she had knowledge of\\nthe injury giving rise to her claims by May 2013. Instead, she\\n\\n14 BIRD V. STATE OF HAWAI‘I\\n\\nraises two arguments against applying the normal discovery\\nrule of accrual. First, Bird argues that “[w]here a plaintiff\\nchallenges the constitutionality of a statutory or regulatory\\nscheme, the plaintiff’s claim does not accrue until the\\noffending law is repealed.” Second, Bird argues that her\\ncomplaint alleges a “continuing violation”—an exception to\\nthe usual discovery rule of accrual. Because neither argument\\nprevails, we conclude that the normal discovery rule of\\naccrual applies. We agree with the district court’s conclusion\\nthat Bird knew of her injury no later than May 2013, and\\naffirm the district court’s order dismissing Bird’s complaint\\nas time-barred.\\n\\n A\\n\\n Bird first contends that a facial challenge to an allegedly\\nunconstitutional statute is exempt from the traditional rules of\\naccrual. More specifically, Bird asserts that a claim seeking\\n“injunctive relief to invalidate an [ongoing] unconstitutional\\nstatutory and regulatory scheme” does not accrue until the\\nstatute is repealed. In essence, Bird’s proposed rule would\\nmean that a person who has been harmed by an allegedly\\nunconstitutional statute can bring suit at any time so long as\\nthe statute remains in operation.\\n\\n Bird acknowledges that “[t]he Ninth Circuit has not yet\\naddressed” the issue, and relies on several out-of-circuit cases\\nto support her position. First, Bird points to Virginia Hospital\\nAss’n v. Baliles, 868 F.2d 653, 656 (4th Cir. 1989) (“VHA”),\\nwhere the plaintiffs—a nonprofit association of hospitals and\\nother healthcare providers—sought to enjoin the procedures\\nthat Virginia used to determine Medicare reimbursements as\\nviolative of the Medicaid Act, 42 U.S.C. § 1396 et seq., and\\nthe due process clause, U.S. CONST. amend. XIV. When the\\n\\n BIRD V. STATE OF HAWAI‘I 15\\n\\ndefendants argued that the plaintiffs’ claim was barred by the\\napplicable statute of limitations, the district court concluded\\nthat “VHA had alleged an ongoing constitutional violation,\\nand . . . the statute [of limitations] would not have begun to\\nrun until the violation ended.” Id. at 663. The Fourth Circuit\\nagreed, concluding that “[t]he continued enforcement of an\\nunconstitutional statute cannot be insulated by the statute of\\nlimitations.” Id. (alteration in original). Bird highlights the\\nsame language in Kuhnle Brothers, Inc. v. County of Geauga,\\n103 F.3d 516, 518, 522 (6th Cir. 1997), where the Sixth\\nCircuit quoted VHA’s assertion that “[t]he continued\\nenforcement of an unconstitutional statute cannot be insulated\\nby the statute of limitations” and concluded that a trucking\\ncompany’s challenge to a county resolution that banned\\nthrough traffic on certain roads was not time-barred.6\\n\\n Bird’s reliance on this language in VHA and Kuhnle\\nexceeds the context of both cases. In both VHA and Kuhnle,\\nthe defendants had argued that the plaintiff’s facial challenges\\nhad accrued when the statute was enacted as opposed to when\\nthe statute was enforced against the plaintiff. See Kuhnle,\\n103 F.3d at 522 (explaining that a statute “does not become\\nimmunized from legal challenge for all time merely because\\nno one challenges it within two years of its enactment”);\\nVHA, 868 F.2d at 663 (explaining that the parties agreed that\\nVHA’s cause of action accrued “when Virginia enacted its\\ncurrent reimbursement plan”). Faced with the unsavory\\nprospect of denying recovery to plaintiffs who had actually\\nbeen injured within the limitations period merely because the\\nstatute had been enacted outside the limitations period, the\\n\\n\\n 6\\n The Kuhnle court’s discussion relies primarily on the continuing\\nviolations doctrine, see 103 F.3d at 520–23, and is discussed below in Part\\nII.B.\\n\\n16 BIRD V. STATE OF HAWAI‘I\\n\\ncourts responded by allowing the suit to proceed. Thus, VHA\\nand Kuhnle did nothing more than bring the Fourth and Sixth\\nCircuits into alignment with our own view that a facial\\nchallenge to a statute generally accrues when “the statute is\\nenforced—in other words, [when] it is applied.” Levald, Inc.\\nv. City of Palm Desert, 998 F.2d 680, 688 (9th Cir. 1993).\\n\\n Although the courts used broader language in reaching\\ntheir holdings, see Kuhnle, 103 F.3d at 522; VHA, 868 F.2d\\nat 663, we do not believe the cases stand for the broad\\nproposition advocated by Bird. Bird’s argument goes far\\nbeyond the actual outcomes in both cases; here, Bird argues\\nthat her claim did not accrue at the time of her injury, but\\nrather continues ad infinitum until the statute is repealed.\\nAlthough Bird frames her argument in terms of accrual, such\\na position would essentially “nullify all statutes of limitations\\nwith respect to statutory challenges.” VHA, 868 F.2d at 663.\\n\\n Bird cites only one case that arguably supports such a\\nbroad definition of accrual. In Tearpock-Martini v. Borough\\nof Shickshinny, 756 F.3d 232, 234 (3d Cir. 2014), the plaintiff\\nbrought a § 1983 suit to enjoin a religious sign that the\\nmunicipality approved to be installed on public grounds near\\nthe plaintiff’s property. While the plaintiff conceded that a\\ntime period greater than the statutory limitations period had\\nelapsed since the sign was installed, she argued that her\\nclaims were nevertheless timely under the continuing\\nviolations doctrine. Id. at 236. The Third Circuit rejected\\napplication of the continuing violations doctrine—a holding\\ndiscussed in greater detail in Part II.B, infra—and instead\\nconsidered whether the plaintiff’s claims were entirely\\nexempt from the traditional statute of limitations. Id. at\\n237–40. The court first concluded that a federal court should\\napply the state-law statute of limitations to a § 1983 claim\\n\\n BIRD V. STATE OF HAWAI‘I 17\\n\\n“only if it is not ‘inconsistent with the Constitution and laws\\nof the United States.’” Id. at 238 (quoting Burnett v. Grattan,\\n468 U.S. 42, 47–48 (1984)). The court went on to “consider\\nthe significance of the federal rights implicated by an\\nEstablishment Clause claim” and concluded that “the\\ntraditional rationales justifying a limitations period . . . simply\\nhave no persuasive force in this context.” Id. at 238–39.\\n\\n Even if this general exemption from the statute of\\nlimitations is justified in the Establishment Clause context—a\\nquestion that we have not decided, see Maldonado v. Harris,\\n370 F.3d 945, 955–56 (9th Cir. 2004) (“[W]hether a statute of\\nlimitations for § 1983 actions can bar a facial challenge under\\nthe First Amendment to a state statute appears to be a\\nquestion that has not been conclusively resolved by any\\nCircuit court . . . but we need not resolve the . . . question\\nhere . . . .”), and which is not before us today—such a\\ndeparture from the discovery rule of accrual is not warranted\\nhere. The court in Tearpock-Martini expressly limited its\\nholding to the “federal rights implicated by an Establishment\\nClause claim.” 756 F.3d at 238. Here, Bird asserts a due\\nprocess claim and the “traditional rationales justifying a\\nlimitations period—‘to protect defendants against stale or\\nunduly delayed claims,’ ‘facilitat[e] the administration of\\nclaims,’ and ‘promot[e] judicial efficiency’”—are applicable\\nto a due process claim. See id. at 238–39 (alterations in\\noriginal). Unlike “a still-existing monument that\\ncommunicates anew an allegedly unconstitutional\\nendorsement of religion by the government each time it is\\nviewed,” see id. at 239, a due process violation is a discrete\\nevent for which evidence and memory will fade over time.\\nThe traditional interests of protecting defendants and\\nfacilitating the administration of such claims are thus\\napplicable to a due process claim and justify enforcing a\\n\\n18 BIRD V. STATE OF HAWAI‘I\\n\\nlimitations period through the discovery rule of accrual.\\nAlthough enforcing the statute of limitations here will make\\nBird’s alleged losses permanent, such a cost is present in\\nevery case where a plaintiff’s claim is denied on statute of\\nlimitations grounds and is alone insufficient to justify\\ndeparting from the normal discovery rule of accrual.\\n\\n Moreover, Bird’s assertion that applying the discovery\\nrule of accrual to her case will forever immunize an\\nunconstitutional state statute is untrue. The only way for the\\nstate to immunize itself from further suit by future plaintiffs\\nwould be to stop enforcing its statute. As long as Hawai’i\\ncontinues to enforce its statute, it is subject to a facial\\nchallenge by every individual it affects. Enforcing the statute\\nof limitations in the case of a facial challenge to an allegedly\\nunconstitutional statute, thus, does not render the statute\\nimmune from challenge; it merely requires that such\\nchallenges be brought in a timely manner. For these reasons,\\nwe decline to adopt Bird’s proposed rule exempting claims\\nseeking “injunctive relief to invalidate an [ongoing]\\nunconstitutional statutory . . . scheme” from the discovery\\nrule of accrual.\\n\\n B\\n\\n We next consider whether Bird’s claim qualifies as a\\ncontinuing violation. The continuing violations doctrine\\nfunctions as an exception to the discovery rule of accrual\\n“allowing a plaintiff to seek relief for events outside of the\\nlimitations period.” Knox, 260 F.3d at 1013. Although the\\ncontinuing violations doctrine is most frequently seen in the\\ncontext of employment discrimination suits, we have held\\nthat the continuing violations doctrine also applies to § 1983\\nclaims. See id. (“The continuing violation theory applies to\\n\\n BIRD V. STATE OF HAWAI‘I 19\\n\\n§ 1983 actions.”); Cherosky v. Henderson, 330 F.3d 1243,\\n1246 n.3 (9th Cir. 2003) (“[T]he Supreme Court’s analysis of\\nthe continuing violations doctrine is not limited to Title VII\\nactions. It applies with equal force to . . . actions arising\\nunder other civil rights laws.”).\\n\\n “The doctrine of continuing violations . . . is actually a\\nconglomeration of several different ideas,” Williams v.\\nOwens-Illinois, Inc., 665 F.2d 918, 924 (9th Cir. 1982)\\n(internal quotation omitted), the essence of which is that\\n“when a defendant’s conduct is part of a continuing practice,\\nan action is timely so long as the last act evidencing the\\ncontinuing practice falls within the limitations period,”\\nTearpock-Martini, 756 F.3d at 236 (quoting Brenner v. Local\\n514, United Bhd. of Carpenters & Joiners of Am., 927 F.2d\\n1283, 1295 (3d Cir. 1991)). We have recognized two\\napplications of the continuing violations doctrine: first, to “a\\nseries of related acts, one or more of which falls within the\\nlimitations period,” and second, to “the maintenance of a\\ndiscriminatory system both before and during [the\\nlimitations] period.” Gutowsky v. County of Placer, 108 F.3d\\n256, 259 (9th Cir. 1997) (quoting Green v. L.A. Cty.\\nSuperintendent of Schs., 883 F.2d 1472, 1480 (9th Cir.\\n1989)).\\n\\n Prior to 2002, we applied the serial acts branch of the\\ncontinuing violations doctrine when the defendant committed\\na series of acts directed against a single plaintiff. See Morgan\\nv. Nat’l R.R. Passenger Corp., 232 F.3d 1008 (9th Cir. 2000)\\n(“Morgan I”); Kyle Graham, The Continuing Violations\\nDoctrine, 43 GONZ. L. REV. 271, 304 (2008). As long as the\\nplaintiff could show that acts preceding the limitations period\\nwere “sufficiently related” to acts that occurred within the\\nlimitations period, the plaintiff could recover damages for all\\n\\n20 BIRD V. STATE OF HAWAI‘I\\n\\nof the related acts—even those that would otherwise be\\nbarred by the statute of limitations. See Morgan I, 232 F.3d\\nat 1016. In 2002, however, the Supreme Court reversed our\\ndecision in Morgan I and limited the serial acts branch of the\\ncontinuing violations doctrine. Nat’l R.R. Passenger Corp.\\nv. Morgan, 536 U.S. 101, 114 (2002) (“Morgan II”). The\\nCourt held that “discrete . . . acts are not actionable if time\\nbarred, even when they are related to acts alleged in timely\\nfiled charges” because “[e]ach discrete . . . act starts a new\\nclock for filing charges alleging that act.” Id. at 113.7\\n\\n Although Morgan II applied Title VII of the Civil Rights\\nAct of 1964, we have relied on Morgan II to abrogate the\\nserial acts branch of the continuing violations doctrine for\\n§ 1983 claims as well. See Carpinteria Valley Farms, Ltd. v.\\nCounty of Santa Barbara, 344 F.3d 822, 829 (9th Cir. 2003)\\n(“Although Morgan was a Title VII case . . . we have applied\\nMorgan to bar § 1983 claims predicated on discrete time-\\nbarred acts, not-withstanding that those acts are related to\\ntimely-filed claims.”); Cherosky, 330 F.3d at 1246 n.3\\n(concluding that the Supreme Court’s holding in Morgan II\\n“applies with equal force . . . to actions arising under other\\ncivil rights laws”).\\n\\n Unlike the serial acts branch, the systematic branch does\\nnot require the plaintiff to allege specific acts that occurred\\nwithin the statute of limitations period. Originally, the\\nsystematic branch allowed the plaintiff to recover for acts that\\noccurred prior to the limitations period as long as (1) those\\n\\n\\n 7\\n The Court carved out one exception for hostile work environment\\nclaims, because “[s]uch claims are based on the cumulative effect of\\nindividual acts” that “cannot be said to occur on any particular day.”\\nMorgan II, 536 U.S. at 115–17.\\n\\n BIRD V. STATE OF HAWAI‘I 21\\n\\nacts were conducted pursuant to a policy or practice that\\nremained in effect within the statute of limitations period and\\n(2) the plaintiff remained subject or susceptible to the policy\\nwithin the limitations period. See Gutowsky, 108 F.3d at 260\\n(applying the systematic branch of the continuing violations\\ndoctrine where the plaintiff’s “specific examples” of\\ndiscrimination occurred prior to the limitations period, but\\nwhere the “widespread policy and practices of discrimination\\nof which [the plaintiff] complain[ed] continued every day of\\nher employment, including days that fall within the limitation\\nperiod”); Graham, 43 GONZ. L. REV. at 303–04. The theory\\nbehind the systematic branch was that every day the plaintiff\\nwas subject to the policy or practice “constituted a new\\nviolation” sufficient to extend the statute of limitations\\nperiod. See Reed v. Lockheed Aircraft Corp., 613 F.2d 757,\\n759–60 (9th Cir. 1980) (“[E]ach day without promotion\\nconstituted a new violation of Title VII . . . .”).\\n\\n Although the Supreme Court in Morgan II addressed only\\nthe serial acts branch of the continuing violations doctrine,\\nsee 536 U.S. at 115 & n.9, we have applied Morgan II to\\nabrogate the systematic branch of the continuing violations\\ndoctrine as well. For example, in Lyons v. England, 307 F.3d\\n1092, 1107 (9th Cir. 2002), we reasoned from Morgan II that\\na plaintiff’s “assertion that [a] series of discrete acts flows\\nfrom a company-wide, or systematic, discriminatory practice\\nwill not succeed in establishing the employer’s liability for\\nacts occurring outside the limitations period.” Similarly, in\\nCherosky, the plaintiffs sought to avoid Morgan II’s\\nunfavorable holding by arguing under the systematic branch\\nof the continuing violations doctrine; the plaintiffs contended\\nthat they could still recover damages for acts of employment\\ndiscrimination that had occurred prior to the statute of\\nlimitations period as long as those acts had been conducted\\n\\n22 BIRD V. STATE OF HAWAI‘I\\n\\npursuant to a discriminatory company policy. 330 F.3d at\\n1246. We rejected that argument, concluding instead that\\n“[t]he allegation that . . . discrete acts were undertaken\\npursuant to a discriminatory policy does not extend the\\nstatutory limitations period.” 330 F.3d at 1247.\\n\\n Thus, after Morgan II, little remains of the continuing\\nviolations doctrine. Except for a limited exception for hostile\\nwork environment claims—not at issue here—the serial acts\\nbranch is virtually non-existent. Moreover, while we have\\nleft room for the systematic branch to apply to class-wide\\npattern-or-practice claims, see Lyons, 307 F.3d at 1107 n.8\\n(emphasizing that its holding did not “suggest that after\\nMorgan the same plaintiff would be precluded from bringing\\na class-wide pattern-or-practice claim”), we have consistently\\nrefused to apply the systematic branch to rescue\\nindividualized claims that are otherwise time-barred.\\n\\n In this light, the district court’s conclusion that Bird is\\ntime-barred from seeking damages for her placement on the\\nCCAR in 2007 is correct. Because “the violation Bird alleges\\nis the placement of her name on the Registry without\\nconstitutionally required due process,” she has brought only\\nan individualized claim.8 As such, the systematic branch of\\nthe continuing violations doctrine is inapplicable here, and\\nthe discovery rule of accrual applies.\\n\\n Bird raises several arguments to the contrary. First, she\\nlikens her case to that of Kuhnle, where the Seventh Circuit\\n\\n\\n 8\\n Bird did style her complaint as a class action, but “there has been no\\nclass certification granted, or even sought” and “the parties have identified\\nno potential class member plaintiffs.” At oral argument, Bird’s counsel\\nconceded that Bird had not pursued her class action claim.\\n\\n BIRD V. STATE OF HAWAI‘I 23\\n\\nheld that a county resolution which barred the plaintiff from\\nusing certain roads for truck travel “deprived [the plaintiff] of\\nliberty interests . . . created by a fundamental constitutional\\nright to intrastate travel . . . every day that it remained in\\neffect” and thus constituted a continuing violation. 103 F.3d\\nat 521–22. Bird argues that “[s]imilar to the plaintiff in\\nKuhnle . . . [her] § 1983 claim arises from a continued\\ndeprivation of liberty interests caused by an unconstitutional\\nrule or statute.” This argument is, in essence, an attempt to\\nrecast Bird’s claim as a substantive due process claim for\\ndeprivation of liberty interests in working with and adopting\\nchildren.\\n\\n Bird’s case, however, is distinguishable from that of\\nKuhnle. In Kuhnle, the resolution directly forbade the\\nplaintiff from driving on certain roads, thereby depriving him\\nevery day of his “fundamental constitutional right to\\nintrastate travel.” 103 F.3d at 521. By comparison, the\\ndeprivation of liberty that Bird continues to suffer is best\\nunderstood as the “continuing impact from [a] past\\nviolation[].” Williams, 665 F.2d at 924 (quoting Reed,\\n613 F.2d at 760); see Del. State Coll. v. Ricks, 449 U.S. 250,\\n252–54, 258 (1980) (holding that a professor’s eventual\\ntermination was merely the continuing effect of the original\\ndecision to deny tenure and not a new violation); Knox,\\n260 F.3d at 1013 (holding that repeated denials of prison\\nvisitation rights did not constitute a continuing due process\\nviolation because each of the subsequent denials merely\\nimplemented the original suspension and did not represent an\\nindependent violation). Although Bird’s injuries resulting\\nfrom the 2007 act of placing her on the CCAR continue to the\\npresent day, continuing effect is insufficient to constitute a\\ncontinuing violation. See Williams, 665 F.2d at 924\\n\\n24 BIRD V. STATE OF HAWAI‘I\\n\\n(“[C]ontinuing impact from past violations is not actionable.”\\n(quoting Reed, 613 F.2d at 670)).\\n\\n Second, Bird argues that her claim is not confined to the\\ndenial of pre-deprivation process; rather, Bird believes that\\nthe continual lack of a post-deprivation process by which she\\ncan challenge her placement on the CCAR constitutes an\\nongoing deprivation of due process. Bird is unable to\\nidentify, however, any specific instance within the limitations\\nperiod where she sought and was denied post-deprivation\\nprocess. See Pouncil v. Tilton, 704 F.3d 568, 581 (9th Cir.\\n2012) (holding that constitutional and statutory claims for\\nreligious discrimination were not barred by the statute of\\nlimitations where the defendant allegedly committed an\\nindependently wrongful, discrete act within the limitations\\nperiod, even if the discrete act was related to a preexisting\\npolicy of which the plaintiff was aware and subject to outside\\nthe limitations period). The only instance Bird has identified\\nwhere she sought and was denied a post-deprivation hearing\\noccurred prior to May 2013.\\n\\n ***\\n\\n Because Bird has alleged only individualized claims for\\ndeprivation of procedural due process, the normal discovery\\nrule of accrual applies. Based on the correspondence between\\nBird and DHS, we agree with the district court that Bird had\\nknowledge of the injury giving rise to her claims by May\\n2013. Accordingly, her suit, filed in July 2015, was outside\\nthe two-year statute of limitations. Because “the complaint\\nwould not [have been] saved by any amendment,” Carvalho\\nv. Equifax Info. Servs., LLC, 629 F.3d 876, 893 (9th Cir.\\n2010) (quoting Leadsinger, Inc. v. BMG Music Publ’g,\\n\\n BIRD V. STATE OF HAWAI‘I 25\\n\\n512 F.3d 522, 532 (9th Cir. 2008)), the district court did not\\nerr in denying Bird leave to amend.\\n\\n III. CONCLUSION\\n\\n We AFFIRM the judgment of the district court.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432771/\", \"author_raw\": \"PER CURIAM\"}, {\"author\": \"BYBEE, Circuit Judge, concurring\", \"type\": \"concurrence\", \"text\": \"BYBEE, Circuit Judge, concurring:\\n\\n I fully concur in the per curiam opinion of the court that\\nMs. Bird’s complaint is barred by the applicable statute of\\nlimitations. I regret that result, however, and write separately\\nbecause I fear that a great injustice may have been done. Ms.\\nBird alleges that the State of Hawai‘i has violated her\\nFourteenth Amendment right to procedural due process when,\\nwithout giving her notice and an opportunity for a hearing,\\nthe state placed her on its Central Child Abuse Registry\\n(“CCAR”). Were we permitted to reach the merits of her\\nargument, Ms. Bird has assuredly stated a plausible due\\nprocess claim. Regrettably, if Ms. Bird was erroneously\\nlisted on the CCAR, she may suffer the effects of the state’s\\nerror for the remainder of her life.\\n\\n We analyze procedural due process claims in two familiar\\nsteps: we first ask “whether there exists a liberty or property\\ninterest which has been interfered with by the State.” Ky.\\nDep’t of Corr. v. Thompson, 490 U.S. 454, 460 (1989). We\\nthen ask “whether the procedures attendant upon that\\ndeprivation were constitutionally sufficient.” Id.\\n\\n Unfortunately, this is not a case of first impression. We\\npreviously addressed the adequacy of state procedures for\\n\\n26 BIRD V. STATE OF HAWAI‘I\\n\\nlisting and maintaining names on child abuse registries in\\nHumphries v. County of Los Angeles, 554 F.3d 1170 (9th Cir.\\n2009), overruled on other grounds by Los Angeles County v.\\nHumphries, 562 U.S. 29 (2010). There, we addressed a\\nchallenge to California’s Child Abuse Central Index\\n(“CACI”) and held at step one of the due process inquiry that\\nan individual’s liberty interest has been “interfered with by\\nthe State” where “a state statute creates both a stigma and a\\ntangible burden on an individual’s ability to obtain a right or\\nstatus recognized by state law.” 554 F.3d at 1185, 1188. We\\nrecognized that “being labeled a child abuser by being placed\\non [a child abuse registry] is ‘unquestionably stigmatizing.’”\\nId. at 1186. Moreover, “a tangible burden . . . exists where\\nthe plaintiff can show that, as a practical matter, the law\\ncreates a framework under which agencies reflexively check\\nthe stigmatizing listing—whether by internal regulation or\\ncustom—prior to conferring a legal right or benefit.” Id. at\\n1188. In this case, there is no question that by listing Ms.\\nBird on its CCAR—irrespective of whether she is\\n“confirmed” or “not confirmed”—Hawai‘i has placed a\\nstigmatizing label on Ms. Bird. Indeed, there is perhaps no\\nname more deserving of our opprobrium than to be called a\\nchild abuser—or, as Hawai‘i euphemistically refers to them:\\na “perpetrator” or “maltreater.” Haw. Admin. R. § 17-1610-\\n11(c).\\n\\n There are real consequences for those who find\\nthemselves on a state registry. State-created child abuse\\nregistries form an organic network of accusations from which\\nconsequences flow: those listed may be denied the privilege\\nof teaching or working with children, adopting, fostering, and\\ncoaching youth sports or other activities. See Wright v.\\nO’Day, 706 F.3d 769, 771 (6th Cir. 2013) (describing how\\nTennessee law prohibits people listed on the state’s child\\n\\n BIRD V. STATE OF HAWAI‘I 27\\n\\nabuse registry from working in child-care agencies, child-care\\nprograms, and adult-daycare centers); Humphries, 554 F.3d\\nat 1177–78 (describing consequences of being listed on\\nCalifornia’s CACI); Behrens v. Regier, 422 F.3d 1255, 1257\\n(11th Cir. 2005) (describing how plaintiff was unable to adopt\\nanother child after inaccurate listing on Florida’s child abuse\\nregistry); Dupuy v. Samuels, 397 F.3d 493, 497–98 (7th Cir.\\n2005) (describing how Illinois law requires licensed facilities\\nin childcare to check the state’s child abuse registry).\\n\\n When a state places legal disabilities on its citizens, we\\nask “whether the procedures attendant upon that deprivation\\nwere constitutionally sufficient.” See Thompson, 490 U.S. at\\n460. The answer here is a resounding “NO.” Hawai‘i offers\\nlittle process—pre-listing or post-listing—for persons it lists\\non its CCAR. In the same provision that creates the CCAR,\\nHawai‘i requires DHS to notify police and prosecutors of any\\nreports DHS receives. Haw. Rev. Stat. § 350-2(a), (b).\\nAlthough nothing in the statute requires DHS to notify the\\nindividual who has been branded a perpetrator or maltreater,\\nDHS’s regulations require it to give notice of the listing and\\nthe right to an administrative appeal. Haw. Admin. R. § 17-\\n1610-11(c). In this case, however, DHS did not follow its\\nregulations; Ms. Bird received no notice. She had no way of\\nknowing that she was on the registry until years later, after\\nher ex-husband had been convicted of killing their daughter,\\nHawai‘i family court restored her older daughter to her\\ncustody, she had left Hawai‘i for her family’s home in\\nTennessee, she remarried, and she and her new husband\\nsough to adopt a special-needs child from a third-world\\ncountry. By the time Ms. Bird learned that she was on\\nHawai‘i’s CCAR, the damage had been done. She was\\ndisqualified by Tennessee from adoption (and probably many\\nother things she hadn’t figured out yet), and her only recourse\\n\\n28 BIRD V. STATE OF HAWAI‘I\\n\\nwas to go back to Hawai‘i five years after the incident and try\\nto persuade an agency that didn’t want to deal with her to\\ngive her a hearing. In the end, DHS begrudgingly offered her\\n“an administrative hearing for the sole purpose of deciding\\nwhether DHS properly confirmed (i) abuse of a minor and\\n(ii) threat of abuse of a minor.”\\n\\n We have seen this bad movie before. In Humphries, we\\nconcluded that California’s procedures for listing and\\nmaintaining individuals on the CACI were constitutionally\\ndeficient because they created a high likelihood that innocent\\nnames would remain on the registry. We first emphasized\\nthat the statutory standard for inclusion on the CACI was “a\\nvery low threshold” that created a high likelihood of “false\\npositives.” 554 F.3d at 1195. We observed that California\\nlaw essentially “reverse[d] . . . the presumption of innocence\\nin our criminal justice system: the accused [was] presumed to\\nbe a child abuser and listed [on the registry] unless the\\ninvestigator [affirmatively] determine[d] that the report [was]\\nfalse, improbable, or accidental.” Id. The danger of an\\nerroneous listing was further compounded by the lack of any\\nstatutory procedure by which an individual could challenge\\nhis or her inclusion on the CACI. Id. at 1195–1200. Rather\\nthan decide for ourselves what procedures California should\\nadopt, we ordered California to “provide ‘some kind of\\nhearing’ by which [an accused individual] can challenge his\\ninclusion.” Id. at 1201.\\n\\n Hawai‘i’s procedures for creating and maintaining its\\nregistry perpetuate some of the same problems we identified\\nin Humphries. First, Hawai‘i regulations require that even\\n“not confirmed” reports of child abuse be listed on the\\nCCAR, thereby creating a high likelihood of “false positives.”\\n\\n BIRD V. STATE OF HAWAI‘I 29\\n\\nHaw. Admin. R. § 17-1610-19(a)(1)(A).1 The danger of\\nerroneous listings in the CCAR is magnified by inadequate\\nprocedures by which an individual can challenge his or her\\ninclusion in the CCAR. Unlike the statutory scheme at issue\\nin Humphries, Hawai‘i regulations do require notice and do\\nallow for administrative review of CCAR listings in some\\ncircumstances. At least one of the two means for review\\noccurs on DHS’s initiative and can be terminated on DHS’s\\ninitiative. See Haw. Rev. Stat. § 350-2(d)(1). If DHS\\npetitions for a separate family court proceeding in which it\\nclaims custody of the listed individual’s child, the listed\\nindividual is not permitted to seek administrative review of\\nhis or her registry listing. See Haw. Admin. R. § 17-1610-12.\\nInstead, the listed individual must prevail at an adjudicatory\\nhearing before the family court can expunge the report. Haw.\\nRev. Stat. § 350-2(d)(2).\\n\\n At first blush, this statutory scheme appears more\\nprotective than the scheme at issue in Humphries, where\\nlisted individuals were offered neither administrative review\\nnor a court hearing capable of expunging the listing. Ms.\\nBird’s unfortunate situation, however, highlights a glaring\\nhole in Hawai‘i’s regulations: In cases where DHS chooses\\nto return a child to his or her parents rather than pursue the\\nfamily court proceeding to the adjudicatory phase, listed\\nindividuals are denied access to both an administrative\\nhearing and a court adjudication. See Haw. Rev. Stat. § 350-\\n\\n 1\\n I recognize that on May 29, 2017, the statute governing listings on\\nthe Registry, Haw. Rev. Stat. § 350-2(d), was amended so that “not\\nconfirmed” reports of child abuse may be expunged from the CCAR. See\\n2017 Haw. Laws Act 16 (H.B. 1099) (May 18, 2017). As our per curiam\\nopinion notes, the regulations governing listings in the CCAR, however,\\nstill require that “not confirmed” reports remain on the CCAR. Haw.\\nAdmin. R. § 17-1610-19; see Panel Op. at 6 n.2.\\n\\n30 BIRD V. STATE OF HAWAI‘I\\n\\n2(d); Haw. Admin. R. § 17-1610-12(c). In other words, under\\nHawai‘i procedures, DHS can both return a child to the\\nparents and keep the parent’s names on the CCAR. In that\\ncircumstance, the only way the parent can get a hearing on\\nthe CCAR listing is to decline the opportunity to get the child\\nback and pursue litigation in the family court to a full\\nadjudication. This makes no sense. The parents who settle\\nwith DHS and thus forego adjudication get their child back,\\nbut may be barred forever from teaching school or coaching\\nLittle League. See, e.g., Humphries, 554 F.3d at 1183 & n.9.\\nPersons such as Ms. Bird have no recourse for expunging\\ntheir names from the CCAR, except to file a civil rights\\naction under 42 U.S.C. § 1983.\\n\\n The state suggests that the hole in its regulations, by\\nwhich certain individuals are offered neither administrative\\nreview nor a court adjudication, is immaterial because parents\\nwho are listed on the CCAR should know better than to\\nregain custody of their children from DHS without enduring\\na court adjudication. According to the State, the onus lies on\\nthe parents to insist that the district court proceed to an\\nadjudication where it would otherwise be unnecessary.\\n\\n The flaw in the state’s procedures is compounded in this\\ncase by the fact that it never told Ms. Bird she was listed on\\nthe CCAR. The state’s position that it was, nevertheless, Ms.\\nBird’s burden to insist on an adjudication is just astounding.\\nEven worse, the State argues that the lack of notice in this\\ncase is also immaterial because Ms. Bird was represented by\\ncounsel in the dependency proceeding before the family court\\nand thus should have “just known” that she was listed on the\\nCCAR and that the only way for her to get off was to insist on\\nproceeding to an adjudicatory hearing in spite of the fact that\\n\\n BIRD V. STATE OF HAWAI‘I 31\\n\\nDHS had effectively mooted the family court proceeding by\\nagreeing that Bird’s daughter could be safely returned home.\\n\\n I don’t understand how the State of Hawai‘i can maintain\\nsuch arguments with a straight face. Ms. Bird was plainly\\ndenied due process of law. And, if the facts are as she alleges\\nthem, not only was she denied notice and an opportunity to\\ncontest her placement on the CCAR, she was erroneously\\nlisted in the first place. The state has twice wronged her:\\nfirst, by listing her at all; second, by denying her an\\nopportunity to prove that she didn’t deserve to be blacklisted.\\n\\n Hawai‘i prevails here only because of a statute of\\nlimitations. Statutes of limitations are important to the search\\nfor truth. They “‘promote justice by preventing surprises\\nthrough [plaintiffs’] revival of claims that have been allowed\\nto slumber until evidence has been lost, memories have faded,\\nand witnesses have disappeared.’” CTS Corp. v. Waldburger,\\n573 U.S. 1, 8 (2014) (alteration in original) (quoting R.R.\\nTelegraphers v. Ry. Express Agency, Inc., 321 U.S. 342,\\n348–49 (1944)). The state gets the benefit of that rule today,\\nbecause Ms. Bird filed her suit just outside the two-year\\nstatute of limitations. The irony is that while Ms. Bird filed\\nher suit just two months late, she didn’t even learn of the\\nCCAR—and her cause of action—until five years after the\\nstate listed her. Hawai‘i will not be compelled by law—at\\nleast not by us—to give Ms. Bird an opportunity to show she\\ndoesn’t deserve to be on its CCAR, but it can always choose\\nto do the right thing voluntarily. In any event, DHS should\\nnot interpret our decision today as in any way condoning the\\nblatantly insufficient procedures by which it maintains Ms.\\nBird’s name in its Central Child Abuse Registry.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432771/\", \"author_raw\": \"BYBEE, Circuit Judge, concurring\"}]}"},"judge1":{"kind":"string","value":"RICHARD C TALLMAN"},"judge2":{"kind":"string","value":"JAY S BYBEE"},"judge3":{"kind":"string","value":"N RANDY SMITH"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":1,"string":"1"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4655518/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Courtney Bird v. State of Hawaii\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-23\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: Yes\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCOURTNEY BIRD, No. 17-16076\n Plaintiff-Appellant,\n D.C. No.\n v. 1:15-cv-00304-\n DKW-KJM\nSTATE OF HAWAI‘I; DEPARTMENT OF\nHUMAN SERVICES; DHS, SOCIAL\nSERVICES DIVISION, CHILD OPINION\nWELFARE SERVICES BRANCH;\nPANKAJ BHANOT, Department of\nHuman Services Director; JEFFREY\nR. WOODLAND; DOES,\n Defendants-Appellees.\n\n\n Appeal from the United States District Court\n for the District of Hawai‘i\n Derrick Kahala Watson, District Judge, Presiding\n\n Argued and Submitted February 12, 2019\n Submission Vacated February 26, 2019\n Resubmitted August 23, 2019\n University of Hawai‘i Manoa\n\n Filed August 23, 2019\n\n Before: Richard C. Tallman, Jay S. Bybee,\n and N. Randy Smith, Circuit Judges.\n\n2 BIRD V. STATE OF HAWAI‘I\n\n Per Curiam Opinion;\n Concurrence by Judge Bybee\n\n\n SUMMARY*\n\n\n Civil Rights\n\n The panel affirmed the district court’s dismissal, on\nstatute of limitations grounds, of a complaint brought\npursuant to 42 U.S.C. § 1983 alleging that the Hawai‘i\nDepartment of Human Services violated plaintiff’s right to\ndue process when it listed her, without notice, on the State’s\nChild Protective Services Central Registry.\n\n Plaintiff and her then-husband were placed on the State’s\nRegistry after their 7-week-old baby died of cardiac arrest in\n2007. Bird’s husband later confessed to killing the infant and\na criminal investigation concluded that plaintiff was not a\nsuspect. The Department of Health Services did not,\nhowever, remove plaintiff’s name from the Registry, nor did\nit notify her that her name was listed. In 2013, plaintiff\ndiscovered, during a background check, that her name was on\nthe Registry. Plaintiff’s attorney communicated with the\nDepartment and threatened to sue when plaintiff’s name was\nnot removed, but did not file suit until more than two years\nlater.\n\n Applying Hawai‘i’s two-year statute of limitations, the\npanel held that plaintiff’s complaint was not subject to any\n\n *\n This summary constitutes no part of the opinion of the court. It has\nbeen prepared by court staff for the convenience of the reader.\n\n BIRD V. STATE OF HAWAI‘I 3\n\nexceptions from the normal discovery rule of accrual. The\npanel held that plaintiff had knowledge of the injury giving\nrise to her claim by May 2013 when she threatened to sue the\nDepartment if she was not removed from the Registry.\nAccordingly, her suit, filed in July 2015, was time-barred.\n\n The panel rejected plaintiff’s assertion that a claim\nseeking injunctive relief to invalidate an ongoing\nunconstitutional statutory and regulatory scheme does not\naccrue until the statute is repealed. The panel held that the\ntraditional interests of protecting defendants and facilitating\nthe administration of claims was applicable to a due process\nclaim and justified enforcing a limitations period through the\ndiscovery rule of accrual. The panel further rejected\nplaintiff’s claim that her complaint alleged a continuing\nviolation. The panel held that because the violation plaintiff\nalleged was the placement of her name on the Registry\nwithout constitutionally required due process, she had\nbrought only an individualized claim. As such, the\nsystematic branch of the continuing violations doctrine was\ninapplicable, and the discovery rule of accrual applied.\n\n Concurring in the per curiam opinion, Judge Bybee stated\nthat plaintiff had assuredly stated a plausible due process\nclaim and that the Department of Health Services should not\ninterpret the panel’s decision, which was based on statute of\nlimitations grounds, as in any way condoning the blatantly\ninsufficient procedures by which the Department maintains\nplaintiff’s name in its Central Child Abuse Registry.\n\n4 BIRD V. STATE OF HAWAI‘I\n\n COUNSEL\n\nMargery S. Bronster (argued), Robert M. Hatch, and Kelly A.\nHiga, Bronster Fujichaku Robbins, Honolulu, Hawai‘i; John\nY. U. Choi, Honolulu, Hawai‘i; for Plaintiff-Appellant.\n\nRyan M. Akamine (argued) and Caron M. Inagaki, Deputy\nAttorneys General; Department of the Attorney General,\nHonolulu, Hawai‘i; for Defendants-Appellees.\n\n\n OPINION\n\nPER CURIAM:\n\n On March 28, 2007, plaintiff-appellant Courtney Bird\nreturned home from a dentist appointment to find her then-\nhusband administering CPR to their seven-week-old baby,\nwho later died at the hospital of cardiac arrest. As a result of\nthe infant’s death, the Hawai‘i Department of Human\nServices (“DHS”) listed both Bird and her husband on the\nstate’s Central Child Abuse Registry (“CCAR”). Bird’s\nhusband later confessed to killing their infant baby and the\ncriminal investigation concluded that Bird was not a suspect.\nDHS, however, did not remove Bird’s name from the CCAR.\nThroughout this time, DHS never told Bird that she was listed\non the registry.\n\n Bird did not learn that she had been listed on the CCAR\nuntil June 2012—more than five years later—when the report\nturned up on a background check. When a request for a\nhearing and various communications with DHS failed to\nresult in Bird’s removal from the CCAR, Bird’s attorney\nthreatened DHS in May 2013 with suit in federal court. Bird\n\n BIRD V. STATE OF HAWAI‘I 5\n\ndid not file suit, however, until July 20, 2015—more than two\nyears after her extensive communications with DHS and\nprevious threat to sue. The district court determined that\nBird’s claims brought under 42 U.S.C. § 1983 were untimely\nand granted summary judgment in favor of the State.\n\n Because Bird’s complaint is not subject to any exception\nfrom the normal discovery rule of accrual, we agree that\nBird’s claim accrued no later than May 2013, when Bird\nthreatened to sue DHS if she was not removed from the\nCCAR. We affirm.\n\n I. FACTS AND PROCEEDINGS BELOW\n\nA. Statutory Background\n\n Like most states, Hawai‘i law requires DHS to maintain\na central registry for reports of child abuse or neglect. See\nHaw. Rev. Stat. § 350-2(d) (“The department shall maintain\na central registry of reported child abuse or neglect cases\n. . . .”). “Central registry reports are typically used to aid\nagencies in the investigation, treatment, and prevention of\nchild abuse cases . . . . Central registry records also are used\nto screen persons who will be entrusted with the care of\nchildren,” such as “individuals [who] apply[] to be foster or\nadoptive parents or child or youth care providers.”\nEstablishment and Maintenance of Central Registrations for\nChild Abuse or Neglect Reports at 2, Child Welfare\nInformation Gateway: U.S. Dep’t of Health & Human Serv.,\nChildren’s Bureau (2018).\n\n At times relevant to this suit, § 350-2 further provided\nthat reports maintained in the central registry “shall [be]\npromptly expunge[d] . . . if: (1) The department has found the\n\n6 BIRD V. STATE OF HAWAI‘I\n\nreports to be unsubstantiated; or (2) The petition arising from\nthe report has been dismissed by order of the family court\nafter an adjudicatory hearing on the merits.” Act of May 18,\n2017, sec. 3, § 350-2(d), 2017 Haw. Sess. Laws 60, 61–62.1\nThe statute stated that “a report is unsubstantiated only when\nthe department has found the allegations to be frivolous or to\nhave been made in bad faith.” Id. The statute provided, and\ncontinues to provide, however, no procedures by which DHS\nshould investigate, record, or expunge reports in the CCAR.\nSuch procedures are left to “the department’s rules.” Haw.\nRev. Stat. § 350-2(a).\n\n The Hawai‘i Administrative Rules require DHS to record\naccused individuals in the CCAR whenever it receives a\nreport of serious child abuse. Haw. Admin. R. § 17-1610-18.\nDHS is simultaneously required to commence an\ninvestigation. Id. At the conclusion of its investigation, DHS\nwill declare the report “confirmed,” “not confirmed,” or\n“unsubstantiated.” Id. § 17-1610-19(a)(1)–(3). Only a report\ndeclared “unsubstantiated” will be expunged from the CCAR.\nId. § 17-1610-19(a)(1). In cases where the report is declared\n“confirmed” or “not confirmed,” the individuals subject to the\nallegations remain on the CCAR. See id. § 17-1610-19(a)(2),\n(4).2 The regulations specify that DHS “shall comply with\n\n\n 1\n Haw. Rev. Stat. § 350-2(d) was amended effective May 29, 2017,\nso that “not confirmed” reports of child abuse are also expunged from the\nregistry. Act of May 18, 2017, sec. 3, § 350-2(d), 2017 Haw. Sess. Laws\n60, 61–62.\n 2\n We note that while the Hawai‘i statute governing the CCAR has\nbeen amended to expunge “not confirmed” reports of child abuse, the\nregulations have not kept pace and still require that “not confirmed”\nreports of child abuse remain on the registry. See Haw. Admin. R. § 17-\n1610–19.\n\n BIRD V. STATE OF HAWAI‘I 7\n\nrequests from other states to check its central registry for the\npurpose of conducting background checks in foster or\nadoptive placement cases.” Id. § 17-1610-19(a)(5).\n\n Once DHS has completed its investigation and\ndetermined that an accused individual should remain on the\nCCAR, state regulations require DHS to give written notice\nto the “identified perpetrator or maltreater.” Id. § 17-1610-\n11(c). Anyone “dissatisfied with the disposition of the\ndepartment’s assessment or action taken by the department”\nhas two routes for challenging his or her inclusion in the\nCCAR. Id. § 17-1610-12(b). First, if DHS has not yet\ncommenced any proceeding against the listed individual in\nfamily court, he or she may file an administrative appeal. Id.\nAt the appeal, however, the listed individual remains subject\nto the requirements for expungement laid out in Hawai‘i\nRevised Statutes § 350-2, which at all relevant times in this\ncase required the listed individual to prove that the report was\n“unsubstantiated,” in other words, that it was “frivolous or . . .\nmade in bad faith.” Act of May 18, 2017, sec. 3, § 350-2(d),\n2017 Haw. Sess. Laws 60, 61–62. If the listed individual was\nunable to meet this exacting standard, he or she remained on\nthe registry.\n\n Alternatively, if DHS has petitioned the family court for\ncustody of the listed individual’s child(ren), the listed\nindividual is not permitted to seek administrative review.\nHaw. Admin. R. § 17-1610-12(c). Instead, the listed\nindividual’s exclusive means by which to expunge his or her\nlisting in the CCAR is to prevail at an adjudicatory hearing\nbefore the family court. Haw. Rev. Stat. § 350-2(d)(2). At\nthe adjudicatory hearing, the family court will determine\n“[w]hether the child’s physical or psychological health or\nwelfare has been harmed or is subject to threatened harm by\n\n8 BIRD V. STATE OF HAWAI‘I\n\nthe acts or omissions of the [listed individual].” Haw. Rev.\nStat. § 587A-28(d)(1). Significantly, if DHS returns custody\nof the child and settles the family court proceeding before it\nreaches the adjudicatory phase, the statutory scheme leaves\nlisted individuals who are wrongfully listed in the CCAR\nwith no recourse for expunging their names. See id.\n\nB. Factual Background\n\n In 2003, Bird married Frank Fontana and two years later\ngave birth to a daughter, T.F. Fontana, who worked for the\nNavy, was subsequently transferred to Hawai‘i. Bird and\ntheir daughter accompanied him to Hawai‘i to live in base\nhousing.\n\n Shortly after their arrival, in February 2007, Bird gave\nbirth to a second daughter, C.F. Over the next six weeks,\nbetween early February and March 20, C.F. was examined by\nover thirty healthcare personnel in multiple well-child checks,\nand as she was treated for jaundice, wheezing, apnea induced\nby acid reflux, and a fever. Over the course of this treatment,\nC.F. underwent two chest x-rays, neither of which showed\nany injuries. None of the doctors and nurses that treated C.F.\never noted any signs of abuse or neglect.\n\n On March 28, Bird returned home from a dentist\nappointment to find Fontana administering CPR to C.F. C.F.\nwas transported by ambulance to Tripler Army Medical\nCenter, where she died of cardiac arrest. The doctors and\nnurses who examined C.F. noted multiple bruises on her\nbody, evidence of a fractured rib, and evidence of possible\nfractures in her femurs. As a result, an emergency\ndepartment nurse at the hospital called the Child Welfare\n\n BIRD V. STATE OF HAWAI‘I 9\n\nServices branch (“CWS”) of DHS to report a case of possible\nchild abuse.\n\n In accordance with its regulations, DHS listed both Bird\nand Fontana on the CCAR and initiated an investigation. On\nApril 2, an investigative multidisciplinary team concluded\nthat C.F.’s death was likely due to child maltreatment, and\nrecommended that DHS seek custody of Bird and Fontana’s\nremaining daughter, T.F. On the basis of this report, CWS\nconcluded on April 5 that the alleged physical abuse of C.F.\nby her parents, Fontana and Bird, was “confirmed.” DHS\nthen filed a Petition for Foster Custody of their remaining\ndaughter, T.F., with the family court. DHS, however, failed\nto give Bird notice that she was now listed on the CCAR as\na “perpetrator or maltreater.”\n\n At the same time, Naval Criminal Investigative Services\n(“NCIS”) initiated an investigation into the death of C.F. In\nJuly 2007, Fontana “confessed [to Navy investigators] to\nharming [C.F.] with actions that caused [her] death.” Fontana\nwas charged with first degree murder and subsequently pled\nguilty to the crime. In November 2007, NCIS reported to\nDHS that “[Bird was] not a suspect” in their on-going\ninvestigation. Nevertheless, because DHS did not conclude\nthat its initial assessment was “frivolous or in bad faith,”\nDHS took no action and Bird remained on the CCAR as a\n“confirmed” child abuser.\n\n Meanwhile, Bird was still engaged with the family court\nin an attempt to regain custody of T.F. “After months of\nhome checks, supervised visits, and evaluations by various\nmedical and social work professionals,” Bird was granted\nphysical custody of T.F. in November 2007. In December,\nthe family court authorized Bird to leave Hawai‘i with T.F.\n\n10 BIRD V. STATE OF HAWAI‘I\n\nand return to Tennessee where she could live with her father\non the condition “that Tennessee DHS approves the . . .\nplacement, and Tennessee DHS puts in writing that services\nfor [Bird] and [T.F.] will be in place when mother and\ndaughter arrive in Tennessee, and both are received by the\nCourt.” In June 2008, the Hawai‘i family court terminated its\njurisdiction and revoked the prior order of family supervision.\nAlthough the family court entered no findings with respect to\nBird’s alleged mistreatment of C.F., it found that “[T.F.’s]\nfamily can provide a safe family home without the assistance\nof a service plan.”\n\n Bird alleges that throughout these events, she was never\ngiven notice, written or otherwise, of her listing on the CCAR\nand that she “did not have any knowledge that a child abuse\nregistry even existed.” Without that notice, Bird believed that\nan adjudicatory trial before the family court concerned only\nher custody of her surviving daughter, T.F. She did not know\nthat obtaining a finding or ruling from the family court would\nbe the only way for her to get her name removed from a\nregistry she did not know existed. Bird had rejected an\nadjudicatory trial before the family court, opting instead to\nwork with DHS to drop the case. Bird contends that without\nnotice, she did not know “that failure to adjudicate [her]\nclaim through the Family Court would result in [her] being on\na list of confirmed child abusers for the rest of [her] life.”\n\n After returning to Tennessee, Bird eventually remarried.\nBird and her new husband decided that they wanted to adopt\na child from Africa, “[p]referably . . . an older child who was\nHIV positive so that [they] could provide him or her with the\nkind of medical and family care that [he or she was] lacking.”\nDuring the adoption process, the adoption agency conducted\nbackground checks with each of the states where Bird had\n\n BIRD V. STATE OF HAWAI‘I 11\n\npreviously lived, and discovered in June 2012 that she was\nlisted on Hawai‘i’s Central Child Abuse Registry. This report\nrendered Bird ineligible to adopt.\n\n Bird contacted DHS in June 2012 and requested an\nadministrative hearing before the agency.3 DHS initially\nrefused to give Bird a hearing, explaining that jurisdiction had\nbeen turned over to the family court five years prior in\nconnection with the custody dispute over T.F. After further\ncorrespondence with Bird and her attorney, DHS offered Bird\nan administrative hearing limited to determining “whether\nDHS properly confirmed (i) abuse of a minor and (ii) threat\nof abuse of a minor.” Bird rejected this hearing, deeming the\nevidentiary standard impossible to meet. In a letter dated\nMay 14, 2013, Bird’s attorney indicated to DHS that Bird\nwould “proceed with litigation” if she was not removed from\nthe registry.\n\nC. Procedural Background\n\n In July 2015, Bird filed a complaint in the First Circuit\nCourt, State of Hawai‘i, seeking both injunctive and\nmonetary relief against the State of Hawai’i, DHS and various\nstate officials, in both their official and individual capacities.\nShe alleged violations of her state constitutional right to due\nprocess and her federal constitutional rights, a claim brought\n\n\n\n\n 3\n Bird was given official notice that she had been placed on the\nregistry for the first time in September 2012 when DHS mailed and\nemailed her a “Notice to Perpetrator.”\n\n12 BIRD V. STATE OF HAWAI‘I\n\nunder 42 U.S.C. § 1983.4 The defendants timely removed the\naction to federal court, and sought summary judgment on the\nground that Bird’s claims fail as a matter of law. At the\nhearing on the motion, the district court raised sua sponte the\nissue of whether Bird’s § 1983 claim is barred by the\napplicable statute of limitations. In response, Bird’s attorney\nrequested leave to amend the complaint.\n\n After supplemental briefing, the district court granted the\ndefendants’ motion for summary judgment, finding that\nBird’s § 1983 claim was untimely. The court concluded that\na two-year statute of limitations applies to Bird’s § 1983\nclaim, and that her claim had accrued “at the very latest—by\nMay 2013, more than two years before Bird filed her July 20,\n2015 Complaint.” The court rejected Bird’s argument that\nher complaint was protected by the continuing violation\ndoctrine and denied Bird leave to amend. Having dismissed\nBird’s § 1983 claim, the court declined to exercise\nsupplemental jurisdiction over Bird’s remaining state law\nclaim, and remanded the case to state court. This appeal\nfollowed.5\n\n\n\n 4\n Bird’s complaint does not actually specify which constitutional\nrights the State violated. Bird’s parallel state constitutional claim,\nhowever, focuses on a violation of due process, and the injunctive relief\nthat Bird seeks under her § 1983 claim would require the State to “provide\nimmediate notice to all individuals who have been reported to the\nRegistry” and “provide procedures for previously reported individuals to\nseek to correct false reports.” Thus, Bird’s claim is most aptly\ncharacterized as a procedural due process claim.\n 5\n We review de novo the district court’s conclusion that a claim is\nbarred by a statute of limitations. Avila v. Spokane Sch. Dist. 81, 852 F.3d\n936, 939 (9th Cir. 2017).\n\n BIRD V. STATE OF HAWAI‘I 13\n\n II. DISCUSSION\n\n Because 42 U.S.C. § 1983 does not contain its own statute\nof limitations, “[a]ctions brought pursuant to 42 U.S.C.\n§ 1983 are governed by the forum state’s statute of\nlimitations for personal injury actions.” Knox v. Davis, 260\nF.3d 1009, 1012–13 (9th Cir. 2001) (citing Wilson v. Garcia,\n471 U.S. 261, 276 (1985)). In Hawai‘i, the statute of\nlimitations for personal injury actions is two years. See Haw.\nRev. Stat. § 657-7.\n\n Here, the parties disagree over when the statute of\nlimitations commenced to run. “Although state law\ndetermines the length of the limitations period, federal law\ndetermines when a civil rights claim accrues.” Knox, 260\nF.3d at 1013 (quoting Morales v. City of Los Angeles, 214\nF.3d 1151, 1153–54 (9th Cir. 2000)). Under federal law, the\n“discovery rule” typically governs the accrual of § 1983\nclaims so that “a claim accrues when the plaintiff knows or\nhas reason to know of the injury which is the basis of the\naction.” Id. (quoting TwoRivers v. Lewis, 174 F.3d 987, 992\n(9th Cir. 1999)).\n\n Because Bird learned in June 2012 that she had been\nplaced on the CCAR, requested and was denied a hearing that\nfall, rejected a proffered hearing in February 2013, and then\nsent a letter in May 2013 to DHS threatening to “proceed\nwith litigation,” the district court concluded that Bird was\naware that she had been placed on the registry, felt this\nplacement was unjustified, and was contemplating her legal\noptions “at the very latest” by May 2013.\n\n Bird does not seriously dispute that she had knowledge of\nthe injury giving rise to her claims by May 2013. Instead, she\n\n14 BIRD V. STATE OF HAWAI‘I\n\nraises two arguments against applying the normal discovery\nrule of accrual. First, Bird argues that “[w]here a plaintiff\nchallenges the constitutionality of a statutory or regulatory\nscheme, the plaintiff’s claim does not accrue until the\noffending law is repealed.” Second, Bird argues that her\ncomplaint alleges a “continuing violation”—an exception to\nthe usual discovery rule of accrual. Because neither argument\nprevails, we conclude that the normal discovery rule of\naccrual applies. We agree with the district court’s conclusion\nthat Bird knew of her injury no later than May 2013, and\naffirm the district court’s order dismissing Bird’s complaint\nas time-barred.\n\n A\n\n Bird first contends that a facial challenge to an allegedly\nunconstitutional statute is exempt from the traditional rules of\naccrual. More specifically, Bird asserts that a claim seeking\n“injunctive relief to invalidate an [ongoing] unconstitutional\nstatutory and regulatory scheme” does not accrue until the\nstatute is repealed. In essence, Bird’s proposed rule would\nmean that a person who has been harmed by an allegedly\nunconstitutional statute can bring suit at any time so long as\nthe statute remains in operation.\n\n Bird acknowledges that “[t]he Ninth Circuit has not yet\naddressed” the issue, and relies on several out-of-circuit cases\nto support her position. First, Bird points to Virginia Hospital\nAss’n v. Baliles, 868 F.2d 653, 656 (4th Cir. 1989) (“VHA”),\nwhere the plaintiffs—a nonprofit association of hospitals and\nother healthcare providers—sought to enjoin the procedures\nthat Virginia used to determine Medicare reimbursements as\nviolative of the Medicaid Act, 42 U.S.C. § 1396 et seq., and\nthe due process clause, U.S. CONST. amend. XIV. When the\n\n BIRD V. STATE OF HAWAI‘I 15\n\ndefendants argued that the plaintiffs’ claim was barred by the\napplicable statute of limitations, the district court concluded\nthat “VHA had alleged an ongoing constitutional violation,\nand . . . the statute [of limitations] would not have begun to\nrun until the violation ended.” Id. at 663. The Fourth Circuit\nagreed, concluding that “[t]he continued enforcement of an\nunconstitutional statute cannot be insulated by the statute of\nlimitations.” Id. (alteration in original). Bird highlights the\nsame language in Kuhnle Brothers, Inc. v. County of Geauga,\n103 F.3d 516, 518, 522 (6th Cir. 1997), where the Sixth\nCircuit quoted VHA’s assertion that “[t]he continued\nenforcement of an unconstitutional statute cannot be insulated\nby the statute of limitations” and concluded that a trucking\ncompany’s challenge to a county resolution that banned\nthrough traffic on certain roads was not time-barred.6\n\n Bird’s reliance on this language in VHA and Kuhnle\nexceeds the context of both cases. In both VHA and Kuhnle,\nthe defendants had argued that the plaintiff’s facial challenges\nhad accrued when the statute was enacted as opposed to when\nthe statute was enforced against the plaintiff. See Kuhnle,\n103 F.3d at 522 (explaining that a statute “does not become\nimmunized from legal challenge for all time merely because\nno one challenges it within two years of its enactment”);\nVHA, 868 F.2d at 663 (explaining that the parties agreed that\nVHA’s cause of action accrued “when Virginia enacted its\ncurrent reimbursement plan”). Faced with the unsavory\nprospect of denying recovery to plaintiffs who had actually\nbeen injured within the limitations period merely because the\nstatute had been enacted outside the limitations period, the\n\n\n 6\n The Kuhnle court’s discussion relies primarily on the continuing\nviolations doctrine, see 103 F.3d at 520–23, and is discussed below in Part\nII.B.\n\n16 BIRD V. STATE OF HAWAI‘I\n\ncourts responded by allowing the suit to proceed. Thus, VHA\nand Kuhnle did nothing more than bring the Fourth and Sixth\nCircuits into alignment with our own view that a facial\nchallenge to a statute generally accrues when “the statute is\nenforced—in other words, [when] it is applied.” Levald, Inc.\nv. City of Palm Desert, 998 F.2d 680, 688 (9th Cir. 1993).\n\n Although the courts used broader language in reaching\ntheir holdings, see Kuhnle, 103 F.3d at 522; VHA, 868 F.2d\nat 663, we do not believe the cases stand for the broad\nproposition advocated by Bird. Bird’s argument goes far\nbeyond the actual outcomes in both cases; here, Bird argues\nthat her claim did not accrue at the time of her injury, but\nrather continues ad infinitum until the statute is repealed.\nAlthough Bird frames her argument in terms of accrual, such\na position would essentially “nullify all statutes of limitations\nwith respect to statutory challenges.” VHA, 868 F.2d at 663.\n\n Bird cites only one case that arguably supports such a\nbroad definition of accrual. In Tearpock-Martini v. Borough\nof Shickshinny, 756 F.3d 232, 234 (3d Cir. 2014), the plaintiff\nbrought a § 1983 suit to enjoin a religious sign that the\nmunicipality approved to be installed on public grounds near\nthe plaintiff’s property. While the plaintiff conceded that a\ntime period greater than the statutory limitations period had\nelapsed since the sign was installed, she argued that her\nclaims were nevertheless timely under the continuing\nviolations doctrine. Id. at 236. The Third Circuit rejected\napplication of the continuing violations doctrine—a holding\ndiscussed in greater detail in Part II.B, infra—and instead\nconsidered whether the plaintiff’s claims were entirely\nexempt from the traditional statute of limitations. Id. at\n237–40. The court first concluded that a federal court should\napply the state-law statute of limitations to a § 1983 claim\n\n BIRD V. STATE OF HAWAI‘I 17\n\n“only if it is not ‘inconsistent with the Constitution and laws\nof the United States.’” Id. at 238 (quoting Burnett v. Grattan,\n468 U.S. 42, 47–48 (1984)). The court went on to “consider\nthe significance of the federal rights implicated by an\nEstablishment Clause claim” and concluded that “the\ntraditional rationales justifying a limitations period . . . simply\nhave no persuasive force in this context.” Id. at 238–39.\n\n Even if this general exemption from the statute of\nlimitations is justified in the Establishment Clause context—a\nquestion that we have not decided, see Maldonado v. Harris,\n370 F.3d 945, 955–56 (9th Cir. 2004) (“[W]hether a statute of\nlimitations for § 1983 actions can bar a facial challenge under\nthe First Amendment to a state statute appears to be a\nquestion that has not been conclusively resolved by any\nCircuit court . . . but we need not resolve the . . . question\nhere . . . .”), and which is not before us today—such a\ndeparture from the discovery rule of accrual is not warranted\nhere. The court in Tearpock-Martini expressly limited its\nholding to the “federal rights implicated by an Establishment\nClause claim.” 756 F.3d at 238. Here, Bird asserts a due\nprocess claim and the “traditional rationales justifying a\nlimitations period—‘to protect defendants against stale or\nunduly delayed claims,’ ‘facilitat[e] the administration of\nclaims,’ and ‘promot[e] judicial efficiency’”—are applicable\nto a due process claim. See id. at 238–39 (alterations in\noriginal). Unlike “a still-existing monument that\ncommunicates anew an allegedly unconstitutional\nendorsement of religion by the government each time it is\nviewed,” see id. at 239, a due process violation is a discrete\nevent for which evidence and memory will fade over time.\nThe traditional interests of protecting defendants and\nfacilitating the administration of such claims are thus\napplicable to a due process claim and justify enforcing a\n\n18 BIRD V. STATE OF HAWAI‘I\n\nlimitations period through the discovery rule of accrual.\nAlthough enforcing the statute of limitations here will make\nBird’s alleged losses permanent, such a cost is present in\nevery case where a plaintiff’s claim is denied on statute of\nlimitations grounds and is alone insufficient to justify\ndeparting from the normal discovery rule of accrual.\n\n Moreover, Bird’s assertion that applying the discovery\nrule of accrual to her case will forever immunize an\nunconstitutional state statute is untrue. The only way for the\nstate to immunize itself from further suit by future plaintiffs\nwould be to stop enforcing its statute. As long as Hawai’i\ncontinues to enforce its statute, it is subject to a facial\nchallenge by every individual it affects. Enforcing the statute\nof limitations in the case of a facial challenge to an allegedly\nunconstitutional statute, thus, does not render the statute\nimmune from challenge; it merely requires that such\nchallenges be brought in a timely manner. For these reasons,\nwe decline to adopt Bird’s proposed rule exempting claims\nseeking “injunctive relief to invalidate an [ongoing]\nunconstitutional statutory . . . scheme” from the discovery\nrule of accrual.\n\n B\n\n We next consider whether Bird’s claim qualifies as a\ncontinuing violation. The continuing violations doctrine\nfunctions as an exception to the discovery rule of accrual\n“allowing a plaintiff to seek relief for events outside of the\nlimitations period.” Knox, 260 F.3d at 1013. Although the\ncontinuing violations doctrine is most frequently seen in the\ncontext of employment discrimination suits, we have held\nthat the continuing violations doctrine also applies to § 1983\nclaims. See id. (“The continuing violation theory applies to\n\n BIRD V. STATE OF HAWAI‘I 19\n\n§ 1983 actions.”); Cherosky v. Henderson, 330 F.3d 1243,\n1246 n.3 (9th Cir. 2003) (“[T]he Supreme Court’s analysis of\nthe continuing violations doctrine is not limited to Title VII\nactions. It applies with equal force to . . . actions arising\nunder other civil rights laws.”).\n\n “The doctrine of continuing violations . . . is actually a\nconglomeration of several different ideas,” Williams v.\nOwens-Illinois, Inc., 665 F.2d 918, 924 (9th Cir. 1982)\n(internal quotation omitted), the essence of which is that\n“when a defendant’s conduct is part of a continuing practice,\nan action is timely so long as the last act evidencing the\ncontinuing practice falls within the limitations period,”\nTearpock-Martini, 756 F.3d at 236 (quoting Brenner v. Local\n514, United Bhd. of Carpenters & Joiners of Am., 927 F.2d\n1283, 1295 (3d Cir. 1991)). We have recognized two\napplications of the continuing violations doctrine: first, to “a\nseries of related acts, one or more of which falls within the\nlimitations period,” and second, to “the maintenance of a\ndiscriminatory system both before and during [the\nlimitations] period.” Gutowsky v. County of Placer, 108 F.3d\n256, 259 (9th Cir. 1997) (quoting Green v. L.A. Cty.\nSuperintendent of Schs., 883 F.2d 1472, 1480 (9th Cir.\n1989)).\n\n Prior to 2002, we applied the serial acts branch of the\ncontinuing violations doctrine when the defendant committed\na series of acts directed against a single plaintiff. See Morgan\nv. Nat’l R.R. Passenger Corp., 232 F.3d 1008 (9th Cir. 2000)\n(“Morgan I”); Kyle Graham, The Continuing Violations\nDoctrine, 43 GONZ. L. REV. 271, 304 (2008). As long as the\nplaintiff could show that acts preceding the limitations period\nwere “sufficiently related” to acts that occurred within the\nlimitations period, the plaintiff could recover damages for all\n\n20 BIRD V. STATE OF HAWAI‘I\n\nof the related acts—even those that would otherwise be\nbarred by the statute of limitations. See Morgan I, 232 F.3d\nat 1016. In 2002, however, the Supreme Court reversed our\ndecision in Morgan I and limited the serial acts branch of the\ncontinuing violations doctrine. Nat’l R.R. Passenger Corp.\nv. Morgan, 536 U.S. 101, 114 (2002) (“Morgan II”). The\nCourt held that “discrete . . . acts are not actionable if time\nbarred, even when they are related to acts alleged in timely\nfiled charges” because “[e]ach discrete . . . act starts a new\nclock for filing charges alleging that act.” Id. at 113.7\n\n Although Morgan II applied Title VII of the Civil Rights\nAct of 1964, we have relied on Morgan II to abrogate the\nserial acts branch of the continuing violations doctrine for\n§ 1983 claims as well. See Carpinteria Valley Farms, Ltd. v.\nCounty of Santa Barbara, 344 F.3d 822, 829 (9th Cir. 2003)\n(“Although Morgan was a Title VII case . . . we have applied\nMorgan to bar § 1983 claims predicated on discrete time-\nbarred acts, not-withstanding that those acts are related to\ntimely-filed claims.”); Cherosky, 330 F.3d at 1246 n.3\n(concluding that the Supreme Court’s holding in Morgan II\n“applies with equal force . . . to actions arising under other\ncivil rights laws”).\n\n Unlike the serial acts branch, the systematic branch does\nnot require the plaintiff to allege specific acts that occurred\nwithin the statute of limitations period. Originally, the\nsystematic branch allowed the plaintiff to recover for acts that\noccurred prior to the limitations period as long as (1) those\n\n\n 7\n The Court carved out one exception for hostile work environment\nclaims, because “[s]uch claims are based on the cumulative effect of\nindividual acts” that “cannot be said to occur on any particular day.”\nMorgan II, 536 U.S. at 115–17.\n\n BIRD V. STATE OF HAWAI‘I 21\n\nacts were conducted pursuant to a policy or practice that\nremained in effect within the statute of limitations period and\n(2) the plaintiff remained subject or susceptible to the policy\nwithin the limitations period. See Gutowsky, 108 F.3d at 260\n(applying the systematic branch of the continuing violations\ndoctrine where the plaintiff’s “specific examples” of\ndiscrimination occurred prior to the limitations period, but\nwhere the “widespread policy and practices of discrimination\nof which [the plaintiff] complain[ed] continued every day of\nher employment, including days that fall within the limitation\nperiod”); Graham, 43 GONZ. L. REV. at 303–04. The theory\nbehind the systematic branch was that every day the plaintiff\nwas subject to the policy or practice “constituted a new\nviolation” sufficient to extend the statute of limitations\nperiod. See Reed v. Lockheed Aircraft Corp., 613 F.2d 757,\n759–60 (9th Cir. 1980) (“[E]ach day without promotion\nconstituted a new violation of Title VII . . . .”).\n\n Although the Supreme Court in Morgan II addressed only\nthe serial acts branch of the continuing violations doctrine,\nsee 536 U.S. at 115 & n.9, we have applied Morgan II to\nabrogate the systematic branch of the continuing violations\ndoctrine as well. For example, in Lyons v. England, 307 F.3d\n1092, 1107 (9th Cir. 2002), we reasoned from Morgan II that\na plaintiff’s “assertion that [a] series of discrete acts flows\nfrom a company-wide, or systematic, discriminatory practice\nwill not succeed in establishing the employer’s liability for\nacts occurring outside the limitations period.” Similarly, in\nCherosky, the plaintiffs sought to avoid Morgan II’s\nunfavorable holding by arguing under the systematic branch\nof the continuing violations doctrine; the plaintiffs contended\nthat they could still recover damages for acts of employment\ndiscrimination that had occurred prior to the statute of\nlimitations period as long as those acts had been conducted\n\n22 BIRD V. STATE OF HAWAI‘I\n\npursuant to a discriminatory company policy. 330 F.3d at\n1246. We rejected that argument, concluding instead that\n“[t]he allegation that . . . discrete acts were undertaken\npursuant to a discriminatory policy does not extend the\nstatutory limitations period.” 330 F.3d at 1247.\n\n Thus, after Morgan II, little remains of the continuing\nviolations doctrine. Except for a limited exception for hostile\nwork environment claims—not at issue here—the serial acts\nbranch is virtually non-existent. Moreover, while we have\nleft room for the systematic branch to apply to class-wide\npattern-or-practice claims, see Lyons, 307 F.3d at 1107 n.8\n(emphasizing that its holding did not “suggest that after\nMorgan the same plaintiff would be precluded from bringing\na class-wide pattern-or-practice claim”), we have consistently\nrefused to apply the systematic branch to rescue\nindividualized claims that are otherwise time-barred.\n\n In this light, the district court’s conclusion that Bird is\ntime-barred from seeking damages for her placement on the\nCCAR in 2007 is correct. Because “the violation Bird alleges\nis the placement of her name on the Registry without\nconstitutionally required due process,” she has brought only\nan individualized claim.8 As such, the systematic branch of\nthe continuing violations doctrine is inapplicable here, and\nthe discovery rule of accrual applies.\n\n Bird raises several arguments to the contrary. First, she\nlikens her case to that of Kuhnle, where the Seventh Circuit\n\n\n 8\n Bird did style her complaint as a class action, but “there has been no\nclass certification granted, or even sought” and “the parties have identified\nno potential class member plaintiffs.” At oral argument, Bird’s counsel\nconceded that Bird had not pursued her class action claim.\n\n BIRD V. STATE OF HAWAI‘I 23\n\nheld that a county resolution which barred the plaintiff from\nusing certain roads for truck travel “deprived [the plaintiff] of\nliberty interests . . . created by a fundamental constitutional\nright to intrastate travel . . . every day that it remained in\neffect” and thus constituted a continuing violation. 103 F.3d\nat 521–22. Bird argues that “[s]imilar to the plaintiff in\nKuhnle . . . [her] § 1983 claim arises from a continued\ndeprivation of liberty interests caused by an unconstitutional\nrule or statute.” This argument is, in essence, an attempt to\nrecast Bird’s claim as a substantive due process claim for\ndeprivation of liberty interests in working with and adopting\nchildren.\n\n Bird’s case, however, is distinguishable from that of\nKuhnle. In Kuhnle, the resolution directly forbade the\nplaintiff from driving on certain roads, thereby depriving him\nevery day of his “fundamental constitutional right to\nintrastate travel.” 103 F.3d at 521. By comparison, the\ndeprivation of liberty that Bird continues to suffer is best\nunderstood as the “continuing impact from [a] past\nviolation[].” Williams, 665 F.2d at 924 (quoting Reed,\n613 F.2d at 760); see Del. State Coll. v. Ricks, 449 U.S. 250,\n252–54, 258 (1980) (holding that a professor’s eventual\ntermination was merely the continuing effect of the original\ndecision to deny tenure and not a new violation); Knox,\n260 F.3d at 1013 (holding that repeated denials of prison\nvisitation rights did not constitute a continuing due process\nviolation because each of the subsequent denials merely\nimplemented the original suspension and did not represent an\nindependent violation). Although Bird’s injuries resulting\nfrom the 2007 act of placing her on the CCAR continue to the\npresent day, continuing effect is insufficient to constitute a\ncontinuing violation. See Williams, 665 F.2d at 924\n\n24 BIRD V. STATE OF HAWAI‘I\n\n(“[C]ontinuing impact from past violations is not actionable.”\n(quoting Reed, 613 F.2d at 670)).\n\n Second, Bird argues that her claim is not confined to the\ndenial of pre-deprivation process; rather, Bird believes that\nthe continual lack of a post-deprivation process by which she\ncan challenge her placement on the CCAR constitutes an\nongoing deprivation of due process. Bird is unable to\nidentify, however, any specific instance within the limitations\nperiod where she sought and was denied post-deprivation\nprocess. See Pouncil v. Tilton, 704 F.3d 568, 581 (9th Cir.\n2012) (holding that constitutional and statutory claims for\nreligious discrimination were not barred by the statute of\nlimitations where the defendant allegedly committed an\nindependently wrongful, discrete act within the limitations\nperiod, even if the discrete act was related to a preexisting\npolicy of which the plaintiff was aware and subject to outside\nthe limitations period). The only instance Bird has identified\nwhere she sought and was denied a post-deprivation hearing\noccurred prior to May 2013.\n\n ***\n\n Because Bird has alleged only individualized claims for\ndeprivation of procedural due process, the normal discovery\nrule of accrual applies. Based on the correspondence between\nBird and DHS, we agree with the district court that Bird had\nknowledge of the injury giving rise to her claims by May\n2013. Accordingly, her suit, filed in July 2015, was outside\nthe two-year statute of limitations. Because “the complaint\nwould not [have been] saved by any amendment,” Carvalho\nv. Equifax Info. Servs., LLC, 629 F.3d 876, 893 (9th Cir.\n2010) (quoting Leadsinger, Inc. v. BMG Music Publ’g,\n\n BIRD V. STATE OF HAWAI‘I 25\n\n512 F.3d 522, 532 (9th Cir. 2008)), the district court did not\nerr in denying Bird leave to amend.\n\n III. CONCLUSION\n\n We AFFIRM the judgment of the district court.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Courtney Bird v. State of Hawaii\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-23\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: Yes\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nCOURTNEY BIRD, No. 17-16076\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 1:15-cv-00304-\\n DKW-KJM\\nSTATE OF HAWAI‘I; DEPARTMENT OF\\nHUMAN SERVICES; DHS, SOCIAL\\nSERVICES DIVISION, CHILD OPINION\\nWELFARE SERVICES BRANCH;\\nPANKAJ BHANOT, Department of\\nHuman Services Director; JEFFREY\\nR. WOODLAND; DOES,\\n Defendants-Appellees.\\n\\n\\n Appeal from the United States District Court\\n for the District of Hawai‘i\\n Derrick Kahala Watson, District Judge, Presiding\\n\\n Argued and Submitted February 12, 2019\\n Submission Vacated February 26, 2019\\n Resubmitted August 23, 2019\\n University of Hawai‘i Manoa\\n\\n Filed August 23, 2019\\n\\n Before: Richard C. Tallman, Jay S. Bybee,\\n and N. Randy Smith, Circuit Judges.\\n\\n2 BIRD V. STATE OF HAWAI‘I\\n\\n Per Curiam Opinion;\\n Concurrence by Judge Bybee\\n\\n\\n SUMMARY*\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s dismissal, on\\nstatute of limitations grounds, of a complaint brought\\npursuant to 42 U.S.C. § 1983 alleging that the Hawai‘i\\nDepartment of Human Services violated plaintiff’s right to\\ndue process when it listed her, without notice, on the State’s\\nChild Protective Services Central Registry.\\n\\n Plaintiff and her then-husband were placed on the State’s\\nRegistry after their 7-week-old baby died of cardiac arrest in\\n2007. Bird’s husband later confessed to killing the infant and\\na criminal investigation concluded that plaintiff was not a\\nsuspect. The Department of Health Services did not,\\nhowever, remove plaintiff’s name from the Registry, nor did\\nit notify her that her name was listed. In 2013, plaintiff\\ndiscovered, during a background check, that her name was on\\nthe Registry. Plaintiff’s attorney communicated with the\\nDepartment and threatened to sue when plaintiff’s name was\\nnot removed, but did not file suit until more than two years\\nlater.\\n\\n Applying Hawai‘i’s two-year statute of limitations, the\\npanel held that plaintiff’s complaint was not subject to any\\n\\n *\\n This summary constitutes no part of the opinion of the court. It has\\nbeen prepared by court staff for the convenience of the reader.\\n\\n BIRD V. STATE OF HAWAI‘I 3\\n\\nexceptions from the normal discovery rule of accrual. The\\npanel held that plaintiff had knowledge of the injury giving\\nrise to her claim by May 2013 when she threatened to sue the\\nDepartment if she was not removed from the Registry.\\nAccordingly, her suit, filed in July 2015, was time-barred.\\n\\n The panel rejected plaintiff’s assertion that a claim\\nseeking injunctive relief to invalidate an ongoing\\nunconstitutional statutory and regulatory scheme does not\\naccrue until the statute is repealed. The panel held that the\\ntraditional interests of protecting defendants and facilitating\\nthe administration of claims was applicable to a due process\\nclaim and justified enforcing a limitations period through the\\ndiscovery rule of accrual. The panel further rejected\\nplaintiff’s claim that her complaint alleged a continuing\\nviolation. The panel held that because the violation plaintiff\\nalleged was the placement of her name on the Registry\\nwithout constitutionally required due process, she had\\nbrought only an individualized claim. As such, the\\nsystematic branch of the continuing violations doctrine was\\ninapplicable, and the discovery rule of accrual applied.\\n\\n Concurring in the per curiam opinion, Judge Bybee stated\\nthat plaintiff had assuredly stated a plausible due process\\nclaim and that the Department of Health Services should not\\ninterpret the panel’s decision, which was based on statute of\\nlimitations grounds, as in any way condoning the blatantly\\ninsufficient procedures by which the Department maintains\\nplaintiff’s name in its Central Child Abuse Registry.\\n\\n4 BIRD V. STATE OF HAWAI‘I\\n\\n COUNSEL\\n\\nMargery S. Bronster (argued), Robert M. Hatch, and Kelly A.\\nHiga, Bronster Fujichaku Robbins, Honolulu, Hawai‘i; John\\nY. U. Choi, Honolulu, Hawai‘i; for Plaintiff-Appellant.\\n\\nRyan M. Akamine (argued) and Caron M. Inagaki, Deputy\\nAttorneys General; Department of the Attorney General,\\nHonolulu, Hawai‘i; for Defendants-Appellees.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n On March 28, 2007, plaintiff-appellant Courtney Bird\\nreturned home from a dentist appointment to find her then-\\nhusband administering CPR to their seven-week-old baby,\\nwho later died at the hospital of cardiac arrest. As a result of\\nthe infant’s death, the Hawai‘i Department of Human\\nServices (“DHS”) listed both Bird and her husband on the\\nstate’s Central Child Abuse Registry (“CCAR”). Bird’s\\nhusband later confessed to killing their infant baby and the\\ncriminal investigation concluded that Bird was not a suspect.\\nDHS, however, did not remove Bird’s name from the CCAR.\\nThroughout this time, DHS never told Bird that she was listed\\non the registry.\\n\\n Bird did not learn that she had been listed on the CCAR\\nuntil June 2012—more than five years later—when the report\\nturned up on a background check. When a request for a\\nhearing and various communications with DHS failed to\\nresult in Bird’s removal from the CCAR, Bird’s attorney\\nthreatened DHS in May 2013 with suit in federal court. Bird\\n\\n BIRD V. STATE OF HAWAI‘I 5\\n\\ndid not file suit, however, until July 20, 2015—more than two\\nyears after her extensive communications with DHS and\\nprevious threat to sue. The district court determined that\\nBird’s claims brought under 42 U.S.C. § 1983 were untimely\\nand granted summary judgment in favor of the State.\\n\\n Because Bird’s complaint is not subject to any exception\\nfrom the normal discovery rule of accrual, we agree that\\nBird’s claim accrued no later than May 2013, when Bird\\nthreatened to sue DHS if she was not removed from the\\nCCAR. We affirm.\\n\\n I. FACTS AND PROCEEDINGS BELOW\\n\\nA. Statutory Background\\n\\n Like most states, Hawai‘i law requires DHS to maintain\\na central registry for reports of child abuse or neglect. See\\nHaw. Rev. Stat. § 350-2(d) (“The department shall maintain\\na central registry of reported child abuse or neglect cases\\n. . . .”). “Central registry reports are typically used to aid\\nagencies in the investigation, treatment, and prevention of\\nchild abuse cases . . . . Central registry records also are used\\nto screen persons who will be entrusted with the care of\\nchildren,” such as “individuals [who] apply[] to be foster or\\nadoptive parents or child or youth care providers.”\\nEstablishment and Maintenance of Central Registrations for\\nChild Abuse or Neglect Reports at 2, Child Welfare\\nInformation Gateway: U.S. Dep’t of Health & Human Serv.,\\nChildren’s Bureau (2018).\\n\\n At times relevant to this suit, § 350-2 further provided\\nthat reports maintained in the central registry “shall [be]\\npromptly expunge[d] . . . if: (1) The department has found the\\n\\n6 BIRD V. STATE OF HAWAI‘I\\n\\nreports to be unsubstantiated; or (2) The petition arising from\\nthe report has been dismissed by order of the family court\\nafter an adjudicatory hearing on the merits.” Act of May 18,\\n2017, sec. 3, § 350-2(d), 2017 Haw. Sess. Laws 60, 61–62.1\\nThe statute stated that “a report is unsubstantiated only when\\nthe department has found the allegations to be frivolous or to\\nhave been made in bad faith.” Id. The statute provided, and\\ncontinues to provide, however, no procedures by which DHS\\nshould investigate, record, or expunge reports in the CCAR.\\nSuch procedures are left to “the department’s rules.” Haw.\\nRev. Stat. § 350-2(a).\\n\\n The Hawai‘i Administrative Rules require DHS to record\\naccused individuals in the CCAR whenever it receives a\\nreport of serious child abuse. Haw. Admin. R. § 17-1610-18.\\nDHS is simultaneously required to commence an\\ninvestigation. Id. At the conclusion of its investigation, DHS\\nwill declare the report “confirmed,” “not confirmed,” or\\n“unsubstantiated.” Id. § 17-1610-19(a)(1)–(3). Only a report\\ndeclared “unsubstantiated” will be expunged from the CCAR.\\nId. § 17-1610-19(a)(1). In cases where the report is declared\\n“confirmed” or “not confirmed,” the individuals subject to the\\nallegations remain on the CCAR. See id. § 17-1610-19(a)(2),\\n(4).2 The regulations specify that DHS “shall comply with\\n\\n\\n 1\\n Haw. Rev. Stat. § 350-2(d) was amended effective May 29, 2017,\\nso that “not confirmed” reports of child abuse are also expunged from the\\nregistry. Act of May 18, 2017, sec. 3, § 350-2(d), 2017 Haw. Sess. Laws\\n60, 61–62.\\n 2\\n We note that while the Hawai‘i statute governing the CCAR has\\nbeen amended to expunge “not confirmed” reports of child abuse, the\\nregulations have not kept pace and still require that “not confirmed”\\nreports of child abuse remain on the registry. See Haw. Admin. R. § 17-\\n1610–19.\\n\\n BIRD V. STATE OF HAWAI‘I 7\\n\\nrequests from other states to check its central registry for the\\npurpose of conducting background checks in foster or\\nadoptive placement cases.” Id. § 17-1610-19(a)(5).\\n\\n Once DHS has completed its investigation and\\ndetermined that an accused individual should remain on the\\nCCAR, state regulations require DHS to give written notice\\nto the “identified perpetrator or maltreater.” Id. § 17-1610-\\n11(c). Anyone “dissatisfied with the disposition of the\\ndepartment’s assessment or action taken by the department”\\nhas two routes for challenging his or her inclusion in the\\nCCAR. Id. § 17-1610-12(b). First, if DHS has not yet\\ncommenced any proceeding against the listed individual in\\nfamily court, he or she may file an administrative appeal. Id.\\nAt the appeal, however, the listed individual remains subject\\nto the requirements for expungement laid out in Hawai‘i\\nRevised Statutes § 350-2, which at all relevant times in this\\ncase required the listed individual to prove that the report was\\n“unsubstantiated,” in other words, that it was “frivolous or . . .\\nmade in bad faith.” Act of May 18, 2017, sec. 3, § 350-2(d),\\n2017 Haw. Sess. Laws 60, 61–62. If the listed individual was\\nunable to meet this exacting standard, he or she remained on\\nthe registry.\\n\\n Alternatively, if DHS has petitioned the family court for\\ncustody of the listed individual’s child(ren), the listed\\nindividual is not permitted to seek administrative review.\\nHaw. Admin. R. § 17-1610-12(c). Instead, the listed\\nindividual’s exclusive means by which to expunge his or her\\nlisting in the CCAR is to prevail at an adjudicatory hearing\\nbefore the family court. Haw. Rev. Stat. § 350-2(d)(2). At\\nthe adjudicatory hearing, the family court will determine\\n“[w]hether the child’s physical or psychological health or\\nwelfare has been harmed or is subject to threatened harm by\\n\\n8 BIRD V. STATE OF HAWAI‘I\\n\\nthe acts or omissions of the [listed individual].” Haw. Rev.\\nStat. § 587A-28(d)(1). Significantly, if DHS returns custody\\nof the child and settles the family court proceeding before it\\nreaches the adjudicatory phase, the statutory scheme leaves\\nlisted individuals who are wrongfully listed in the CCAR\\nwith no recourse for expunging their names. See id.\\n\\nB. Factual Background\\n\\n In 2003, Bird married Frank Fontana and two years later\\ngave birth to a daughter, T.F. Fontana, who worked for the\\nNavy, was subsequently transferred to Hawai‘i. Bird and\\ntheir daughter accompanied him to Hawai‘i to live in base\\nhousing.\\n\\n Shortly after their arrival, in February 2007, Bird gave\\nbirth to a second daughter, C.F. Over the next six weeks,\\nbetween early February and March 20, C.F. was examined by\\nover thirty healthcare personnel in multiple well-child checks,\\nand as she was treated for jaundice, wheezing, apnea induced\\nby acid reflux, and a fever. Over the course of this treatment,\\nC.F. underwent two chest x-rays, neither of which showed\\nany injuries. None of the doctors and nurses that treated C.F.\\never noted any signs of abuse or neglect.\\n\\n On March 28, Bird returned home from a dentist\\nappointment to find Fontana administering CPR to C.F. C.F.\\nwas transported by ambulance to Tripler Army Medical\\nCenter, where she died of cardiac arrest. The doctors and\\nnurses who examined C.F. noted multiple bruises on her\\nbody, evidence of a fractured rib, and evidence of possible\\nfractures in her femurs. As a result, an emergency\\ndepartment nurse at the hospital called the Child Welfare\\n\\n BIRD V. STATE OF HAWAI‘I 9\\n\\nServices branch (“CWS”) of DHS to report a case of possible\\nchild abuse.\\n\\n In accordance with its regulations, DHS listed both Bird\\nand Fontana on the CCAR and initiated an investigation. On\\nApril 2, an investigative multidisciplinary team concluded\\nthat C.F.’s death was likely due to child maltreatment, and\\nrecommended that DHS seek custody of Bird and Fontana’s\\nremaining daughter, T.F. On the basis of this report, CWS\\nconcluded on April 5 that the alleged physical abuse of C.F.\\nby her parents, Fontana and Bird, was “confirmed.” DHS\\nthen filed a Petition for Foster Custody of their remaining\\ndaughter, T.F., with the family court. DHS, however, failed\\nto give Bird notice that she was now listed on the CCAR as\\na “perpetrator or maltreater.”\\n\\n At the same time, Naval Criminal Investigative Services\\n(“NCIS”) initiated an investigation into the death of C.F. In\\nJuly 2007, Fontana “confessed [to Navy investigators] to\\nharming [C.F.] with actions that caused [her] death.” Fontana\\nwas charged with first degree murder and subsequently pled\\nguilty to the crime. In November 2007, NCIS reported to\\nDHS that “[Bird was] not a suspect” in their on-going\\ninvestigation. Nevertheless, because DHS did not conclude\\nthat its initial assessment was “frivolous or in bad faith,”\\nDHS took no action and Bird remained on the CCAR as a\\n“confirmed” child abuser.\\n\\n Meanwhile, Bird was still engaged with the family court\\nin an attempt to regain custody of T.F. “After months of\\nhome checks, supervised visits, and evaluations by various\\nmedical and social work professionals,” Bird was granted\\nphysical custody of T.F. in November 2007. In December,\\nthe family court authorized Bird to leave Hawai‘i with T.F.\\n\\n10 BIRD V. STATE OF HAWAI‘I\\n\\nand return to Tennessee where she could live with her father\\non the condition “that Tennessee DHS approves the . . .\\nplacement, and Tennessee DHS puts in writing that services\\nfor [Bird] and [T.F.] will be in place when mother and\\ndaughter arrive in Tennessee, and both are received by the\\nCourt.” In June 2008, the Hawai‘i family court terminated its\\njurisdiction and revoked the prior order of family supervision.\\nAlthough the family court entered no findings with respect to\\nBird’s alleged mistreatment of C.F., it found that “[T.F.’s]\\nfamily can provide a safe family home without the assistance\\nof a service plan.”\\n\\n Bird alleges that throughout these events, she was never\\ngiven notice, written or otherwise, of her listing on the CCAR\\nand that she “did not have any knowledge that a child abuse\\nregistry even existed.” Without that notice, Bird believed that\\nan adjudicatory trial before the family court concerned only\\nher custody of her surviving daughter, T.F. She did not know\\nthat obtaining a finding or ruling from the family court would\\nbe the only way for her to get her name removed from a\\nregistry she did not know existed. Bird had rejected an\\nadjudicatory trial before the family court, opting instead to\\nwork with DHS to drop the case. Bird contends that without\\nnotice, she did not know “that failure to adjudicate [her]\\nclaim through the Family Court would result in [her] being on\\na list of confirmed child abusers for the rest of [her] life.”\\n\\n After returning to Tennessee, Bird eventually remarried.\\nBird and her new husband decided that they wanted to adopt\\na child from Africa, “[p]referably . . . an older child who was\\nHIV positive so that [they] could provide him or her with the\\nkind of medical and family care that [he or she was] lacking.”\\nDuring the adoption process, the adoption agency conducted\\nbackground checks with each of the states where Bird had\\n\\n BIRD V. STATE OF HAWAI‘I 11\\n\\npreviously lived, and discovered in June 2012 that she was\\nlisted on Hawai‘i’s Central Child Abuse Registry. This report\\nrendered Bird ineligible to adopt.\\n\\n Bird contacted DHS in June 2012 and requested an\\nadministrative hearing before the agency.3 DHS initially\\nrefused to give Bird a hearing, explaining that jurisdiction had\\nbeen turned over to the family court five years prior in\\nconnection with the custody dispute over T.F. After further\\ncorrespondence with Bird and her attorney, DHS offered Bird\\nan administrative hearing limited to determining “whether\\nDHS properly confirmed (i) abuse of a minor and (ii) threat\\nof abuse of a minor.” Bird rejected this hearing, deeming the\\nevidentiary standard impossible to meet. In a letter dated\\nMay 14, 2013, Bird’s attorney indicated to DHS that Bird\\nwould “proceed with litigation” if she was not removed from\\nthe registry.\\n\\nC. Procedural Background\\n\\n In July 2015, Bird filed a complaint in the First Circuit\\nCourt, State of Hawai‘i, seeking both injunctive and\\nmonetary relief against the State of Hawai’i, DHS and various\\nstate officials, in both their official and individual capacities.\\nShe alleged violations of her state constitutional right to due\\nprocess and her federal constitutional rights, a claim brought\\n\\n\\n\\n\\n 3\\n Bird was given official notice that she had been placed on the\\nregistry for the first time in September 2012 when DHS mailed and\\nemailed her a “Notice to Perpetrator.”\\n\\n12 BIRD V. STATE OF HAWAI‘I\\n\\nunder 42 U.S.C. § 1983.4 The defendants timely removed the\\naction to federal court, and sought summary judgment on the\\nground that Bird’s claims fail as a matter of law. At the\\nhearing on the motion, the district court raised sua sponte the\\nissue of whether Bird’s § 1983 claim is barred by the\\napplicable statute of limitations. In response, Bird’s attorney\\nrequested leave to amend the complaint.\\n\\n After supplemental briefing, the district court granted the\\ndefendants’ motion for summary judgment, finding that\\nBird’s § 1983 claim was untimely. The court concluded that\\na two-year statute of limitations applies to Bird’s § 1983\\nclaim, and that her claim had accrued “at the very latest—by\\nMay 2013, more than two years before Bird filed her July 20,\\n2015 Complaint.” The court rejected Bird’s argument that\\nher complaint was protected by the continuing violation\\ndoctrine and denied Bird leave to amend. Having dismissed\\nBird’s § 1983 claim, the court declined to exercise\\nsupplemental jurisdiction over Bird’s remaining state law\\nclaim, and remanded the case to state court. This appeal\\nfollowed.5\\n\\n\\n\\n 4\\n Bird’s complaint does not actually specify which constitutional\\nrights the State violated. Bird’s parallel state constitutional claim,\\nhowever, focuses on a violation of due process, and the injunctive relief\\nthat Bird seeks under her § 1983 claim would require the State to “provide\\nimmediate notice to all individuals who have been reported to the\\nRegistry” and “provide procedures for previously reported individuals to\\nseek to correct false reports.” Thus, Bird’s claim is most aptly\\ncharacterized as a procedural due process claim.\\n 5\\n We review de novo the district court’s conclusion that a claim is\\nbarred by a statute of limitations. Avila v. Spokane Sch. Dist. 81, 852 F.3d\\n936, 939 (9th Cir. 2017).\\n\\n BIRD V. STATE OF HAWAI‘I 13\\n\\n II. DISCUSSION\\n\\n Because 42 U.S.C. § 1983 does not contain its own statute\\nof limitations, “[a]ctions brought pursuant to 42 U.S.C.\\n§ 1983 are governed by the forum state’s statute of\\nlimitations for personal injury actions.” Knox v. Davis, 260\\nF.3d 1009, 1012–13 (9th Cir. 2001) (citing Wilson v. Garcia,\\n471 U.S. 261, 276 (1985)). In Hawai‘i, the statute of\\nlimitations for personal injury actions is two years. See Haw.\\nRev. Stat. § 657-7.\\n\\n Here, the parties disagree over when the statute of\\nlimitations commenced to run. “Although state law\\ndetermines the length of the limitations period, federal law\\ndetermines when a civil rights claim accrues.” Knox, 260\\nF.3d at 1013 (quoting Morales v. City of Los Angeles, 214\\nF.3d 1151, 1153–54 (9th Cir. 2000)). Under federal law, the\\n“discovery rule” typically governs the accrual of § 1983\\nclaims so that “a claim accrues when the plaintiff knows or\\nhas reason to know of the injury which is the basis of the\\naction.” Id. (quoting TwoRivers v. Lewis, 174 F.3d 987, 992\\n(9th Cir. 1999)).\\n\\n Because Bird learned in June 2012 that she had been\\nplaced on the CCAR, requested and was denied a hearing that\\nfall, rejected a proffered hearing in February 2013, and then\\nsent a letter in May 2013 to DHS threatening to “proceed\\nwith litigation,” the district court concluded that Bird was\\naware that she had been placed on the registry, felt this\\nplacement was unjustified, and was contemplating her legal\\noptions “at the very latest” by May 2013.\\n\\n Bird does not seriously dispute that she had knowledge of\\nthe injury giving rise to her claims by May 2013. Instead, she\\n\\n14 BIRD V. STATE OF HAWAI‘I\\n\\nraises two arguments against applying the normal discovery\\nrule of accrual. First, Bird argues that “[w]here a plaintiff\\nchallenges the constitutionality of a statutory or regulatory\\nscheme, the plaintiff’s claim does not accrue until the\\noffending law is repealed.” Second, Bird argues that her\\ncomplaint alleges a “continuing violation”—an exception to\\nthe usual discovery rule of accrual. Because neither argument\\nprevails, we conclude that the normal discovery rule of\\naccrual applies. We agree with the district court’s conclusion\\nthat Bird knew of her injury no later than May 2013, and\\naffirm the district court’s order dismissing Bird’s complaint\\nas time-barred.\\n\\n A\\n\\n Bird first contends that a facial challenge to an allegedly\\nunconstitutional statute is exempt from the traditional rules of\\naccrual. More specifically, Bird asserts that a claim seeking\\n“injunctive relief to invalidate an [ongoing] unconstitutional\\nstatutory and regulatory scheme” does not accrue until the\\nstatute is repealed. In essence, Bird’s proposed rule would\\nmean that a person who has been harmed by an allegedly\\nunconstitutional statute can bring suit at any time so long as\\nthe statute remains in operation.\\n\\n Bird acknowledges that “[t]he Ninth Circuit has not yet\\naddressed” the issue, and relies on several out-of-circuit cases\\nto support her position. First, Bird points to Virginia Hospital\\nAss’n v. Baliles, 868 F.2d 653, 656 (4th Cir. 1989) (“VHA”),\\nwhere the plaintiffs—a nonprofit association of hospitals and\\nother healthcare providers—sought to enjoin the procedures\\nthat Virginia used to determine Medicare reimbursements as\\nviolative of the Medicaid Act, 42 U.S.C. § 1396 et seq., and\\nthe due process clause, U.S. CONST. amend. XIV. When the\\n\\n BIRD V. STATE OF HAWAI‘I 15\\n\\ndefendants argued that the plaintiffs’ claim was barred by the\\napplicable statute of limitations, the district court concluded\\nthat “VHA had alleged an ongoing constitutional violation,\\nand . . . the statute [of limitations] would not have begun to\\nrun until the violation ended.” Id. at 663. The Fourth Circuit\\nagreed, concluding that “[t]he continued enforcement of an\\nunconstitutional statute cannot be insulated by the statute of\\nlimitations.” Id. (alteration in original). Bird highlights the\\nsame language in Kuhnle Brothers, Inc. v. County of Geauga,\\n103 F.3d 516, 518, 522 (6th Cir. 1997), where the Sixth\\nCircuit quoted VHA’s assertion that “[t]he continued\\nenforcement of an unconstitutional statute cannot be insulated\\nby the statute of limitations” and concluded that a trucking\\ncompany’s challenge to a county resolution that banned\\nthrough traffic on certain roads was not time-barred.6\\n\\n Bird’s reliance on this language in VHA and Kuhnle\\nexceeds the context of both cases. In both VHA and Kuhnle,\\nthe defendants had argued that the plaintiff’s facial challenges\\nhad accrued when the statute was enacted as opposed to when\\nthe statute was enforced against the plaintiff. See Kuhnle,\\n103 F.3d at 522 (explaining that a statute “does not become\\nimmunized from legal challenge for all time merely because\\nno one challenges it within two years of its enactment”);\\nVHA, 868 F.2d at 663 (explaining that the parties agreed that\\nVHA’s cause of action accrued “when Virginia enacted its\\ncurrent reimbursement plan”). Faced with the unsavory\\nprospect of denying recovery to plaintiffs who had actually\\nbeen injured within the limitations period merely because the\\nstatute had been enacted outside the limitations period, the\\n\\n\\n 6\\n The Kuhnle court’s discussion relies primarily on the continuing\\nviolations doctrine, see 103 F.3d at 520–23, and is discussed below in Part\\nII.B.\\n\\n16 BIRD V. STATE OF HAWAI‘I\\n\\ncourts responded by allowing the suit to proceed. Thus, VHA\\nand Kuhnle did nothing more than bring the Fourth and Sixth\\nCircuits into alignment with our own view that a facial\\nchallenge to a statute generally accrues when “the statute is\\nenforced—in other words, [when] it is applied.” Levald, Inc.\\nv. City of Palm Desert, 998 F.2d 680, 688 (9th Cir. 1993).\\n\\n Although the courts used broader language in reaching\\ntheir holdings, see Kuhnle, 103 F.3d at 522; VHA, 868 F.2d\\nat 663, we do not believe the cases stand for the broad\\nproposition advocated by Bird. Bird’s argument goes far\\nbeyond the actual outcomes in both cases; here, Bird argues\\nthat her claim did not accrue at the time of her injury, but\\nrather continues ad infinitum until the statute is repealed.\\nAlthough Bird frames her argument in terms of accrual, such\\na position would essentially “nullify all statutes of limitations\\nwith respect to statutory challenges.” VHA, 868 F.2d at 663.\\n\\n Bird cites only one case that arguably supports such a\\nbroad definition of accrual. In Tearpock-Martini v. Borough\\nof Shickshinny, 756 F.3d 232, 234 (3d Cir. 2014), the plaintiff\\nbrought a § 1983 suit to enjoin a religious sign that the\\nmunicipality approved to be installed on public grounds near\\nthe plaintiff’s property. While the plaintiff conceded that a\\ntime period greater than the statutory limitations period had\\nelapsed since the sign was installed, she argued that her\\nclaims were nevertheless timely under the continuing\\nviolations doctrine. Id. at 236. The Third Circuit rejected\\napplication of the continuing violations doctrine—a holding\\ndiscussed in greater detail in Part II.B, infra—and instead\\nconsidered whether the plaintiff’s claims were entirely\\nexempt from the traditional statute of limitations. Id. at\\n237–40. The court first concluded that a federal court should\\napply the state-law statute of limitations to a § 1983 claim\\n\\n BIRD V. STATE OF HAWAI‘I 17\\n\\n“only if it is not ‘inconsistent with the Constitution and laws\\nof the United States.’” Id. at 238 (quoting Burnett v. Grattan,\\n468 U.S. 42, 47–48 (1984)). The court went on to “consider\\nthe significance of the federal rights implicated by an\\nEstablishment Clause claim” and concluded that “the\\ntraditional rationales justifying a limitations period . . . simply\\nhave no persuasive force in this context.” Id. at 238–39.\\n\\n Even if this general exemption from the statute of\\nlimitations is justified in the Establishment Clause context—a\\nquestion that we have not decided, see Maldonado v. Harris,\\n370 F.3d 945, 955–56 (9th Cir. 2004) (“[W]hether a statute of\\nlimitations for § 1983 actions can bar a facial challenge under\\nthe First Amendment to a state statute appears to be a\\nquestion that has not been conclusively resolved by any\\nCircuit court . . . but we need not resolve the . . . question\\nhere . . . .”), and which is not before us today—such a\\ndeparture from the discovery rule of accrual is not warranted\\nhere. The court in Tearpock-Martini expressly limited its\\nholding to the “federal rights implicated by an Establishment\\nClause claim.” 756 F.3d at 238. Here, Bird asserts a due\\nprocess claim and the “traditional rationales justifying a\\nlimitations period—‘to protect defendants against stale or\\nunduly delayed claims,’ ‘facilitat[e] the administration of\\nclaims,’ and ‘promot[e] judicial efficiency’”—are applicable\\nto a due process claim. See id. at 238–39 (alterations in\\noriginal). Unlike “a still-existing monument that\\ncommunicates anew an allegedly unconstitutional\\nendorsement of religion by the government each time it is\\nviewed,” see id. at 239, a due process violation is a discrete\\nevent for which evidence and memory will fade over time.\\nThe traditional interests of protecting defendants and\\nfacilitating the administration of such claims are thus\\napplicable to a due process claim and justify enforcing a\\n\\n18 BIRD V. STATE OF HAWAI‘I\\n\\nlimitations period through the discovery rule of accrual.\\nAlthough enforcing the statute of limitations here will make\\nBird’s alleged losses permanent, such a cost is present in\\nevery case where a plaintiff’s claim is denied on statute of\\nlimitations grounds and is alone insufficient to justify\\ndeparting from the normal discovery rule of accrual.\\n\\n Moreover, Bird’s assertion that applying the discovery\\nrule of accrual to her case will forever immunize an\\nunconstitutional state statute is untrue. The only way for the\\nstate to immunize itself from further suit by future plaintiffs\\nwould be to stop enforcing its statute. As long as Hawai’i\\ncontinues to enforce its statute, it is subject to a facial\\nchallenge by every individual it affects. Enforcing the statute\\nof limitations in the case of a facial challenge to an allegedly\\nunconstitutional statute, thus, does not render the statute\\nimmune from challenge; it merely requires that such\\nchallenges be brought in a timely manner. For these reasons,\\nwe decline to adopt Bird’s proposed rule exempting claims\\nseeking “injunctive relief to invalidate an [ongoing]\\nunconstitutional statutory . . . scheme” from the discovery\\nrule of accrual.\\n\\n B\\n\\n We next consider whether Bird’s claim qualifies as a\\ncontinuing violation. The continuing violations doctrine\\nfunctions as an exception to the discovery rule of accrual\\n“allowing a plaintiff to seek relief for events outside of the\\nlimitations period.” Knox, 260 F.3d at 1013. Although the\\ncontinuing violations doctrine is most frequently seen in the\\ncontext of employment discrimination suits, we have held\\nthat the continuing violations doctrine also applies to § 1983\\nclaims. See id. (“The continuing violation theory applies to\\n\\n BIRD V. STATE OF HAWAI‘I 19\\n\\n§ 1983 actions.”); Cherosky v. Henderson, 330 F.3d 1243,\\n1246 n.3 (9th Cir. 2003) (“[T]he Supreme Court’s analysis of\\nthe continuing violations doctrine is not limited to Title VII\\nactions. It applies with equal force to . . . actions arising\\nunder other civil rights laws.”).\\n\\n “The doctrine of continuing violations . . . is actually a\\nconglomeration of several different ideas,” Williams v.\\nOwens-Illinois, Inc., 665 F.2d 918, 924 (9th Cir. 1982)\\n(internal quotation omitted), the essence of which is that\\n“when a defendant’s conduct is part of a continuing practice,\\nan action is timely so long as the last act evidencing the\\ncontinuing practice falls within the limitations period,”\\nTearpock-Martini, 756 F.3d at 236 (quoting Brenner v. Local\\n514, United Bhd. of Carpenters & Joiners of Am., 927 F.2d\\n1283, 1295 (3d Cir. 1991)). We have recognized two\\napplications of the continuing violations doctrine: first, to “a\\nseries of related acts, one or more of which falls within the\\nlimitations period,” and second, to “the maintenance of a\\ndiscriminatory system both before and during [the\\nlimitations] period.” Gutowsky v. County of Placer, 108 F.3d\\n256, 259 (9th Cir. 1997) (quoting Green v. L.A. Cty.\\nSuperintendent of Schs., 883 F.2d 1472, 1480 (9th Cir.\\n1989)).\\n\\n Prior to 2002, we applied the serial acts branch of the\\ncontinuing violations doctrine when the defendant committed\\na series of acts directed against a single plaintiff. See Morgan\\nv. Nat’l R.R. Passenger Corp., 232 F.3d 1008 (9th Cir. 2000)\\n(“Morgan I”); Kyle Graham, The Continuing Violations\\nDoctrine, 43 GONZ. L. REV. 271, 304 (2008). As long as the\\nplaintiff could show that acts preceding the limitations period\\nwere “sufficiently related” to acts that occurred within the\\nlimitations period, the plaintiff could recover damages for all\\n\\n20 BIRD V. STATE OF HAWAI‘I\\n\\nof the related acts—even those that would otherwise be\\nbarred by the statute of limitations. See Morgan I, 232 F.3d\\nat 1016. In 2002, however, the Supreme Court reversed our\\ndecision in Morgan I and limited the serial acts branch of the\\ncontinuing violations doctrine. Nat’l R.R. Passenger Corp.\\nv. Morgan, 536 U.S. 101, 114 (2002) (“Morgan II”). The\\nCourt held that “discrete . . . acts are not actionable if time\\nbarred, even when they are related to acts alleged in timely\\nfiled charges” because “[e]ach discrete . . . act starts a new\\nclock for filing charges alleging that act.” Id. at 113.7\\n\\n Although Morgan II applied Title VII of the Civil Rights\\nAct of 1964, we have relied on Morgan II to abrogate the\\nserial acts branch of the continuing violations doctrine for\\n§ 1983 claims as well. See Carpinteria Valley Farms, Ltd. v.\\nCounty of Santa Barbara, 344 F.3d 822, 829 (9th Cir. 2003)\\n(“Although Morgan was a Title VII case . . . we have applied\\nMorgan to bar § 1983 claims predicated on discrete time-\\nbarred acts, not-withstanding that those acts are related to\\ntimely-filed claims.”); Cherosky, 330 F.3d at 1246 n.3\\n(concluding that the Supreme Court’s holding in Morgan II\\n“applies with equal force . . . to actions arising under other\\ncivil rights laws”).\\n\\n Unlike the serial acts branch, the systematic branch does\\nnot require the plaintiff to allege specific acts that occurred\\nwithin the statute of limitations period. Originally, the\\nsystematic branch allowed the plaintiff to recover for acts that\\noccurred prior to the limitations period as long as (1) those\\n\\n\\n 7\\n The Court carved out one exception for hostile work environment\\nclaims, because “[s]uch claims are based on the cumulative effect of\\nindividual acts” that “cannot be said to occur on any particular day.”\\nMorgan II, 536 U.S. at 115–17.\\n\\n BIRD V. STATE OF HAWAI‘I 21\\n\\nacts were conducted pursuant to a policy or practice that\\nremained in effect within the statute of limitations period and\\n(2) the plaintiff remained subject or susceptible to the policy\\nwithin the limitations period. See Gutowsky, 108 F.3d at 260\\n(applying the systematic branch of the continuing violations\\ndoctrine where the plaintiff’s “specific examples” of\\ndiscrimination occurred prior to the limitations period, but\\nwhere the “widespread policy and practices of discrimination\\nof which [the plaintiff] complain[ed] continued every day of\\nher employment, including days that fall within the limitation\\nperiod”); Graham, 43 GONZ. L. REV. at 303–04. The theory\\nbehind the systematic branch was that every day the plaintiff\\nwas subject to the policy or practice “constituted a new\\nviolation” sufficient to extend the statute of limitations\\nperiod. See Reed v. Lockheed Aircraft Corp., 613 F.2d 757,\\n759–60 (9th Cir. 1980) (“[E]ach day without promotion\\nconstituted a new violation of Title VII . . . .”).\\n\\n Although the Supreme Court in Morgan II addressed only\\nthe serial acts branch of the continuing violations doctrine,\\nsee 536 U.S. at 115 & n.9, we have applied Morgan II to\\nabrogate the systematic branch of the continuing violations\\ndoctrine as well. For example, in Lyons v. England, 307 F.3d\\n1092, 1107 (9th Cir. 2002), we reasoned from Morgan II that\\na plaintiff’s “assertion that [a] series of discrete acts flows\\nfrom a company-wide, or systematic, discriminatory practice\\nwill not succeed in establishing the employer’s liability for\\nacts occurring outside the limitations period.” Similarly, in\\nCherosky, the plaintiffs sought to avoid Morgan II’s\\nunfavorable holding by arguing under the systematic branch\\nof the continuing violations doctrine; the plaintiffs contended\\nthat they could still recover damages for acts of employment\\ndiscrimination that had occurred prior to the statute of\\nlimitations period as long as those acts had been conducted\\n\\n22 BIRD V. STATE OF HAWAI‘I\\n\\npursuant to a discriminatory company policy. 330 F.3d at\\n1246. We rejected that argument, concluding instead that\\n“[t]he allegation that . . . discrete acts were undertaken\\npursuant to a discriminatory policy does not extend the\\nstatutory limitations period.” 330 F.3d at 1247.\\n\\n Thus, after Morgan II, little remains of the continuing\\nviolations doctrine. Except for a limited exception for hostile\\nwork environment claims—not at issue here—the serial acts\\nbranch is virtually non-existent. Moreover, while we have\\nleft room for the systematic branch to apply to class-wide\\npattern-or-practice claims, see Lyons, 307 F.3d at 1107 n.8\\n(emphasizing that its holding did not “suggest that after\\nMorgan the same plaintiff would be precluded from bringing\\na class-wide pattern-or-practice claim”), we have consistently\\nrefused to apply the systematic branch to rescue\\nindividualized claims that are otherwise time-barred.\\n\\n In this light, the district court’s conclusion that Bird is\\ntime-barred from seeking damages for her placement on the\\nCCAR in 2007 is correct. Because “the violation Bird alleges\\nis the placement of her name on the Registry without\\nconstitutionally required due process,” she has brought only\\nan individualized claim.8 As such, the systematic branch of\\nthe continuing violations doctrine is inapplicable here, and\\nthe discovery rule of accrual applies.\\n\\n Bird raises several arguments to the contrary. First, she\\nlikens her case to that of Kuhnle, where the Seventh Circuit\\n\\n\\n 8\\n Bird did style her complaint as a class action, but “there has been no\\nclass certification granted, or even sought” and “the parties have identified\\nno potential class member plaintiffs.” At oral argument, Bird’s counsel\\nconceded that Bird had not pursued her class action claim.\\n\\n BIRD V. STATE OF HAWAI‘I 23\\n\\nheld that a county resolution which barred the plaintiff from\\nusing certain roads for truck travel “deprived [the plaintiff] of\\nliberty interests . . . created by a fundamental constitutional\\nright to intrastate travel . . . every day that it remained in\\neffect” and thus constituted a continuing violation. 103 F.3d\\nat 521–22. Bird argues that “[s]imilar to the plaintiff in\\nKuhnle . . . [her] § 1983 claim arises from a continued\\ndeprivation of liberty interests caused by an unconstitutional\\nrule or statute.” This argument is, in essence, an attempt to\\nrecast Bird’s claim as a substantive due process claim for\\ndeprivation of liberty interests in working with and adopting\\nchildren.\\n\\n Bird’s case, however, is distinguishable from that of\\nKuhnle. In Kuhnle, the resolution directly forbade the\\nplaintiff from driving on certain roads, thereby depriving him\\nevery day of his “fundamental constitutional right to\\nintrastate travel.” 103 F.3d at 521. By comparison, the\\ndeprivation of liberty that Bird continues to suffer is best\\nunderstood as the “continuing impact from [a] past\\nviolation[].” Williams, 665 F.2d at 924 (quoting Reed,\\n613 F.2d at 760); see Del. State Coll. v. Ricks, 449 U.S. 250,\\n252–54, 258 (1980) (holding that a professor’s eventual\\ntermination was merely the continuing effect of the original\\ndecision to deny tenure and not a new violation); Knox,\\n260 F.3d at 1013 (holding that repeated denials of prison\\nvisitation rights did not constitute a continuing due process\\nviolation because each of the subsequent denials merely\\nimplemented the original suspension and did not represent an\\nindependent violation). Although Bird’s injuries resulting\\nfrom the 2007 act of placing her on the CCAR continue to the\\npresent day, continuing effect is insufficient to constitute a\\ncontinuing violation. See Williams, 665 F.2d at 924\\n\\n24 BIRD V. STATE OF HAWAI‘I\\n\\n(“[C]ontinuing impact from past violations is not actionable.”\\n(quoting Reed, 613 F.2d at 670)).\\n\\n Second, Bird argues that her claim is not confined to the\\ndenial of pre-deprivation process; rather, Bird believes that\\nthe continual lack of a post-deprivation process by which she\\ncan challenge her placement on the CCAR constitutes an\\nongoing deprivation of due process. Bird is unable to\\nidentify, however, any specific instance within the limitations\\nperiod where she sought and was denied post-deprivation\\nprocess. See Pouncil v. Tilton, 704 F.3d 568, 581 (9th Cir.\\n2012) (holding that constitutional and statutory claims for\\nreligious discrimination were not barred by the statute of\\nlimitations where the defendant allegedly committed an\\nindependently wrongful, discrete act within the limitations\\nperiod, even if the discrete act was related to a preexisting\\npolicy of which the plaintiff was aware and subject to outside\\nthe limitations period). The only instance Bird has identified\\nwhere she sought and was denied a post-deprivation hearing\\noccurred prior to May 2013.\\n\\n ***\\n\\n Because Bird has alleged only individualized claims for\\ndeprivation of procedural due process, the normal discovery\\nrule of accrual applies. Based on the correspondence between\\nBird and DHS, we agree with the district court that Bird had\\nknowledge of the injury giving rise to her claims by May\\n2013. Accordingly, her suit, filed in July 2015, was outside\\nthe two-year statute of limitations. Because “the complaint\\nwould not [have been] saved by any amendment,” Carvalho\\nv. Equifax Info. Servs., LLC, 629 F.3d 876, 893 (9th Cir.\\n2010) (quoting Leadsinger, Inc. v. BMG Music Publ’g,\\n\\n BIRD V. STATE OF HAWAI‘I 25\\n\\n512 F.3d 522, 532 (9th Cir. 2008)), the district court did not\\nerr in denying Bird leave to amend.\\n\\n III. CONCLUSION\\n\\n We AFFIRM the judgment of the district court.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4097,"cells":{"id":{"kind":"number","value":4655519,"string":"4,655,519"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"Adree Edmo v. Corizon, Inc."},"decision_date":{"kind":"string","value":"2019-08-23"},"docket_number":{"kind":"string","value":"19-35017"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: M. Margaret McKeown and Ronald M. Gould, Circuit Judges, and Robert S. Lasnik, ** District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"B. Lynn Winmill, Chief District Judge, Presiding\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n ADREE EDMO, AKA Mason Edmo, No. 19-35017\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 1:17-cv-00151-\\n BLW\\n CORIZON, INC.; SCOTT ELIASON;\\n MURRAY YOUNG; CATHERINE\\n WHINNERY,\\n Defendants-Appellants,\\n\\n and\\n\\n IDAHO DEPARTMENT OF\\n CORRECTIONS; AL RAMIREZ, in his\\n official capacity as warden of Idaho\\n State Correctional Institution; *\\n HENRY ATENCIO; JEFF ZMUDA;\\n HOWARD KEITH YORDY; RICHARD\\n CRAIG; RONA SIEGERT,\\n Defendants.\\n\\n\\n ADREE EDMO, AKA Mason Edmo, No. 19-35019\\n Plaintiff-Appellee,\\n\\n\\n\\n\\n *\\n Al Ramirez is substituted in his official capacity for his\\npredecessor, Howard Keith Yordy, pursuant to Rule 43(c)(2) of the\\nFederal Rules of Appellate Procedure.\\n\\f2 EDMO V. CORIZON\\n\\n\\n v. D.C. No.\\n 1:17-cv-00151-\\n IDAHO DEPARTMENT OF BLW\\n CORRECTIONS; AL RAMIREZ, in his\\n official capacity as warden of Idaho\\n State Correctional Institution; HENRY OPINION\\n ATENCIO; JEFF ZMUDA; HOWARD\\n KEITH YORDY; RICHARD CRAIG;\\n RONA SIEGERT,\\n Defendants-Appellants,\\n\\n and\\n\\n CORIZON, INC.; SCOTT ELIASON;\\n MURRAY YOUNG; CATHERINE\\n WHINNERY,\\n Defendants.\\n\\n Appeal from the United States District Court\\n for the District of Idaho\\n B. Lynn Winmill, Chief District Judge, Presiding\\n\\n Argued and Submitted May 16, 2019\\n San Francisco, California\\n\\n Filed August 23, 2019\\n\\n Before: M. Margaret McKeown and Ronald M. Gould,\\n Circuit Judges, and Robert S. Lasnik, ** District Judge.\\n\\n Per Curiam Opinion\\n\\n **\\n The Honorable Robert S. Lasnik, United States District Judge for\\nthe Western District of Washington, sitting by designation.\\n\\f EDMO V. CORIZON 3\\n\\n SUMMARY ***\\n\\n\\n Eighth Amendment / Prisoner Rights\\n\\n The panel affirmed the district court’s entry of a\\npermanent injunction in favor of Idaho state prisoner Adree\\nEdmo, but vacated the injunction to the extent it applied to\\ndefendants Corizon, Howard Yordy, Rona Siegert, Dr.\\nYoung, Dr. Craig, and Dr. Whinnery, in their individual\\ncapacities, in Edmo’s action seeking medical treatment for\\ngender dysphoria.\\n\\n The district court concluded that Edmo had established\\nher Eighth Amendment claim. The district court further\\nconcluded that gender confirmation surgery (“GCS”) was\\nmedically necessary for Edmo, and ordered the State to\\nprovide the surgery.\\n\\n The panel credited the district court’s factual findings as\\nlogical and well-supported, and held that the responsible\\nprison authorities were deliberately indifferent to Edmo’s\\ngender dysphoria, in violation of the Eighth Amendment.\\nThe panel held that the record, as construed by the district\\ncourt, established that Edmo had a serious medical need, that\\nthe appropriate medical treatment was GCS, and that prison\\nauthorities had not provided that treatment despite full\\nknowledge of Edmo’s ongoing and extreme suffering and\\nmedical needs. The panel rejected the State’s position that\\nthere was a reasoned disagreement between qualified\\nmedical professionals. The panel emphasized that its\\n\\n\\n ***\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f4 EDMO V. CORIZON\\n\\nanalysis was individual to Edmo, and rested on the record of\\nthis case.\\n\\n Addressing further aspects of the appeal, the panel\\nrejected the State’s contention that the district court did not\\nmake the Prison Litigation Reform Act’s requisite “need-\\nnarrowness-intrusiveness” findings, causing the injunction\\nto automatically expire and mooting the appeal. The panel\\nheld that the district court’s order, considered as a whole,\\nmade all the findings required by 18 U.S.C. § 3626(a)(1)(A),\\nand Ninth Circuit precedent. The panel also held that the\\npermanent injunction that the district court entered had not\\nexpired, and remained in place, albeit stayed. The panel\\naccordingly denied the State’s motion to dismiss.\\n\\n The panel held that the district court did not err in\\ngranting a permanent injunction. Specifically, the panel\\nheld, based on the district court’s factual findings, that Edmo\\nestablished her Eighth Amendment claim and that she will\\nsuffer irreparable harm – in the form of ongoing mental\\nanguish and possible physical harm – if GCS is not provided.\\nThe State did not dispute that Edmo’s gender dysphoria was\\na sufficiently serious medical need to trigger the State’s\\nobligations under the Eighth Amendment. The panel held\\nthat the district court did not err in crediting the testimony of\\nEdmo’s experts that GCS was medically necessary to treat\\nEdmo’s gender dysphoria and that the State’s failure to\\nprovide that treatment was medically unacceptable. The\\npanel further held that the district court did not err in\\ndiscrediting the State’s experts because aspects of their\\nopinions were illogical and unpersuasive. Also, the panel\\nheld that the record demonstrated that Dr. Eliason acted with\\ndeliberate indifference to Edmo’s serious medical needs.\\nThe panel noted that its decision was in tension with the Fifth\\nCircuit’s decision in Gibson v. Collier, 920 F.3d 212 (5th\\n\\f EDMO V. CORIZON 5\\n\\nCir. 2019), and the panel rejected that decision’s categorical\\nholding that denying GCS cannot, as a matter of law, violate\\nthe Eighth Amendment.\\n\\n The panel held that the district court did not err in finding\\nthat Edmo would be irreparably harmed absent an\\ninjunction. The panel rejected the State’s contentions as to\\nwhy the district court erred in this finding.\\n\\n The panel next considered the State’s challenges to the\\nscope of the injunction. The panel held that the injunction\\nwas properly entered against Dr. Eliason because he\\npersonally participated in the deprivation of Edmo’s\\nconstitutional rights. The panel also held that because Edmo\\nmay properly pursue her Eighth Amendment claim for\\ninjunctive relief against Attencio, Zmuda and Ramirez in\\ntheir official capacities, they were properly included within\\nthe scope of the district court’s injunction. On remand, the\\ndistrict court shall amend the injunction to substitute the\\ncurrent warden as a party for Yordy. The panel vacated the\\ndistrict court’s injunction to the extent it applied to Yordy,\\nSiegert, Dr. Young, Dr. Craig, and Dr. Whinnery in their\\nindividual capacities because the evidence in the record was\\ninsufficient to conclude that they were deliberately\\nindifferent to Edmo’s serious medical needs. The panel\\nvacated the injunction as to Corizon, and remanded with\\ninstructions to the district court to modify the injunction to\\nexclude Corizon. Finally, the panel held that the injunctive\\nrelief ordered was not overbroad.\\n\\n The panel considered the State’s challenges to the\\nprocedure used by the district court. The panel rejected the\\nState’s contention that the district court erroneously\\nconverted the evidentiary hearing into a final trial on the\\nmerits without giving proper notice. The panel held that the\\nState did receive notice, and in any event, the State had not\\n\\f6 EDMO V. CORIZON\\n\\nshown any prejudice. The panel also rejected the State’s\\ncontention that the district court violated defendants’\\nSeventh Amendment right to a jury trial by converting the\\nevidentiary hearing into a trial on the merits. The panel held\\nthat the State’s conduct waived its right to a jury trial with\\nrespect to issues common to Edmo’s request for an\\ninjunction ordering GCS and her legal claims.\\n\\n\\n COUNSEL\\n\\nBrady J. Hall (argued), Special Deputy Attorney General;\\nLawrence G. Wasden, Attorney General; Office of the\\nAttorney General, Boise, Idaho; Marisa S. Crecelius, Moore\\nElia Kraft & Hall LLP, Boise, Idaho; for Defendants-\\nAppellants Idaho Department of Corrections, Henry\\nAtencio, Jeff Zmuda, Howard Keith Yordy, Richard Craig,\\nand Rona Siegert.\\n\\nDylan A. Eaton (argued), J. Kevin West, and Bryce Jensen,\\nParsons Behle & Latimer, Boise, Idaho, for Defendants-\\nAppellants Corizon, Inc.; Scott Eliason; Murray Young; and\\nCatherine Whinnery.\\n\\nLori Rifkin (argued), Hadsell Stormer & Renick LLP,\\nEmeryville, California; Dan Stormer and Shaleen Shanbhag,\\nHadsell Stormer & Renick LLP, Pasadena, California; Craig\\nDurham and Deborah Ferguson, Ferguson Durham PLLC,\\nBoise, Idaho; Amy Whelan and Julie Wilensky, National\\nCenter for Lesbian Rights, San Francisco, California; for\\nPlaintiff-Appellee.\\n\\nDavid M. Shapiro, Sheila A. Bedi, and Vanessa del Valle,\\nRoderick & Solange MacArthur Justice Center, Chicago,\\nIllinois; Molly E. Whitman, Akin Gump Strauss Hauer &\\n\\f EDMO V. CORIZON 7\\n\\nFeld LLP, Dallas, Texas; for Amici Curiae Andrea\\nArmstrong, Sharon Dolovich, Betsy Ginsberg, Michael B.\\nMushlin, Alexander A. Reinert, Laura Rovner, and Margo\\nSchlanger.\\n\\nMolly Kafka and Richard Alan Eppink, ACLU of Idaho\\nFoundation, Boise, Idaho; Devon A. Little and Derek\\nBorchardt, Walden Macht & Haran LLP, New York, New\\nYork; Amy Fettig and Jennifer Wedekind, ACLU National\\nPrison Project, Washington, D.C.; Gabriel Arkles and Rose\\nSaxe, ACLU LGBT & HIV Project/ACLU Foundation, New\\nYork, New York; for Amici Curiae Former Corrections\\nOfficials.\\n\\nDevi M. Rao and Jason T. Perkins, Jenner & Block LLP,\\nWashington, D.C., for Amici Curiae Medical and Mental\\nHealth Professional Organizations.\\n\\nSharif E. Jacob, Ryan K. M. Wong, Kristin E. Hucek, and\\nPatrick E. Murray, Keker Van Nest & Peters LLP, San\\nFrancisco, California, for Amicus Curiae Jody L. Herman.\\n\\nAlan E. Schoenfeld, Wilmer Cutler Pickering Hale and Dorr\\nLLP, New York, New York; Michael Posada, Wilmer Cutler\\nPickering Hale and Dorr LLP, Washington, D.C.; Richard\\nSaenz, Lambda Legal Defense & Education Fund Inc., New\\nYork, New York; A. Chinyere Ezie, Center for\\nConstitutional Rights, New York, New York; for Amici\\nCuriae Civil Rights & Non-Profit Organizations.\\n\\f8 EDMO V. CORIZON\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n The Eighth Amendment prohibits “cruel and unusual\\npunishments.” U.S. Const. amend. VIII. “The Amendment\\nembodies broad and idealistic concepts of dignity, civilized\\nstandards, humanity, and decency . . . .” Estelle v. Gamble,\\n429 U.S. 97, 102 (1976) (quotation omitted). Our society\\nrecognizes that prisoners “retain the essence of human\\ndignity inherent in all persons.” Brown v. Plata, 563 U.S.\\n493, 510 (2011).\\n\\n Consistent with the values embodied by the Eighth\\nAmendment, for more than 40 years the Supreme Court has\\nheld that “deliberate indifference to serious medical needs”\\nof prisoners constitutes cruel and unusual punishment.\\nEstelle, 429 U.S. at 106. When prison authorities do not\\nabide by their Eighth Amendment duty, “the courts have a\\nresponsibility to remedy the resulting . . . violation.” Brown,\\n563 U.S. at 511. We do so here.\\n\\n Adree Edmo (formerly Mason Dean Edmo) is a male-to-\\nfemale transgender prisoner in the custody of the Idaho\\nDepartment of Correction (“IDOC”). Edmo’s sex assigned\\nat birth (male) differs from her gender identity (female). The\\nincongruity causes Edmo to experience persistent distress so\\nsevere it limits her ability to function. She has twice\\nattempted self-castration to remove her male genitalia,\\nwhich cause her profound anguish.\\n\\n Both sides and their medical experts agree: Edmo suffers\\nfrom gender dysphoria, a serious medical condition. They\\nalso agree that the appropriate benchmark regarding\\ntreatment for gender dysphoria is the World Professional\\nAssociation of Transgender Health Standards of Care for the\\n\\f EDMO V. CORIZON 9\\n\\nHealth of Transsexual, Transgender, and Gender\\nNonconforming People (“WPATH Standards of Care”).\\nAnd the State 1 does not seriously dispute that in certain\\ncircumstances, gender confirmation surgery (“GCS”) can be\\na medically necessary treatment for gender dysphoria. The\\nparties’ dispute centers around whether GCS is medically\\nnecessary for Edmo—a question we analyze with deference\\nto the district court’s factual findings.\\n\\n Following four months of intensive discovery and a\\nthree-day evidentiary hearing, the district court concluded\\nthat GCS is medically necessary for Edmo and ordered the\\nState to provide the surgery. Its ruling hinged on findings\\nindividual to Edmo’s medical condition. The ruling also\\nrested on the finding that Edmo’s medical experts testified\\npersuasively that GCS was medically necessary, whereas\\ntestimony from the State’s medical experts deserved little\\nweight. In contrast to Edmo’s experts, the State’s witnesses\\nlacked relevant experience, could not explain their\\ndeviations from generally accepted guidelines, and testified\\nillogically and inconsistently in important ways.\\n\\n The district court’s detailed factual findings were amply\\nsupported by its careful review of the extensive evidence and\\ntestimony. Indeed, they are essentially unchallenged. The\\nappeal boils down to a disagreement about the implications\\nof the factual findings.\\n\\n Crediting, as we must, the district court’s logical, well-\\nsupported factual findings, we hold that the responsible\\n 1\\n In addition to IDOC, Edmo sued Corizon, Inc. (a private for-profit\\ncorporation that provides health care to inmates in IDOC custody) and\\nvarious employees of IDOC and Corizon. The defendants briefed the\\ncase jointly, and for ease of reference we refer to them collectively as\\n“the State.”\\n\\f10 EDMO V. CORIZON\\n\\nprison authorities have been deliberately indifferent to\\nEdmo’s gender dysphoria, in violation of the Eighth\\nAmendment. The record before us, as construed by the\\ndistrict court, establishes that Edmo has a serious medical\\nneed, that the appropriate medical treatment is GCS, and that\\nprison authorities have not provided that treatment despite\\nfull knowledge of Edmo’s ongoing and extreme suffering\\nand medical needs. In so holding, we reject the State’s\\nportrait of a reasoned disagreement between qualified\\nmedical professionals. We also emphasize that the analysis\\nhere is individual to Edmo and rests on the record in this\\ncase. We do not endeavor to project whether individuals in\\nother cases will meet the threshold to establish an Eighth\\nAmendment violation. The district court’s order entering\\ninjunctive relief for Edmo is affirmed, with minor\\nmodifications noted below.\\n\\n Our opinion proceeds as follows. In Part I, we provide\\nbackground on gender dysphoria, the standard of care, and\\nthe evidence considered and factual findings made by the\\ndistrict court. Part II explains why this appeal complies with\\nthe Prison Litigation Reform Act (“PLRA”) and is not moot.\\nIn Part III, we turn to the gravamen of the appeal: Edmo’s\\nEighth Amendment claim and showing of irreparable injury.\\nPart IV addresses the State’s challenges to the injunction’s\\nscope and narrows the injunction as to certain defendants.\\nPart V rejects the State’s objections to the procedure\\nemployed by the district court. We conclude in Part VI.\\n\\f EDMO V. CORIZON 11\\n\\n I. Background 2\\n\\n A. Gender Dysphoria and its Treatment\\n\\n Transgender individuals have a “[g]ender identity”—a\\n“deeply felt, inherent sense” of their gender—that does not\\nalign with their sex assigned at birth. 3 Am. Psychol. Ass’n,\\nGuidelines for Psychological Practice with Transgender\\nand Gender Nonconforming People, 70 Am. Psychologist\\n832, 834 (2015). Recent estimates suggest that\\napproximately 1.4 million transgender adults live in the\\nUnited States, or 0.6 percent of the adult population.\\nAndrew R. Flores et al., The Williams Inst., How Many\\nAdults Identify as Transgender in the United States?, at\\n2 (2016), http://williamsinstitute.law.ucla.edu/wp-content/\\nuploads/How-Many-Adults-Identify-as-Transgender-in-the-\\nUnited-States.pdf.\\n\\n Gender dysphoria 4 is “[d]istress that is caused by a\\ndiscrepancy between a person’s gender identity and that\\nperson’s sex assigned at birth (and the associated gender role\\nand/or primary and secondary sex characteristics).” World\\nProf’l Ass’n for Transgender Health, Standards of Care for\\nthe Health of Transsexual, Transgender, and Gender-\\n\\n 2\\n The following sections are derived from the district court’s factual\\nfindings and the record on appeal.\\n 3\\n At birth, infants are classified as male or female based on visual\\nobservation of their external genitalia. This is a person’s “sex assigned\\nat birth,” but it may not be the person’s gender identity.\\n 4\\n Until recently, the medical community commonly referred to\\ngender dysphoria as “gender identity disorder.” See Kosilek v. Spencer,\\n774 F.3d 63, 68 n.1 (1st Cir. 2014).\\n\\f12 EDMO V. CORIZON\\n\\nNonconforming People 2 (7th ed. 2011) (hereinafter\\n“WPATH SOC”). The Fifth Edition of the American\\nPsychiatric Association’s Diagnostic and Statistical Manual\\nof Mental Disorders (“DSM-5”) sets forth two conditions\\nthat must be met for a person to be diagnosed with gender\\ndysphoria. 5\\n\\n First, there must be “[a] marked incongruence between\\none’s experienced/expressed gender and assigned gender, of\\nat least 6 months’ duration, as manifested by at least two of\\nthe following”:\\n\\n (1) “a marked incongruence between one’s\\n experienced/expressed gender and primary\\n and/or secondary sex characteristics”;\\n\\n (2) “a strong desire to be rid of one’s primary\\n and/or secondary sex characteristics because\\n of a marked incongruence with one’s\\n experienced/expressed gender”;\\n\\n (3) “a strong desire for the primary and/or\\n secondary sex characteristics of the other\\n gender”;\\n\\n (4) “a strong desire to be of the other gender”;\\n\\n (5) “a strong desire to be treated as the other\\n gender”; or\\n\\n\\n\\n\\n 5\\n Each expert in the case used these criteria to determine whether\\nEdmo has gender dysphoria.\\n\\f EDMO V. CORIZON 13\\n\\n (6) “a strong conviction that one has the\\n typical feelings and reactions of the other\\n gender.”\\n\\nAm. Psychiatric Ass’n, Diagnostic and Statistical Manual of\\nMental Disorders 452 (5th ed. 2013) (hereinafter “DSM-5”).\\nSecond, the person’s condition must be associated with\\n“clinically significant distress”—i.e., distress that impairs or\\nseverely limits the person’s ability to function in a\\nmeaningful way and has reached a threshold that requires\\nmedical or surgical intervention, or both. Id. at 453, 458.\\nNot every transgender person has gender dysphoria, and not\\nevery gender dysphoric person has the same medical needs.\\n\\n Gender dysphoria is a serious but treatable medical\\ncondition. Left untreated, however, it can lead to debilitating\\ndistress, depression, impairment of function, substance use,\\nself-surgery to alter one’s genitals or secondary sex\\ncharacteristics, self-injurious behaviors, and even suicide.\\n\\n The district court found that the World Professional\\nAssociation of Transgender Health Standards of Care for the\\nHealth of Transsexual, Transgender, and Gender\\nNonconforming People (“WPATH Standards of Care”) 6\\n“are the internationally recognized guidelines for the\\ntreatment of individuals with gender dysphoria.” Edmo v.\\nIdaho Dep’t of Corr., 358 F. Supp. 3d 1103, 1111 (D. Idaho\\n2018). Most courts agree. See, e.g., De’lonta v. Johnson,\\n708 F.3d 520, 522–23 (4th Cir. 2013); Keohane v. Jones,\\n328 F. Supp. 3d 1288, 1294 (N.D. Fla. 2018), appeal filed,\\n\\n 6\\n The WPATH Standards of Care were formerly referred to as the\\n“Harry Benjamin Standards of Care” and were promulgated by WPATH\\nunder its former name, the “Harry Benjamin International Gender\\nDysphoria Association.” Kosilek, 774 F.3d at 70 & n.3.\\n\\f14 EDMO V. CORIZON\\n\\nNo. 18-14096 (11th Cir. 2018); Norsworthy v. Beard, 87 F.\\nSupp. 3d 1164, 1170 (N.D. Cal.), appeal dismissed &\\nremanded, 802 F.3d 1090 (9th Cir. 2015); Soneeya v.\\nSpencer, 851 F. Supp. 2d 228, 231–32 (D. Mass. 2012). But\\nsee Gibson v. Collier, 920 F.3d 212, 221 (5th Cir. 2019)\\n(“[T]he WPATH Standards of Care reflect not consensus,\\nbut merely one side in a sharply contested medical debate\\nover [GCS].”); cf. Kosilek, 774 F.3d at 76–79 (recounting\\ntestimony questioning the WPATH Standards of Care). And\\nmany of the major medical and mental health groups in the\\nUnited States—including the American Medical\\nAssociation, the American Medical Student Association, the\\nAmerican Psychiatric Association, the American\\nPsychological Association, the American Family Practice\\nAssociation, the Endocrine Society, the National\\nAssociation of Social Workers, the American Academy of\\nPlastic Surgeons, the American College of Surgeons, Health\\nProfessionals Advancing LGBTQ Equality, the HIV\\nMedicine Association, the Lesbian, Bisexual, Gay and\\nTransgender Physician Assistant Caucus, and Mental Health\\nAmerica—recognize the WPATH Standards of Care as\\nrepresenting the consensus of the medical and mental health\\ncommunities regarding the appropriate treatment for\\ntransgender and gender dysphoric individuals.\\n\\n Each expert in this case relied on the WPATH Standards\\nof Care in rendering an opinion. As the State acknowledged\\nto the district court, the WPATH Standards of Care “provide\\nthe best guidance,” and “are the best standards out there.”\\n“There are no other competing, evidence-based standards\\nthat are accepted by any nationally or internationally\\nrecognized medical professional groups.” Edmo, 358 F.\\nSupp. 3d at 1125.\\n\\f EDMO V. CORIZON 15\\n\\n “[B]ased on the best available science and expert\\nprofessional consensus,” the WPATH Standards of Care\\nprovide “flexible clinical guidelines” “to meet the diverse\\nhealth care needs of transsexual, transgender, and gender\\nnonconforming people.” WPATH SOC at 1–2. Treatment\\nunder the WPATH Standards of Care must be\\nindividualized: “[w]hat helps one person alleviate gender\\ndysphoria might be very different from what helps another\\nperson.” Id. at 5. “Clinical departures from the [WPATH\\nStandards of Care] may come about because of a patient’s\\nunique anatomic, social, or psychological situation; an\\nexperienced health professional’s evolving method of\\nhandling a common situation; a research protocol; lack of\\nresources in various parts of the world; or the need for\\nspecific harm reduction strategies.” Id. at 2.\\n\\n The WPATH Standards of Care identify the following\\nevidence-based treatment options for individuals with\\ngender dysphoria:\\n\\n (1) “changes in gender expression and role\\n (which may involve living part time or full\\n time in another gender role, consistent with\\n one’s gender identity)”;\\n\\n (2) “psychotherapy (individual, couple,\\n family, or group) for purposes such as\\n exploring gender identity, role, and\\n expression[,] addressing the negative impact\\n of gender dysphoria and stigma on mental\\n health[,] alleviating internalized\\n transphobia[,] enhancing social and peer\\n support[,] improving body image[,] or\\n promoting resilience”;\\n\\f16 EDMO V. CORIZON\\n\\n (3) “hormone therapy to feminize or\\n masculinize the body”; and\\n\\n (4) “surgery to change primary and/or\\n secondary sex characteristics (e.g.,\\n breasts/chest, external and/or internal\\n genitalia, facial features, body contouring).”\\n\\nId. at 10. The WPATH Standards of Care state that many\\nindividuals “find comfort with their gender identity, role,\\nand expression without surgery.” Id. at 54. For others,\\nhowever, “surgery is essential and medically necessary to\\nalleviate their gender dysphoria.” Id. That group cannot\\nachieve “relief from gender dysphoria . . . without\\nmodification of their primary and/or secondary sex\\ncharacteristics to establish greater congruence with their\\ngender identity.” Id. at 55; see also Jae Sevelius & Valerie\\nJenness, Challenges and Opportunities for Gender-\\nAffirming Healthcare for Transgender Women in Prison,\\n13 Int’l J. Prisoner Health 32, 36 (2017) (“Negative\\noutcomes such as genital self-harm, including autocastration\\nand/or autopenectomy, can arise when gender-affirming\\nsurgeries are delayed or denied.”); George R. Brown &\\nEverett McDuffie, Health Care Policies Addressing\\nTransgender Inmates in Prison Systems in the United States,\\n15 J. Corr. Health Care 280, 287–88 (2009) (describing the\\nauthors’ “firsthand knowledge of completed autocastration\\nand/or autopenectomy in six facilities in four states”).\\n\\n The weight of opinion in the medical and mental health\\ncommunities agrees that GCS is safe, effective, and\\nmedically necessary in appropriate circumstances. See, e.g.,\\nU.S. Dep’t of Health & Human Servs., No. A-13-87,\\nDecision No. 2576, (Dep’t Appeals Bd. May 30, 2014);\\nRandi Ettner, et al., Principles of Transgender Medicine and\\n\\f EDMO V. CORIZON 17\\n\\nSurgery 109–11 (2d ed. 2016); Jordan D. Frey, et al., A\\nHistorical Review of Gender-Affirming Medicine: Focus on\\nGenital Reconstruction Surgery, 14 J. Sexual Med. 991, 991\\n(2017); Cynthia S. Osborne & Anne A. Lawrence, Male\\nPrison Inmates With Gender Dysphoria: When Is Sex\\nReassignment Surgery Appropriate?, 45 Archives of Sexual\\nBehav. 1649, 1651–53 (2016); see also De’lonta, 708 F.3d\\nat 523 (“Pursuant to the Standards of Care, after at least one\\nyear of hormone therapy and living in the patient’s identified\\ngender role, sex reassignment surgery may be necessary for\\nsome individuals for whom serious symptoms persist. In\\nthese cases, the surgery is not considered experimental or\\ncosmetic; it is an accepted, effective, medically indicated\\ntreatment for [gender dysphoria].”).\\n\\n The WPATH criteria for genital reconstruction surgery\\nin male-to-female patients include the following:\\n\\n (1) “persistent, well documented gender\\n dysphoria”;\\n\\n (2) “capacity to make a fully informed\\n decision and to consent for treatment”;\\n\\n (3) “age of majority in a given country”;\\n\\n (4) “if significant medical or mental health\\n concerns are present, they must be well\\n controlled”;\\n\\n (5) “12 continuous months of hormone\\n therapy as appropriate to the patient’s gender\\n goals”; and\\n\\f18 EDMO V. CORIZON\\n\\n (6) “12 continuous months of living in a\\n gender role that is congruent with their\\n gender identity.”\\n\\nWPATH SOC at 60. The parties’ dispute focuses on\\nwhether Edmo satisfied the fourth and sixth criteria.\\n\\n With respect to the fourth criterion, the WPATH\\nStandards of Care provide that coexisting medical or mental\\nhealth concerns unrelated to the person’s gender dysphoria\\ndo not necessarily preclude surgery. Id. at 25. But those\\nconcerns need to be managed prior to, or concurrent with,\\ntreatment of a person’s gender dysphoria. Id. Coexisting\\nmedical or mental health issues resulting from a person’s\\ngender dysphoria are not an impediment under the fourth\\ncriterion. It may be difficult to determine, however, whether\\nmental or medical health concerns result from the gender\\ndysphoria or are unrelated.\\n\\n The WPATH Standards of Care explain that the sixth\\ncriterion—living for 12 months in an identity-congruent\\nrole—is intended to ensure that the person experiences the\\nfull range of “different life experiences and events that may\\noccur throughout the year.” Id. at 61. During that time, the\\npatient should present consistently in her desired gender\\nrole. Id.\\n\\n Scientific studies show that the regret rate for individuals\\nwho undergo GCS is low, in the range of one to two percent.\\nSee, e.g., Osborne & Lawrence, Male Prison Inmates With\\nGender Dysphoria, 45 Archives of Sexual Behav. at 1660;\\nWilliam Byne, et al., Report of the American Psychiatric\\nAssociation Task Force on Treatment of Gender Identity\\nDisorder, 41 Archives of Sexual Behav. 759, 780–81\\n(2012). The district court found, and the State does not\\ndispute on appeal, that Edmo does not have any of the risk\\n\\f EDMO V. CORIZON 19\\n\\nfactors that would make her likely to regret GCS. See Edmo,\\n358 F. Supp. 3d at 1121.\\n\\n The WPATH Standards of Care apply equally to all\\nindividuals “irrespective of their housing situation” and\\nexplicitly state that health care for transgender individuals\\n“living in an institutional environment should mirror that\\nwhich would be available to them if they were living in a\\nnon-institutional setting within the same community.”\\nWPATH SOC at 67. The next update to the WPATH\\nStandards of Care will likewise apply equally to incarcerated\\npersons. The National Commission on Correctional Health\\nCare (“NCCHC”), a leading professional organization in\\nhealth care delivery in the correctional context, endorses the\\nWPATH Standards of Care as the accepted standards for the\\ntreatment of transgender prisoners.\\n\\n In summary, the broad medical consensus in the area of\\ntransgender health care requires providers to individually\\ndiagnose, assess, and treat individuals’ gender dysphoria,\\nincluding for those individuals in institutionalized\\nenvironments. Treatment can and should include GCS when\\nmedically necessary. Failure to follow an appropriate\\ntreatment plan can expose transgender individuals to a\\nserious risk of psychological and physical harm. The State\\ndoes not dispute these points; it contends that GCS is not\\nmedically necessary for Edmo.\\n\\n B. Edmo’s Treatment\\n\\n Edmo is a transgender woman in IDOC custody. Her sex\\nassigned at birth was male, but she identifies as female. In\\nher words, “my brain typically operates female, even though\\nmy body hasn’t corresponded with my brain.”\\n\\f20 EDMO V. CORIZON\\n\\n Edmo has been incarcerated since pleading guilty in\\n2012 to sexual abuse of a 15-year-old male at a house party.\\nEdmo was 21 years old at the time of the criminal offense.\\nEdmo is currently incarcerated at the Idaho State\\nCorrectional Institution (“ISCI”). At the time of the\\nevidentiary hearing, she was 30 years old and due to be\\nreleased from prison in 2021.\\n\\n Edmo has viewed herself as female since age 5 or 6. She\\nstruggled with her gender identity as a child and teenager,\\npresenting herself intermittently as female, but around age\\n20 or 21 she began living fulltime as a woman.\\n\\n Although she identified as female from an early age,\\nEdmo first learned the term “gender dysphoria” and the\\ncontours of that diagnosis around the time of her\\nincarceration. Shortly thereafter, Corizon psychiatrist\\nDr. Scott Eliason diagnosed her with “gender identity\\ndisorder,” now referred to as gender dysphoria. Corizon\\npsychologist Dr. Claudia Lake confirmed that diagnosis.\\n\\n While incarcerated, Edmo has changed her legal name to\\nAdree Edmo and the sex on her birth certificate to “female”\\nto affirm her gender identity. Throughout her incarceration,\\nEdmo has consistently presented as female, despite receiving\\nmany disciplinary offense reports for doing so. For example,\\nwhen able to do so, Edmo has worn her hair in feminine\\nhairstyles and worn makeup, for which she has received\\nmultiple disciplinary offense reports. 7 Medical providers\\nhave documented Edmo’s feminine presentation since 2012.\\n\\n\\n 7\\n Before the evidentiary hearing, Edmo tried to receive access to\\nfemale commissary items, such as women’s underwear. Most of her\\nrequests were denied. On the eve of the evidentiary hearing, IDOC\\n\\f EDMO V. CORIZON 21\\n\\n Neither the parties nor their experts dispute that Edmo\\nsuffers from gender dysphoria. That dysphoria causes Edmo\\nto feel “depressed,” “disgusting,” “tormented,” and\\n“hopeless.”\\n\\n To alleviate Edmo’s gender dysphoria, prison officials\\nhave, since 2012, provided hormone therapy. Edmo has\\nfollowed and complied with her hormone therapy regimen,\\nwhich helps alleviate her gender dysphoria to some extent.\\nThe hormones “clear[] [her] mind” and have resulted in\\nbreast growth, body fat redistribution, and changes in her\\nskin. Today, Edmo is hormonally confirmed, which means\\nthat she has the hormones and secondary sex characteristics\\n(characteristics, such as women’s breasts, that appear during\\npuberty but are not part of the reproductive system) of an\\nadult female. Edmo has gained the maximum physical\\nchanges associated with hormone treatment.\\n\\n Hormone therapy has not completely alleviated Edmo’s\\ngender dysphoria. Edmo continues to experience significant\\ndistress related to gender incongruence. Much of that\\ndistress is caused by her male genitalia. Edmo testified that\\nshe feels “depressed, embarrassed, [and] disgusted” by her\\nmale genitalia and that this is an “everyday reoccurring\\nthought.” Her medical records confirm her disgust, noting\\nrepeated efforts by Edmo to purchase underwear to keep, in\\nEdmo’s words, her “disgusting penis” out of sight.\\n\\n In addition to her gender dysphoria, Edmo suffers from\\nmajor depressive disorder with anxiety and drug and alcohol\\naddiction, although her addiction has been in remission\\n\\n\\namended its policy concerning the treatment of gender dysphoric\\nprisoners to increase transgender women’s access to female commissary\\nitems.\\n\\f22 EDMO V. CORIZON\\n\\nwhile incarcerated. Edmo has taken her prescribed\\nmedications for depression and anxiety. Prison officials\\nhave also provided Edmo mental health treatment to help her\\nwork through her serious underlying mental health issues\\nand a pre-incarceration history of trauma, abuse, and suicide\\nattempts. Edmo sees her psychiatrist when scheduled. But\\nEdmo does not see her treating clinician, Krina Stewart,\\nbecause Edmo does not believe Stewart is qualified to treat\\nher gender dysphoria. Edmo has attended group therapy\\nsessions inconsistently.\\n\\n In September 2015, Edmo attempted to castrate herself\\nfor the first time using a disposable razor blade. 8 Before\\ndoing so, she left a note to alert officials that she was not\\n“trying to commit suicide,” and was instead “only trying to\\nhelp [her]self.” Edmo did not complete the castration,\\nthough she continued to report thoughts of self-castration in\\nthe following months.\\n\\n On April 20, 2016, Dr. Eliason evaluated Edmo for GCS.\\nAt the time, IDOC’s policy concerning the treatment of\\ngender dysphoric prisoners provided that GCS “will not be\\nconsidered for individuals within [IDOC], unless determined\\nmedically necessary by” the treating physician. 9 Corizon’s\\npolicy does not mention GCS.\\n\\n In his evaluation, Dr. Eliason noted that Edmo reported\\nshe was “doing alright.” He also noted that Edmo had been\\non hormone replacement therapy for the last year and a half,\\nbut that she felt she needed more. He reported that Edmo\\n 8\\n She had previously reported thoughts of self-castration to\\nclinicians.\\n 9\\n IDOC revised its policy shortly before the evidentiary hearing, but\\nits revised policy contains functionally identical language.\\n\\f EDMO V. CORIZON 23\\n\\nhad stated that hormone replacement therapy helped\\nalleviate her gender dysphoria, but she remained frustrated\\nwith her male anatomy.\\n\\n Dr. Eliason indicated that Edmo appeared feminine in\\ndemeanor and interaction style. He also indicated that Edmo\\nhad previously attempted to “mutilate her genitalia” because\\nof the severity of her distress. Dr. Eliason later testified that,\\nat the time of his evaluation, he felt that Edmo’s gender\\ndysphoria “had risen to another level,” as evidenced by her\\nself-castration attempt.\\n\\n But Dr. Eliason also flagged that he had spoken to prison\\nstaff about Edmo’s behavior and they explained it was\\n“notable for animated affect and no observed distress.” He\\nsimilarly noted that he had personally observed Edmo and\\ndid not see significant dysphoria; instead, she “looked\\npleasant and had a good mood.”\\n\\n As to GCS, Dr. Eliason explained in his notes that while\\nmedical necessity for GCS is “not very well defined and is\\nconstantly shifting,” in his view, GCS would be medically\\nnecessary in at least three situations: (1) “congenital\\nmalformations or ambiguous genitalia,” (2) “severe and\\ndevastating dysphoria that is primarily due to genitals,” or\\n(3) “some type of medical problem in which endogenous\\nsexual hormones were causing severe physiological\\ndamage.” Dr. Eliason concluded that Edmo “does not meet\\nany of those . . . criteria” and, for that reason, GCS is not\\nmedically necessary for her.\\n\\n Dr. Eliason instead concluded that hormone therapy and\\nsupportive counseling suffice to treat Edmo’s gender\\ndysphoria for the time being, despite recognizing that Edmo\\nhad attempted self-castration on that regimen. Dr. Eliason\\n\\f24 EDMO V. CORIZON\\n\\nindicated that he would continue to monitor and assess\\nEdmo.\\n\\n Dr. Eliason staffed Edmo’s evaluation with Dr. Jeremy\\nStoddart, Dr. Murray Young, and Jeremy Clark, who all\\nagreed with his assessment. They did not observe Edmo;\\nrather, they agreed with Dr. Eliason’s recommended\\ntreatment as he presented it to them. The record is sparse on\\nthe qualifications of Dr. Stoddart and Dr. Young, but Clark\\nhas never personally treated anyone with gender dysphoria\\nand was not qualified under IDOC policy to assess whether\\nGCS would be appropriate for Edmo.\\n\\n Dr. Eliason also discussed his evaluation with IDOC’s\\nManagement and Treatment Committee (“MTC”), a multi-\\ndisciplinary team composed of medical providers, mental\\nhealth clinicians, IDOC’s Chief Psychologist, and prison\\nleadership. The MTC meets periodically to evaluate and\\naddress the unique medical, mental health, and housing\\nneeds of prisoners with gender dysphoria. The committee\\n“does not make any individual treatment decisions\\nregarding” treatment for inmates with gender dysphoria.\\n“Those determinations are made by the individual clinicians\\nor the medical staff employed by Corizon.” The MTC\\nagreed with Dr. Eliason’s assessment.\\n\\n Although not mentioned in his April 20, 2016 notes,\\nDr. Eliason testified at the evidentiary hearing that he\\nconsidered the WPATH Standards of Care when\\ndetermining Edmo’s treatment. Citing those standards,\\nDr. Eliason testified that he did not believe GCS was\\nappropriate for two reasons: (1) because mental health\\nissues separate from Edmo’s gender dysphoria were not\\n“fully in adequate control” and (2) because Edmo had not\\nlived in her identified gender role for 12 months outside of\\nprison. He explained that Edmo needed to experience\\n\\f EDMO V. CORIZON 25\\n\\n“living as a woman” around “her real social network – her\\nfamily and friends on the outside” so that she could\\n“determine whether or not she felt like that was her real\\nidentity.”\\n\\n Edmo was never evaluated for GCS again, but the MTC\\nconsidered her gender dysphoria and treatment plan during\\nlater meetings. The MTC continues to believe that GCS is\\nnot medically necessary or appropriate for Edmo.\\n\\n In December 2016, Edmo tried to castrate herself for the\\nsecond time. A medical note from the incident reports that\\nEdmo said she no longer wanted her testicles. Edmo\\nreported to medical providers that she was “feeling\\nangry/frustrated that [she] was not receiving the help desired\\nrelated to [her] gender dysphoria. Inmate Edmo’s actions\\nwere reported as a method to stop/cease testosterone\\nproduction in Edmo’s body. Edmo denied suicidal ideation\\n. . . .”\\n\\n Edmo’s second attempt was more successful than the\\nfirst. She was able to open her testicle sac with a razor blade\\nand remove one testicle. She abandoned her attempt,\\nhowever, when there was too much blood to continue. She\\nthen sought medical assistance and was transported to a\\nhospital, where her testicle was repaired. Edmo was\\nreceiving hormone therapy both times she attempted self-\\ncastration.\\n\\n Edmo testified that she was disappointed in herself for\\ncoming so close but failing to complete her self-castration\\nattempts. She also testified that she continues to actively\\nthink about self-castration. To avoid acting on those\\nthoughts and impulses, Edmo “self-medicat[es]” by cutting\\nher arms with a razor. She says that the physical pain helps\\n\\f26 EDMO V. CORIZON\\n\\nto ease the “emotional torment” and mental anguish her\\ngender dysphoria causes her.\\n\\n Edmo further testified that she expects GCS to help\\nalleviate some of her gender dysphoria. In particular, she\\ntestified that she expects GCS to help her avoid having “as\\nmuch depression about myself and my physical body. I\\ndon’t think I will be so anxious that people are always\\nknowing I’m different . . . .” Edmo recognizes, however,\\nthat GCS “is not a fix-all”: “[i]t’s not a magic operation. . . .\\nI’m still going to have to face the same stressors that we all\\nface in everyday life . . . .”\\n\\n C. Initiation of this Action\\n\\n Edmo filed a pro se complaint on April 6, 2017. She also\\nmoved for a temporary restraining order, a preliminary\\ninjunction, and the appointment of counsel.\\n\\n Edmo’s motion for appointment of counsel was granted\\nin part, and counsel for Edmo appeared in June and August\\n2017. Counsel withdrew Edmo’s pro se motion for\\npreliminary injunction shortly thereafter.\\n\\n On September 1, 2017, Edmo filed an amended\\ncomplaint asserting claims under 42 U.S.C. § 1983, the\\nEighth Amendment, the Fourteenth Amendment, the\\nAmericans with Disabilities Act, the Affordable Care Act,\\nand for common law negligence. She named as defendants\\nIDOC, Henry Atencio (Director of IDOC), Jeff Zmuda\\n(Deputy Director of IDOC), Howard Keith Yordy (former\\nWarden of ISCI), Dr. Richard Craig (Chief Psychologist at\\nISCI), Rona Siegert (Health Services Director at ISCI),\\nCorizon, Dr. Eliason, Dr. Young, and Dr. Catherine\\nWhinnery (Corizon employee).\\n\\f EDMO V. CORIZON 27\\n\\n Through counsel, Edmo filed a renewed motion for a\\npreliminary injunction on June 1, 2018. Among other relief,\\nEdmo sought an order requiring the State to provide her with\\na referral to a qualified surgeon and access to GCS.\\n\\n The State moved to extend the time to respond to Edmo’s\\nmotion. After a status conference, the district court set an\\nevidentiary hearing for October 10, 11, and 12, 2018. The\\ncourt permitted the parties to undertake four months of\\nextensive fact and expert discovery in preparation for the\\nhearing.\\n\\n D. The Evidentiary Hearing\\n\\n At the evidentiary hearing, each side had eight hours to\\npresent its case. The district court heard live testimony from\\nseven witnesses over three days. It also considered\\nthousands of pages of exhibits, including Edmo’s medical\\nrecords. With the parties’ agreement, the court also\\npermitted the State to submit declarations in lieu of live\\ntestimony and permitted Edmo to impeach the declarations\\nwith deposition testimony.\\n\\n At the outset of the hearing, the district court noted that\\n“[w]e’re here on a hearing for a temporary injunction,” but\\nit explained that “it’s hard for me to envision this hearing\\nbeing anything but a hearing on a final injunction[,] at least\\nas to” the injunctive relief ordering GCS. The court stated\\nthat it was unsure whether that made a difference, and it\\nasked the parties to address at some point whether the\\nhearing was for a preliminary injunction or a permanent\\ninjunction. Notably, the State did not do so.\\n\\n The district court heard testimony from three percipient\\nwitnesses: Edmo, Dr. Eliason (the Corizon physician), and\\nJeremy Clark (an IDOC clinician who did not meet IDOC’s\\n\\f28 EDMO V. CORIZON\\n\\ncriteria to assess Edmo for GCS). Their relevant testimony\\nis largely recounted above.\\n\\n It also heard testimony from four expert witnesses, two\\neach for Edmo and the State. Dr. Randi Ettner, Ph.D. in\\npsychology, testified first for Edmo. Dr. Ettner is one of the\\nauthors of the current (seventh) version of the WPATH\\nStandards of Care. She has been a WPATH member since\\n1993 and chairs its Institutionalized Persons Committee. Dr.\\nEttner has authored or edited many peer-reviewed\\npublications on the treatment of gender dysphoria and\\ntransgender health care more broadly, including the leading\\ntextbook used in medical schools on the subject. She also\\ntrains medical and mental health providers on treating people\\nwith gender dysphoria. Dr. Ettner has been retained as an\\nexpert witness on gender dysphoria and its treatment in\\nmany court cases, and she has been appointed as an\\nindependent expert by one federal court to evaluate an\\nincarcerated person for GCS.\\n\\n Dr. Ettner has evaluated, diagnosed, and treated between\\n2,500 and 3,000 individuals with gender dysphoria. She has\\nreferred about 300 people for GCS. She has also refused to\\nrecommend surgery for some patients who have requested it.\\nShe believes that not everyone who has gender dysphoria\\nneeds GCS. Dr. Ettner also has “[e]xtensive experience”\\ntreating and providing post-operative care for patients who\\nhave undergone GCS.\\n\\n Dr. Ettner has assessed approximately 30 incarcerated\\nindividuals with gender dysphoria for GCS and other\\nmedical care, but she has not treated incarcerated patients.\\nShe has not worked in a prison and she is not a Certified\\nCorrectional Healthcare Professional.\\n\\f EDMO V. CORIZON 29\\n\\n Based on her evaluation of Edmo and a review of\\nEdmo’s medical records, Dr. Ettner diagnosed Edmo with\\ngender dysphoria, depressive disorder, anxiety, and suicidal\\nideation. In Dr. Ettner’s opinion, GCS is medically\\nnecessary for Edmo and should be immediately performed.\\nShe explained that most patients with gender dysphoria do\\nnot require GCS, but Edmo requires it because hormone\\ntherapy has been inadequate for her and Edmo has attempted\\nto remove her own testicles. Dr. Ettner further explained that\\nGCS would give Edmo congruent genitalia, eliminating the\\nsevere distress Edmo experiences due to her male anatomy.\\n\\n Dr. Ettner further opined that Edmo meets the WPATH\\ncriteria for GCS. She explained that Edmo has “persistent\\nand well-documented long-standing gender dysphoria”;\\nEdmo “has no thought disorders and no impaired reality\\ntesting”; Edmo is the age of majority in this country;\\nalthough Edmo has depression and anxiety, those conditions\\ndo not “impair her ability to undergo surgery” because they\\nare “as controlled as [they] can be”; Edmo has had six years\\nof hormone therapy; and Edmo has lived for more than one\\nyear “as a woman to the best of her ability in a male prison.”\\n\\n More specifically, as to the fourth criterion, Dr. Ettner\\nopined that Edmo does not have mental health concerns that\\nwould preclude GCS. She explained that Edmo’s depression\\nand anxiety are as “controlled as can be” because Edmo “is\\ntaking the maximum amount of medication that controls\\ndepression.” Dr. Ettner noted that Edmo has complied with\\ntaking her prescribed medications and that psychotherapy is\\nnot “a precondition for surgery” under the WPATH\\nStandards of Care. She also flagged that Edmo has the\\ncapacity to comply with her postsurgical treatment, as\\nevidenced by her compliance with her hormone therapy to\\ndate.\\n\\f30 EDMO V. CORIZON\\n\\n As to the clinical significance of Edmo’s self-castration\\nattempts and cutting behaviors, Dr. Ettner explained that\\nneither behavior indicates that Edmo has inadequately\\ncontrolled mental health concerns. Rather, those behaviors\\nindicate “the need for treatment for gender dysphoria.”\\nDr. Ettner explained that\\n\\n when an individual who is not psychotic or\\n delusional attempts what we call surgical\\n self-treatment – because we don’t regard\\n removal of the testicles or attempted removal\\n of the testicles as either mutilation or self-\\n harm – we regard it as an intentional attempt\\n to remove the target organ that produces\\n testosterone, which, in fact, is the cure for\\n gender dysphoria.\\n\\nIn Dr. Ettner’s opinion, Edmo’s depression and anxiety “will\\nbe attenuated post surgery.”\\n\\n Dr. Ettner opined that Edmo satisfies the sixth criterion\\nbecause she has lived “as a woman to the best of her ability\\nin a male prison.” Dr. Ettner based her opinion on Edmo’s\\n“appearance . . . , her disciplinary records, which indicated\\nthat she had attempted to wear her hair in a feminine\\nhairstyle and to wear makeup even though that was against\\nthe rules and she was – received some sort of disciplinary\\naction for that, and her – the way that she was receiving\\nfemale undergarments and had developed the stigma of\\nfemininity, the secondary sex characteristics, breast\\ndevelopment, et cetera.”\\n\\n Dr. Ettner opined that if Edmo does not receive GCS,\\n“[t]he risks would be, as typical in inadequately treated or\\nuntreated gender dysphoria, either surgical self-treatment,\\nemotional decompensation, or suicide.” Dr. Ettner\\n\\f EDMO V. CORIZON 31\\n\\nexplained that Edmo “is at particular risk of suicide given\\nthat she has a high degree of suicide ideation.” If, on the\\nother hand, Edmo receives surgery, Dr. Ettner opined that\\n\\n [i]t would eliminate the gender dysphoria. It\\n would provide a level of wellbeing that she\\n hasn’t had previously. It would eliminate\\n 80 percent of the testosterone in her body,\\n necessitating a lower dose of hormones going\\n forward, which would be particularly helpful\\n given that she has elevated liver enzymes.\\n And it would, I believe, eliminate much of the\\n depression and the attendant symptoms that\\n she is experiencing.\\n\\n Dr. Ryan Gorton, M.D., also testified for Edmo.\\nDr. Gorton is an emergency medicine physician. He also\\nworks pro bono at a clinic serving uninsured patients or those\\nwith Medicare or Medicaid. Many of those patients have\\nmental health conditions or have been in prison. He has\\npublished peer-reviewed articles on the treatment of gender\\ndysphoria, and he has been qualified as an expert witness in\\ncases involving transgender health care. Dr. Gorton also\\nprovides training on transgender health care issues to many\\ngroups, is a member of WPATH, and serves on WPATH’s\\nTransgender Medicine and Research Committee and its\\nInstitutionalized Persons Committee.\\n\\n Dr. Gorton has been the primary care physician for about\\n400 patients with gender dysphoria. At the time of the\\nevidentiary hearing, Dr. Gorton was treating approximately\\n100 patients with gender dysphoria. Dr. Gorton has assessed\\npatients for gender dysphoria, initiated and monitored\\nhormone treatment, referred patients for mental health\\ntreatment, and determined the appropriateness of GCS. At\\n\\f32 EDMO V. CORIZON\\n\\nthe time of the evidentiary hearing, Dr. Gorton was\\nproviding follow-up care for about 30 patients who had\\nvaginoplasty. Dr. Gorton has no experience treating\\ntransgender inmates and is not a Certified Correctional\\nHealthcare Professional.\\n\\n Based on his review of Edmo’s medical records and his\\nin-person evaluation of Edmo, Dr. Gorton opined that GCS\\nis medically necessary for Edmo and that she meets the\\nWPATH criteria for GCS. He explained that Edmo has\\n“persistent well-documented gender dysphoria,” as shown in\\nher prison medical records; she has the capacity “to make a\\nfully informed decision and to consent for treatment”\\nbecause “she didn’t seem at all impaired in her decision-\\nmaking capacity”; she is the age of majority; she has\\ndepression and anxiety, “but they are not to a level that\\nwould preclude her getting [GCS]”; she had 12 consecutive\\nmonths of hormone therapy; and she has been living in her\\n“target gender role . . . despite an environment that’s very\\nhostile to that and some negative consequences that she has\\nexperienced because of that.”\\n\\n Dr. Gorton further opined that if Edmo “is not provided\\nsurgery, there is a very substantial chance she will try to\\nattempt self-surgery again. And that’s especially worrisome\\ngiven her attempts have been progressive. . . . So I think she\\nmight be successful” on her next attempt. He predicted that\\nthere is little chance that Edmo’s gender dysphoria will\\nimprove without surgery. Conversely, Dr. Gorton\\nanticipated that Edmo is unlikely to regret surgery because\\n“her gender dysphoria is very genital-focused” and regret\\nrates among GCS patients are very low.\\n\\n Dr. Gorton also opined that Edmo’s self-castration\\nattempts demonstrate “that she has severe genital-focused\\ngender dysphoria and that she is not getting the medically\\n\\f EDMO V. CORIZON 33\\n\\nnecessary treatment to alleviate that.” He elaborated that\\nEdmo’s depression and anxiety are not driving Edmo’s self-\\ncastration attempts: “there [are] a lot of people with\\ndepression and anxiety who don’t remove their testicles.”\\n\\n Finally, Dr. Gorton criticized Dr. Eliason’s evaluation of\\nEdmo. He explained that he disagreed with Dr. Eliason’s\\nconclusion that Edmo does not need GCS and he also\\ndisagreed with the three “criteria” Dr. Eliason gave for when\\nGCS would be necessary. Dr. Gorton criticized Dr.\\nEliason’s first criterion—that GCS could be needed where\\nthere is “congenital malformation or ambiguous genitalia”—\\nbecause that situation “isn’t even germane to transgender\\npeople”; rather, it relates to “people with intersex\\nconditions.” As to the second criterion—that GCS could be\\nneeded when a patient is suffering from “severe and\\ndevastating gender dysphoria that is primarily due to\\ngenitals”—Dr. Gorton pointed out that the WPATH\\nStandards of Care for surgery require only “clear and\\nsignificant dysphoria.” And even applying Dr. Eliason’s\\nhigher bar, Dr. Gorton explained that Edmo would still\\nqualify for GCS because she has twice attempted self-\\ncastration, demonstrating “severe genital-focused\\ndysphoria.” Finally, Dr. Gorton characterized Dr. Eliason’s\\nthird criterion—that GCS could be needed in situations when\\n“endogenous sexual hormones were causing severe\\nphysiological damage”—as “bizarre.” Dr. Gorton could not\\nconjure “a clinical circumstance where that would be the\\ncase that your hormones that your body produces are\\nattacking you . . . . I just don’t understand what [Dr. Eliason]\\nis talking about there.”\\n\\n Dr. Keelin Garvey, M.D., testified for the State.\\nDr. Garvey is a psychiatrist and Certified Correctional\\nHealthcare Professional. As the former Chief Psychiatrist of\\n\\f34 EDMO V. CORIZON\\n\\nthe Massachusetts Department of Corrections, Dr. Garvey\\nchaired the Gender Dysphoria Treatment Committee. She\\ndirectly treated a “couple of patients” with gender dysphoria\\nearlier in her career as Deputy Medical Director, but she has\\nnot done so in recent years. Prior to evaluating Edmo, Dr.\\nGarvey had never evaluated a patient in person to determine\\nwhether that person needed GCS. Dr. Garvey has never\\nrecommended a patient for GCS, and she has not done\\nfollow-up care with a person who has received GCS.\\n\\n Based on her evaluation of Edmo and a review of\\nEdmo’s medical records, Dr. Garvey diagnosed Edmo with\\ngender dysphoria, major depressive disorder, alcohol use\\ndisorder, stimulant use disorder, and opioid use disorder.\\nShe explained that the latter three are in remission.\\n\\n Relying on the WPATH Standards of Care, Dr. Garvey\\nopined that GCS is not medically necessary for Edmo. 10 Dr.\\nGarvey first explained that Edmo does not meet the first\\nWPATH Standards of Care criterion—“persistent, well\\ndocumented gender dysphoria”—because of a lack of\\nevidence in pre-incarceration medical records that Edmo\\npresented as female before her time in prison. Dr. Garvey\\nacknowledged, however, that Edmo has been presenting as\\nfemale since 2012 and that she has been diagnosed with\\ngender dysphoria since that time.\\n\\n Dr. Garvey then explained that Edmo does not meet the\\nfourth criterion—“medical/mental health concerns must be\\nwell controlled”—because Edmo “is actively self-injuring.”\\nDr. Garvey elaborated that “self-injury in any form is never\\n\\n 10\\n Dr. Garvey testified that she relies on the WPATH Standards of\\nCare and the NCCHC guidelines adopting those standards when treating\\ninmates with gender dysphoria.\\n\\f EDMO V. CORIZON 35\\n\\nconsidered a healthy or productive coping mechanism” and\\nthat she would like to see Edmo “develop further coping\\nskills that she would be able to use following surgery so that\\nshe is not engaging in self-injury after surgery.” Dr.\\nGarvey’s concern is that GCS is a “stressful undertaking”\\nand Edmo lacks “effective coping strategies” to deal with the\\nstress.\\n\\n Finally, Dr. Garvey testified that Edmo does not meet the\\nsixth criterion—“12 continuous months of living in a gender\\nrole that is congruent with gender identity”—because Edmo\\nhas not presented as female outside of prison and “there [are]\\nchallenges to using her time in a men’s prison as this real-\\nlife experience because it doesn’t offer her the opportunity\\nto actually experience all those things she is going to go\\nthrough on the outside.”\\n\\n Dr. Joel Andrade, Ph.D. in social work, also testified for\\nthe State. He is a licensed clinical social worker and is a\\nCertified Correctional Healthcare Professional with an\\nemphasis in mental health. Dr. Andrade has over a decade\\nof experience providing and supervising the provision of\\ncorrectional mental health care, including directing and\\noverseeing the treatment of inmates diagnosed with gender\\ndysphoria in the custody of the Massachusetts Department\\nof Corrections in his roles as clinical director, chair of the\\nGender Dysphoria Supervision Group, and member of the\\nGender Dysphoria Treatment Committee.\\n\\n As a member of the Gender Dysphoria Treatment\\nCommittee, Dr. Andrade recommended GCS for two\\ninmates. But the recommendations were contingent on the\\ninmates living in a women’s prison for approximately\\n12 months before the surgery. The Massachusetts\\nDepartment of Corrections, like IDOC, houses prisoners\\n\\f36 EDMO V. CORIZON\\n\\naccording to their genitals, so the inmates had not been\\nmoved (nor had their surgery occurred).\\n\\n Dr. Andrade has never directly treated patients with\\ngender dysphoria, nor has he been a treating clinician for a\\npatient who has had GCS. His “experience with gender\\ndysphoria comes almost exclusively from [his] participation\\non the Massachusetts Department of Corrections[’] Gender\\nDysphoria Treatment Committee and Supervision Group.”\\nDr. Andrade did not qualify, under the IDOC gender\\ndysphoria policy in effect at the time of his assessment of\\nEdmo, to assess a person for GCS because he is neither a\\npsychologist nor a physician.\\n\\n Based on his evaluation of Edmo and a review of her\\nmedical records, Dr. Andrade diagnosed Edmo with “major\\ndepressive disorder, recurrent, in partial remission,”\\n“generalized anxiety disorder,” “alcohol use disorder,\\nsevere,” and gender dysphoria. Dr. Andrade also diagnosed\\nEdmo with borderline personality disorder. The district\\ncourt did not credit this diagnosis, however, because no other\\nperson (including the State’s other expert, Dr. Garvey) has\\never diagnosed Edmo with borderline personality disorder\\nand Dr. Andrade was unable to identify his criteria for this\\ndiagnosis. Edmo, 358 F. Supp. 3d at 1120. The record\\namply supports the district court’s finding in this respect.\\n\\n Dr. Andrade opined that Edmo does not meet the\\nWPATH criteria for GCS. He explained that, based on his\\nreview of Edmo’s pre-incarceration records, Edmo did not\\npresent as female or discuss her gender dysphoria before\\nincarceration. Dr. Andrade testified that he would like to see\\nEdmo live as female outside of a correctional setting before\\nreceiving GCS, or, at the least, live in a women’s prison first.\\nIDOC, however, houses prisoners according to their\\ngenitals. Dr. Andrade also explained that Edmo needs to\\n\\f EDMO V. CORIZON 37\\n\\nwork through some of her trauma, particularly sexual abuse\\nthat she suffered, and other mental health concerns before\\nreceiving surgery. Dr. Andrade opined that Edmo’s mental\\nhealth issues will not be cured by GCS.\\n\\n At the close of the hearing, the district court reiterated\\nthat it was unsure “how we can hear [Edmo’s request for\\nGCS] on a preliminary injunction. . . . [I]f I order it, then it’s\\ndone.” The court further suggested that the request for GCS\\ncould “only be resolved in a final hearing” and noted that it\\nhad, in effect, “treated this hearing as [a] final hearing on the\\nissue.”\\n\\n The court, as it had done at the outset of the hearing,\\nasked the parties to address whether the hearing was for a\\npreliminary or permanent injunction. In response, Edmo\\ncontended that the court could order GCS in a preliminary\\ninjunction. The State did not address the court’s question. It\\ninstead contended that the standard for a mandatory\\ninjunction—which can be preliminary or permanent—\\nshould apply.\\n\\n E. The District Court’s Decision\\n\\n The district court rendered its decision on December 13,\\n2018. After recounting the evidence and making extensive\\nfactual findings, the district court began its analysis by\\nnoting that it was unsure whether the standard for a\\npreliminary injunction or the standard for a permanent\\ninjunction applied. The court noted that “the nature of the\\nrelief requested in this case, coupled with the extensive\\nevidence presented by the parties over a 3-day evidentiary\\nhearing, [may have] effectively converted these proceedings\\ninto a final trial on the merits of the plaintiff’s request for\\npermanent injunctive relief.” Edmo, 358 F. Supp. 3d at 1122\\nn.1. It also indicated that “both parties appear to have treated\\n\\f38 EDMO V. CORIZON\\n\\nthe evidentiary hearing” as a final trial on the merits. Id. The\\ndistrict court explained that the difference was immaterial,\\nhowever, because Edmo was entitled to relief under either\\nstandard. Id.\\n\\n On the merits, the district court concluded that Edmo had\\nestablished her Eighth Amendment claim. The district court\\nfirst held that Edmo suffers from gender dysphoria, which is\\nundisputedly “a serious medical condition.” Id. at 1124.\\n\\n It then concluded that GCS is medically necessary to\\ntreat Edmo’s gender dysphoria. See id. at 1124–26. In a\\ncarefully considered, 45-page opinion, the district court\\nspecifically found “credible the testimony of Plaintiff’s\\nexperts Drs. Ettner and Gorton, who have extensive personal\\nexperience treating individuals with gender dysphoria both\\nbefore and after receiving gender confirmation surgery,” and\\nwho opined that GCS was medically necessary. Id. at 1125.\\nThe court rejected the contrary opinions of the State’s\\nexperts because “neither Dr. Garvey nor Dr. Andrade has\\nany direct experience with patients receiving gender\\nconfirmation surgery or assessing patients for the medical\\nnecessity of gender confirmation surgery,” and neither of the\\nState’s experts had meaningful “experience treating patients\\nwith gender dysphoria other than assessing them for the\\nexistence of the condition.” Id. The district court also noted\\nthat the State’s “experts appear to misrepresent the WPATH\\nStandards of Care by concluding that Ms. Edmo, despite\\npresenting as female since her incarceration in 2012, cannot\\nsatisfy the WPATH criteria because she has not presented as\\nfemale outside of the prison setting.” Id. As the district\\ncourt noted, “there is no requirement in the WPATH\\nStandards of Care that a patient live for twelve months in his\\nor her gender role outside of prison before becoming eligible\\nfor” GCS. Id. (quotation omitted).\\n\\f EDMO V. CORIZON 39\\n\\n Finally, the district court explained that the State was\\ndeliberately indifferent to Edmo’s gender dysphoria because\\nit “fail[ed] to provide her with available treatment that is\\ngenerally accepted in the field as safe and effective, despite\\nher actual harm and ongoing risk of future harm including\\nself-castration attempts, cutting, and suicidal ideation.” Id.\\nat 1126–27. The district court also stated that the evidence\\n“suggest[ed] that Ms. Edmo has not been provided gender\\nconfirmation surgery because Corizon and IDOC have a de\\nfacto policy or practice of refusing this treatment for gender\\ndysphoria to prisoners,” which amounts to deliberate\\nindifference. Id. at 1127.\\n\\n After analyzing the merits, the district court concluded\\nthat Edmo satisfied the other prerequisites to injunctive\\nrelief. Id. at 1127–28. The district court found that, given\\nEdmo’s continuing emotional distress and self-castration\\nattempts, “Edmo is at serious risk of life-threatening self-\\nharm” if she does not receive GCS. Id. at 1128. The State,\\non the other hand, had not shown that it would be harmed if\\nordered to provide GCS, so the equities favored Edmo. Id.\\n\\n Having concluded that Edmo was entitled to an\\ninjunction, the court ordered the State “to provide Plaintiff\\nwith adequate medical care, including gender confirmation\\nsurgery.” Id. at 1129. It ordered the State to “take all actions\\nreasonably necessary to provide Ms. Edmo gender\\nconfirmation surgery as promptly as possible and no later\\nthan six months from the date of this order.” Id.\\n\\n F. Appellate Proceedings\\n\\n The State filed timely notices of appeal on January 9,\\n2019. It also asked the district court to stay its order pending\\nappeal. The district court denied the State’s motion on\\nMarch 4.\\n\\f40 EDMO V. CORIZON\\n\\n The State then filed in this court a motion to stay pending\\nappeal. A motions panel granted that motion. Edmo\\nsubsequently moved to amend the stay to allow her to\\nundergo a previously scheduled pre-surgery consultation.\\nThe motions panel granted that motion and amended the\\nstay.\\n\\n On April 3, the State filed an “urgent motion” to dismiss\\nthis appeal as moot. We indicated on April 5 that our court\\nwould consider that motion with the merits, not on an urgent\\nbasis.\\n\\n After hearing oral argument on May 16, we ordered a\\nlimited remand to the district court to clarify three points.\\nRelevant here, we asked the district court to clarify whether\\nit granted Edmo a permanent injunction in its December 13,\\n2018 order. The district court clarified that it “granted\\npermanent injunctive relief.” Edmo v. Idaho Dep’t of Corr.,\\nNo. 1:17-CV-00151-BLW, 2019 WL 2319527, at *2 (D.\\nIdaho May 31, 2019). We also asked the district court to\\nclarify whether it had concluded that Edmo had succeeded\\non the merits of her Eighth Amendment claim. The district\\ncourt responded that it had. Id.\\n\\n Having received the district court’s response to our\\nlimited remand order, we proceed to the issues on appeal.\\nThe State challenges the district court’s grant of injunctive\\nrelief to Edmo on multiple grounds. It contends that this\\nappeal is moot because the injunction did not comply with\\nthe PLRA and has, for that reason, automatically expired. It\\ncontends that the decision not to provide GCS to Edmo\\nreflects a difference of prudent medical opinion and cannot\\nsupport an Eighth Amendment claim. It contends that Edmo\\nwill not be irreparably harmed absent an injunction. It\\ncontends that the injunction is overbroad. Finally, it\\ncontends that, to the extent the district court converted the\\n\\f EDMO V. CORIZON 41\\n\\nevidentiary hearing into a final trial on the merits of Edmo’s\\nrequest for GCS, it was provided inadequate notice and the\\ncourt violated its right to a jury trial.\\n\\n II. Mootness\\n\\n “We first address, as we must, the question of mootness\\n. . . .” Shell Offshore Inc. v. Greenpeace, Inc., 815 F.3d 623,\\n628 (9th Cir. 2016). An appeal is moot “[w]hen events\\nchange such that the appellate court can no longer grant ‘any\\neffectual relief whatever to the prevailing party.’” Id.\\n(quoting City of Erie v. Pap’s A.M., 529 U.S. 277, 287\\n(2000)). In those circumstances, we “lack[] jurisdiction and\\nmust dismiss the appeal.” Id.\\n\\n The State contends that the injunction does not comply\\nwith provisions of the PLRA and, for that reason, has\\nautomatically expired under the terms of the statute.\\nRelevant here, the PLRA provides that a\\n\\n court shall not grant or approve any\\n prospective relief unless the court finds that\\n such relief is narrowly drawn, extends no\\n further than necessary to correct the violation\\n of the Federal right, and is the least intrusive\\n means necessary to correct the violation of\\n the Federal right. The court shall give\\n substantial weight to any adverse impact on\\n public safety or the operation of a criminal\\n justice system caused by the relief.\\n\\n18 U.S.C. § 3626(a)(1)(A). Courts often refer to this\\nprovision as the “need-narrowness-intrusiveness” inquiry.\\nGraves v. Arpaio, 623 F.3d 1043, 1048 n.1 (9th Cir. 2010)\\n(per curiam) (quoting Pierce v. County of Orange, 526 F.3d\\n1190, 1205 (9th Cir. 2008)). The PLRA further provides that\\n\\f42 EDMO V. CORIZON\\n\\nany “[p]reliminary injunctive relief shall automatically\\nexpire on the date that is 90 days after its entry, unless the\\ncourt makes the findings required under subsection (a)(1)\\n[quoted above] for the entry of prospective relief and makes\\nthe order final before the expiration of the 90-day period.”\\n18 U.S.C. § 3626(a)(2).\\n\\n The State contends that the district court did not make\\nthe PLRA’s requisite need-narrowness-intrusiveness\\nfindings or make its order final within 90 days, causing the\\ninjunction to expire under 18 U.S.C. § 3626(a)(2).\\nGenerally, the expiration of an injunction challenged on\\nappeal moots the appeal. See Kitlutsisti v. ARCO Alaska,\\nInc., 782 F.2d 800, 801 (9th Cir. 1986); see also United\\nStates v. Sec’y, Fla. Dep’t of Corr., 778 F.3d 1223, 1228–29\\n(11th Cir. 2015). The State asserts separate, albeit\\noverlapping, contentions in their motion to dismiss this\\nappeal and in their briefing. We reject those arguments.\\n\\n A. Need-Narrowness-Intrusiveness Findings\\n\\n The State first contends that the district court did not\\nmake the PLRA’s need-narrowness-intrusiveness findings,\\ncausing the injunction to automatically expire and mooting\\nthis appeal. 11 As we have explained in prior decisions, the\\nPLRA “has not substantially changed the threshold findings\\nand standards required to justify an injunction.” Gomez v.\\nVernon, 255 F.3d 1118, 1129 (9th Cir. 2001). When\\n“determining the appropriateness of the relief ordered,”\\nappellate “courts must do what they have always done”:\\n\\n 11\\n We question whether the State’s need-narrowness-intrusiveness\\nchallenge, properly understood, implicates mootness. But because the\\nresult is the same, we accept the State’s framing for purposes of our\\nanalysis.\\n\\f EDMO V. CORIZON 43\\n\\n“consider the order as a whole.” Armstrong v.\\nSchwarzenegger, 622 F.3d 1058, 1070 (9th Cir. 2010).\\nDistrict courts must make need-narrowness-intrusiveness\\n“findings sufficient to allow a ‘clear understanding’ of the\\nruling,” but they need not “make such findings on a\\nparagraph by paragraph, or even sentence by sentence,\\nbasis.” Id. (quotation omitted). “What is important, and\\nwhat the PLRA requires, is a finding that the set of reforms\\nbeing ordered—the ‘relief’—corrects the violations of\\nprisoners’ rights with the minimal impact possible on\\ndefendants’ discretion over their policies and procedures.”\\nId.\\n\\n Here, the district court made the necessary need-\\nnarrowness-intrusiveness findings. At the start of its\\nDecember 13, 2018 order, the district court explained that\\nany injunction must meet the PLRA’s need-narrowness-\\nintrusiveness requirement. See Edmo, 358 F. Supp. 3d at\\n1122. The district court then explained how the relief being\\nordered, GCS, “corrects the violations of” Edmo’s rights.\\nSee Armstrong, 622 F.3d at 1071. Specifically, the district\\ncourt explained that GCS is medically necessary to alleviate\\nEdmo’s gender dysphoria and that the State’s denial of GCS\\namounts to deliberate indifference in violation of the Eighth\\nAmendment. See Edmo, 358 F. Supp. 3d at 1116–21, 1123–\\n27, 1129. The district court limited the relief ordered to have\\n“the minimal impact possible on [the State’s] discretion over\\ntheir policies and procedures.” See Armstrong, 622 F.3d at\\n1071. Specifically, the district court limited the relief to\\n“actions reasonably necessary” to provide GCS, cautioned\\nthat its conclusion is based on “the unique facts and\\ncircumstances presented” by Edmo, and noted that its\\n“decision is not intended, and should not be construed, as a\\ngeneral finding that all inmates suffering from gender\\ndysphoria are entitled to [GCS].” Edmo, 358 F. Supp. 3d at\\n\\f44 EDMO V. CORIZON\\n\\n1110, 1129. Finally, the district court rejected the notion that\\ninjunctive relief would have “any adverse impact on public\\nsafety or the operation of a criminal justice system.”\\n18 U.S.C. § 3626(a)(1)(A). It explained that the State had\\n“made no showing that an order requiring them to provide”\\nGCS to Edmo “causes them injury.” Edmo, 358 F. Supp. 3d\\nat 1128. The district court’s order, considered as a whole,\\nmade all the findings required by 18 U.S.C. § 3626(a)(1)(A)\\nand our precedent. See Armstrong, 622 F.3d at 1070.\\n\\n B. Finality\\n\\n The State next argues that the injunction has\\nautomatically expired under the PLRA because the district\\ncourt did not make its order “final” within 90 days of\\nentering injunctive relief. See 18 U.S.C. § 3626(a)(2); see\\nalso Sec’y, Fla. Dep’t of Corr., 778 F.3d at 1228–29\\n(holding that an appeal of a preliminary injunction was moot\\nbecause the district court “did not issue an order finalizing\\nits [preliminary-injunction] order,” and “[a]s a result, the\\npreliminary injunction expired by operation of law” 90 days\\nlater). The PLRA provision cited by the State applies to\\npreliminary injunctive relief, not permanent injunctive\\nrelief. See 18 U.S.C. § 3626(a)(2). The permanent\\ninjunction that the district court entered has not expired. See\\nEdmo, 358 F. Supp. 3d at 1122 n.1 (concluding that Edmo is\\n“entitled to relief” under the permanent injunction standard);\\nsee also Edmo, 2019 WL 2319527, at *2 (clarifying on\\nlimited remand that the district court granted Edmo a\\npermanent injunction). It remains in place, albeit stayed.\\n\\f EDMO V. CORIZON 45\\n\\n There is a live controversy on appeal. 12 We accordingly\\nDENY the State’s motion to dismiss and proceed to the\\nmerits of the appeal.\\n\\n III. Challenges to the District Court’s Grant of\\n Injunctive Relief\\n\\n An injunction is an “extraordinary remedy never\\nawarded as of right.” Winter v. Nat. Res. Def. Council, Inc.,\\n555 U.S. 7, 24 (2008). “To be entitled to a permanent\\ninjunction, a plaintiff must demonstrate: (1) actual success\\non the merits; (2) that it has suffered an irreparable injury;\\n(3) that remedies available at law are inadequate; (4) that the\\nbalance of hardships justify a remedy in equity; and (5) that\\nthe public interest would not be disserved by a permanent\\ninjunction.” 13 Indep. Training & Apprenticeship Program\\n\\n\\n\\n 12\\n Even construed as a preliminary injunction, the district court’s\\nDecember 13, 2018 order is not moot. On May 31, 2019, the district\\ncourt, incorporating its previous findings, renewed the injunction. See\\nEdmo, 2019 WL 2319527, at *2. Because the district court renewed the\\ninjunction, we can consider its merits. See Mayweathers v. Newland,\\n258 F.3d 930, 935–36 (9th Cir. 2001) (holding that district courts may\\nrenew preliminary injunctions under the PLRA while an appeal is\\npending, and considering the merits of the renewed injunction). And we\\nhave jurisdiction under 28 U.S.C. § 1292(a)(1) regardless of whether the\\ndistrict court’s order is considered a preliminary or permanent\\ninjunction. See Hendricks v. Bank of Am., N.A., 408 F.3d 1127, 1131\\n(9th Cir. 2005) (preliminary injunction); TransWorld Airlines, Inc. v.\\nAm. Coupon Exch., Inc., 913 F.2d 676, 680–81 (9th Cir. 1990)\\n(permanent injunction where the “district court retained jurisdiction to\\ndetermine damages” and to adjudicate a separate claim).\\n 13\\n We agree with the State that the injunction is mandatory, as\\nopposed to prohibitory, because it requires the State to act. Based on that\\ndistinction, the State argues that Edmo must satisfy a higher burden of\\n\\f46 EDMO V. CORIZON\\n\\nv. Cal. Dep’t of Indus. Relations, 730 F.3d 1024, 1032 (9th\\nCir. 2013) (citing eBay Inc. v. MercExch., L.L.C., 547 U.S.\\n388, 391 (2006)).\\n\\n We review for abuse of discretion the district court’s\\ndecision to grant a permanent injunction. Ariz. Dream Act\\nCoal. v. Brewer, 855 F.3d 957, 965 (9th Cir. 2017). We\\n\\n\\nproof to be entitled to injunctive relief, and that the district court failed\\nto hold Edmo to that burden. On that point, we disagree.\\n\\n The State errs by relying on cases that concern mandatory\\npreliminary injunctions. Because mandatory preliminary injunctions go\\n“well beyond simply maintaining the status quo [p]endente lite,” they are\\n“particularly disfavored” and “are not issued in doubtful cases.” Marlyn\\nNutraceuticals, Inc. v. Mucos Pharma GmbH & Co., 571 F.3d 873, 879\\n(9th Cir. 2009) (alteration in original) (quoting Anderson v. United\\nStates, 612 F.2d 1112, 1114–15 (9th Cir. 1980)). The calculus is\\ndifferent in the context of permanent injunctions. A plaintiff must show\\nactual success on the merits, see Amoco Prod. Co. v. Village of Gambell,\\n480 U.S. 531, 546 n.12 (1987), so there is no concern that a mandatory\\npermanent injunction will upset the status quo only for a later trial on the\\nmerits to show that the plaintiff was not entitled to equitable relief. As a\\nresult, a plaintiff need not show that “extreme or very serious damage\\nwill result,” as is required for mandatory preliminary injunctions.\\n\\n As we have explained, the district court granted Edmo injunctive\\nrelief under both the preliminary and permanent injunction standards.\\nSee Edmo, 358 F. Supp. 3d at 1122 n.1; see also Edmo, 2019 WL\\n2319527, at *2. Because the standard for granting permanent injunctive\\nrelief is higher (in that it requires actual success on the merits) and the\\nState contends in its opening brief that we should review the injunction\\nas a permanent injunction, we consider whether the district court erred\\nin granting Edmo permanent injunctive relief. But we would also affirm\\nunder the mandatory preliminary injunction standard, because the district\\ncourt correctly applied the proper standard for mandatory preliminary\\ninjunctive relief, and not the lower standard for prohibitory preliminary\\ninjunctions. See Edmo, 358 F. Supp. 3d at 1122, 1128.\\n\\f EDMO V. CORIZON 47\\n\\nreview “any determination underlying the grant of an\\ninjunction by the standard that applies to that\\ndetermination.” Ting v. AT&T, 319 F.3d 1126, 1134–35 (9th\\nCir. 2003). Accordingly, the district court’s factual findings\\non Edmo’s Eighth Amendment claim are reviewed for clear\\nerror. See Graves, 623 F.3d at 1048. Clear error exists if the\\nfinding is “illogical, implausible, or without support in\\ninferences that may be drawn from the facts in the record.”\\nLa Quinta Worldwide LLC v. Q.R.T.M., S.A. de C.V.,\\n762 F.3d 867, 879 (9th Cir. 2014) (quoting Herb Reed\\nEnters., LLC v. Florida Entm’t Mgmt., Inc., 736 F.3d 1239,\\n1247 (9th Cir. 2013)). We review de novo the district court’s\\n“conclusion that the facts . . . demonstrate an Eighth\\nAmendment violation.” Hallett v. Morgan, 296 F.3d 732,\\n744 (9th Cir. 2002).\\n\\n The State contends that the district court erred in\\ngranting an injunction because (1) Edmo’s Eighth\\nAmendment claim fails and (2) Edmo has not shown that she\\nwill suffer irreparable injury in the absence of an\\ninjunction. 14 We disagree. We hold, based on the district\\ncourt’s factual findings, that Edmo established her Eighth\\nAmendment claim and that she will suffer irreparable\\nharm—in the form of ongoing mental anguish and possible\\nphysical harm—if GCS is not provided.\\n\\n A. The Merits of Edmo’s Eighth Amendment Claim\\n\\n “[D]eliberate indifference to serious medical needs of\\nprisoners” violates the Eighth Amendment. Estelle,\\n429 U.S. at 104. Because “society takes from prisoners the\\nmeans to provide for their own needs,” Brown, 563 U.S.\\n\\n 14\\n Because the State does not contest the other injunction factors, we\\ndo not address them.\\n\\f48 EDMO V. CORIZON\\n\\nat 510, the government has an “obligation to provide medical\\ncare for those whom it is punishing by incarceration,”\\nEstelle, 429 U.S. at 103.\\n\\n To establish a claim of inadequate medical care, a\\nprisoner must first “show a ‘serious medical need’ by\\ndemonstrating that ‘failure to treat a prisoner’s condition\\ncould result in further significant injury or the ‘unnecessary\\nand wanton infliction of pain.’” Jett v. Penner, 439 F.3d\\n1091, 1096 (9th Cir. 2006) (quoting McGuckin v. Smith,\\n974 F.2d 1050, 1059 (9th Cir. 1991), overruled on other\\ngrounds by WMX Techs., Inc. v. Miller, 104 F.3d 1133 (9th\\nCir. 1997) (en banc)). Serious medical needs can relate to\\n“physical, dental and mental health.” Hoptowit v. Ray,\\n682 F.2d 1237, 1253 (9th Cir. 1982), abrogated on other\\ngrounds by Sandin v. Conner, 515 U.S. 472 (1995).\\n\\n The State does not dispute that Edmo’s gender dysphoria\\nis a sufficiently serious medical need to trigger the State’s\\nobligations under the Eighth Amendment. Nor could it.\\nGender dysphoria is a “serious . . . medical condition” that\\ncauses “clinically significant distress”—distress that impairs\\nor severely limits an individual’s ability to function in a\\nmeaningful way. DSM-5 at 453, 458. As Edmo testified,\\nher gender dysphoria causes her to feel “depressed,”\\n“disgusting,” “tormented,” and “hopeless,” and it has caused\\npast efforts and active thoughts of self-castration. As this\\nand many other courts have recognized, Edmo’s gender\\ndysphoria is a sufficiently serious medical need to implicate\\nthe Eighth Amendment. See Rosati v. Igbinoso, 791 F.3d\\n1037, 1039–40 (9th Cir. 2015); Kosilek, 774 F.3d at 86;\\nDe’lonta, 708 F.3d at 525; Battista v. Clarke, 645 F.3d 449,\\n452 (1st Cir. 2011); Allard v. Gomez, 9 F. App’x 793, 794\\n(9th Cir. 2001); White v. Farrier, 849 F.2d 322, 325 (8th Cir.\\n1988); Meriwether v. Faulkner, 821 F.2d 408, 412 (7th Cir.\\n\\f EDMO V. CORIZON 49\\n\\n1987) (and cases cited therein); Norsworthy, 87 F. Supp. 3d\\nat 1187; Konitzer v. Frank, 711 F. Supp. 2d 874, 905 (E.D.\\nWis. 2010).\\n\\n If, as here, a prisoner establishes a sufficiently serious\\nmedical need, that prisoner must then “show the [official’s]\\nresponse to the need was deliberately indifferent.” Jett,\\n439 F.3d at 1096. An inadvertent or negligent failure to\\nprovide adequate medical care is insufficient to establish a\\nclaim under the Eighth Amendment. Estelle, 429 U.S.\\nat 105–06; see also Farmer v. Brennan, 511 U.S. 825, 835\\n(1994) (“ordinary lack of due care” is insufficient to\\nestablish an Eighth Amendment claim). In other words,\\n“[m]edical malpractice does not become a constitutional\\nviolation merely because the victim is a prisoner.” Estelle,\\n429 U.S. at 106. To “show deliberate indifference, the\\nplaintiff must show that the course of treatment the [official]\\nchose was medically unacceptable under the circumstances\\nand that the [official] chose this course in conscious\\ndisregard of an excessive risk to the plaintiff’s health.”\\nHamby v. Hammond, 821 F.3d 1085, 1092 (9th Cir. 2016)\\n(quoting Snow v. McDaniel, 681 F.3d 978, 988 (9th Cir.\\n2012), overruled in part on other grounds by Peralta v.\\nDillard, 744 F.3d 1076 (9th Cir. 2014) (en banc)).\\n\\n 1. The Medical Necessity of GCS for Edmo\\n\\n The crux of the State’s appeal is that it provided adequate\\nand medically acceptable care to Edmo.\\n\\n Accepted standards of care and practice within the\\nmedical community are highly relevant in determining what\\ncare is medically acceptable and unacceptable. See Allard v.\\nBaldwin, 779 F.3d 768, 772 (8th Cir. 2015); Henderson v.\\nGhosh, 755 F.3d 559, 566 (7th Cir. 2014) (per curiam).\\nTypically, “[a] difference of opinion between a physician\\n\\f50 EDMO V. CORIZON\\n\\nand the prisoner—or between medical professionals—\\nconcerning what medical care is appropriate does not\\namount to deliberate indifference.” Snow, 681 F.3d at 987;\\nsee also Gibson, 920 F.3d at 220. But that is true only if the\\ndueling opinions are medically acceptable under the\\ncircumstances. See Toguchi v. Chung, 391 F.3d 1051, 1058\\n(9th Cir. 2004) (a mere “difference of medical opinion . . .\\n[is] insufficient, as a matter of law, to establish deliberate\\nindifference,” but not if the “chosen course of treatment ‘was\\nmedically unacceptable under the circumstances’”\\n(alterations in original) (quoting Jackson v. McIntosh,\\n90 F.3d 330, 332 (9th Cir. 1996))).\\n\\n “In deciding whether there has been deliberate\\nindifference to an inmate’s serious medical needs, we need\\nnot defer to the judgment of prison doctors or\\nadministrators.” Hunt v. Dental Dep’t, 865 F.2d 198, 200\\n(9th Cir. 1989). Nor does it suffice for “correctional\\nadministrators wishing to avoid treatment . . . simply to find\\na single practitioner willing to attest that some well-accepted\\ntreatment is not necessary.” Kosilek, 774 F.3d at 90 n.12. In\\nthe final analysis under the Eighth Amendment, we must\\ndetermine, considering the record, the judgments of prison\\nmedical officials, and the views of prudent professionals in\\nthe field, whether the treatment decision of responsible\\nprison authorities was medically acceptable.\\n\\n Reviewing the record and the district court’s extensive\\nfactual findings, we conclude that Edmo has established that\\nthe “course of treatment” chosen to alleviate her gender\\ndysphoria “was medically unacceptable under the\\ncircumstances.” Hamby, 821 F.3d at 1092 (quoting Snow,\\n681 F.3d at 988). This conclusion derives from the district\\ncourt’s factual findings, which are not “illogical,\\nimplausible, or without support in inferences that may be\\n\\f EDMO V. CORIZON 51\\n\\ndrawn from the facts in the record.” La Quinta Worldwide\\nLLC, 762 F.3d at 879 (quotation omitted).\\n\\n In particular, and as we will explain, this is not a case of\\ndueling experts, as the State paints it. The district court\\npermissibly credited the opinions of Edmo’s experts that\\nGCS is medically necessary to treat Edmo’s gender\\ndysphoria and that the State’s failure to provide that\\ntreatment is medically unacceptable. Edmo’s experts are\\nwell-qualified to render such opinions, and they logically\\nand persuasively explained the necessity of GCS and applied\\nthe WPATH Standards of Care—the undisputed starting\\npoint in determining the appropriate treatment for gender\\ndysphoric individuals. On the other side of the coin, the\\ndistrict court permissibly discredited the contrary opinions\\nof the State’s treating physician and medical experts. Those\\nindividuals lacked expertise and incredibly applied (or did\\nnot apply, in the case of the State’s treating physician) the\\nWPATH Standards of Care. In other words, the district court\\ndid not clearly err in making its credibility determinations,\\nso it is not our role to reevaluate them. The credited\\ntestimony establishes that GCS is medically necessary.\\n\\n a. Expert Testimony\\n\\n Turning first to the expert testimony offered, the district\\ncourt credited the testimony of Edmo’s experts that GCS is\\nmedically necessary to treat Edmo’s gender dysphoria and\\nthat the State’s failure to provide that treatment is medically\\nunacceptable. See Edmo, 358 F. Supp. 3d at 1120–21, 1125.\\nDr. Ettner and Dr. Gorton opined that GCS is medically\\nnecessary because Edmo’s current treatment has been\\ninadequate, as evidenced by her self-castration attempts.\\nThey also opined that if Edmo does not receive GCS, there\\nis little chance that her gender dysphoria will improve and\\nshe is at risk of committing self-surgery again, suicide, and\\n\\f52 EDMO V. CORIZON\\n\\nfurther emotional decompensation. On the other hand,\\nproviding GCS to Edmo would, in the opinions of Dr. Ettner\\nand Dr. Gorton, align Edmo’s genitalia with her gender\\nidentity, thereby eliminating the severe distress Edmo\\nexperiences from her male genitalia.\\n\\n In sharp contrast, the district court gave “virtually no\\nweight” to the opinions of the State’s experts. Edmo, 358 F.\\nSupp. 3d at 1126. Dr. Garvey and Dr. Andrade, who\\npurported to rely on the WPATH Standards of Care, opined\\nthat GCS is not medically necessary for Edmo.\\n\\n The district court did not err in crediting the testimony of\\nEdmo’s experts and discounting the testimony of the State’s\\nexperts. Dr. Ettner and Dr. Gorton are well-qualified to\\nopine on the medical necessity of GCS. Both have\\nsubstantial experience treating individuals with gender\\ndysphoria. Dr. Ettner has evaluated, diagnosed, and treated\\nbetween 2,500 and 3,000 individuals with gender dysphoria,\\nwhile Dr. Gorton has been the primary care physician for\\napproximately 400 patients with gender dysphoria. Both\\nhave substantial experience evaluating whether GCS is\\nmedically necessary for patients. Dr. Ettner has evaluated\\nhundreds of people for GCS, referring approximately 300\\nwhile refusing others, and Dr. Gorton routinely determines\\nthe appropriateness of GCS for patients. They also have\\nexperience providing follow-up care for patients who have\\nundergone GCS. And both have published peer-reviewed\\narticles concerning the treatment of gender dysphoria.\\n\\n The State’s experts, by contrast, have substantial\\nexperience providing health care in institutional settings, but\\nlack meaningful experience directly treating people with\\ngender dysphoria. Dr. Garvey directly treated a “couple of\\npatients” with gender dysphoria early in her career, while\\nDr. Andrade has never provided direct treatment for patients\\n\\f EDMO V. CORIZON 53\\n\\nwith gender dysphoria. Moreover, prior to evaluating Edmo,\\nneither had ever evaluated someone in person to determine\\nthe medical necessity of GCS. Relatedly, Dr. Garvey and\\nDr. Andrade have never provided follow-up care for a person\\nwho has received GCS. Indeed, Dr. Andrade did not even\\nqualify under IDOC policy to assess a person for GCS. And\\nneither Dr. Garvey nor Dr. Andrade has published a peer-\\nreviewed article concerning the treatment of gender\\ndysphoria.\\n\\n Neither Dr. Ettner nor Dr. Gorton have treated prisoners\\nwith gender dysphoria, nor are they Certified Correctional\\nHealthcare Professionals. But both serve on WPATH’s\\nInstitutionalized Persons Committee, which “looks at the\\ncare and the assessment of individuals who are incarcerated\\nand develops standards for treatment” of such individuals.\\nThey are thus familiar with medical treatment in prison\\nsettings. Moreover, Dr. Ettner has assessed approximately\\n30 incarcerated persons with gender dysphoria for GCS and\\nother medical care.\\n\\n More to the point, the more relevant experience for\\ndetermining the medical necessity of GCS is having treated\\nindividuals with gender dysphoria, having evaluated\\nindividuals for GCS, and having treated them post-\\noperatively. Such experience lends itself to fundamental\\nknowledge of whether GCS is necessary and the potential\\nrisks of providing or foregoing the surgery. Edmo’s experts\\nhave the requisite experience; the State’s experts do not. For\\nthat reason alone, the district court did not clearly err in\\ncrediting the opinions of Edmo’s experts over those of the\\nState. 15 See Caro v. Woodford, 280 F.3d 1247, 1253 (9th\\n\\n 15\\n The State contends that neither Dr. Ettner nor Dr. Gorton was\\nqualified to offer expert opinions as to the appropriate medical care for\\n\\f54 EDMO V. CORIZON\\n\\nCir. 2002) (explaining that we “must afford the District\\nCourt considerable deference in its determination that the\\nwitnesses were qualified to draw [their] conclusions”).\\n\\n Independent of the experts’ qualifications, the district\\ncourt did not err in crediting the opinions of Edmo’s experts\\nover those of the State because aspects of Dr. Garvey’s and\\nDr. Andrade’s opinions ran contrary to the established\\nstandards of care in the area of transgender health care—the\\nWPATH Standards of Care—which they purported to\\napply. 16 See Edmo, 358 F. Supp. 3d at 1125.\\n\\n\\nEdmo because neither is a psychiatrist. So far as we can discern, the\\nargument is that because a psychiatrist (Dr. Eliason) evaluated Edmo for\\nGCS, only other psychiatrists are qualified to opine as to the medical\\nnecessity of GCS and to contradict his assessment. See Oral Arg.\\nat 10:00–10:30. We reject that contention. Edmo’s experts, as\\nexplained, have significant experience evaluating patients for GCS—\\nprecisely what Dr. Eliason did. On the basis of their medical experience\\ntreating persons with gender dysphoria, they are well-qualified to render\\nan opinion on the medical necessity of GCS and whether failure to\\nprovide the surgery is medically acceptable. See Fed. R. Evid. 702.\\n 16\\n The State contends that the district court erred in requiring strict\\nadherence to the flexible WPATH Standards of Care and in concluding\\nthat any deviation from those standards is medically unacceptable. But\\nthe district court correctly recognized that the WPATH Standards of\\nCare are flexible, see Edmo, 358 F. Supp. 3d at 1111, and it appropriately\\nused them as a starting point to gauge the credibility of each expert’s\\ntestimony, see id. at 1125–26. Tellingly, each expert for Edmo and the\\nState likewise used the WPATH Standards of Care as a starting point.\\nAs the district court recognized: “There are no other competing,\\nevidence-based standards that are accepted by any nationally or\\ninternationally recognized medical professional groups.” Id. at 1125.\\nAnd as the State acknowledged at the evidentiary hearing, the “WPATH\\nstandards of care in the seventh edition do provide the best guidance”\\nand “are the best standards out there.” For these reasons, the WPATH\\nStandards of Care establish a useful starting point for analyzing the\\n\\f EDMO V. CORIZON 55\\n\\n For example, both Dr. Garvey and Dr. Andrade\\nexpressed the view that Edmo does not meet the sixth\\nWPATH criterion, “12 continuous months of living in a\\ngender role that is congruent with gender identity.” WPATH\\nSOC at 60. They pointed out that Edmo has not presented\\nas female outside of prison and urged that she needs real-life\\nexperiences in the community before undergoing GCS.\\n\\n These opinions run head-on into the WPATH Standards\\nof Care. The WPATH standards, which the NCCHC\\nendorses as the accepted standards for the treatment of\\ntransgender inmates, apply\\n\\n in their entirety . . . to all transsexual,\\n transgender, and gender nonconforming\\n people, irrespective of their housing\\n situation. People should not be discriminated\\n against in their access to appropriate health\\n care based on where they live, including\\n institutional environments such as prisons\\n .... Health care for transsexual,\\n transgender, and gender nonconforming\\n people living in an institutional environment\\n should mirror that which would be available\\n to them if they were living in a non-\\n institutional setting within the same\\n community.\\n\\n\\n\\ncredibility and weight to be given to each expert’s opinion and whether\\nthat opinion was consistent with established standards of care. The State\\ndoes not contest the district court’s finding that the WPATH Standards\\nof Care are the “internationally recognized guidelines for the treatment\\nof individuals with gender dysphoria.” Id. at 1111. They are the gold\\nstandard on this issue.\\n\\f56 EDMO V. CORIZON\\n\\n All elements of assessment and treatment as\\n described in the [Standards of Care] can be\\n provided to people living in institutions.\\n Access to these medically necessary\\n treatments should not be denied on the basis\\n of institutionalization or housing\\n arrangements.\\n\\nWPATH SOC at 67. Dr. Garvey and Dr. Andrade’s view—\\nthat GCS cannot be medically indicated for transgender\\ninmates who did not present in a gender-congruent manner\\nbefore incarceration—contradicts these accepted standards.\\nDr. Garvey and Dr. Andrade would deny GCS to a class of\\npeople because of their “institutionalization,” which the\\nWPATH Standards of Care explicitly disavow. They\\nprovide no persuasive explanation for their deviation. 17 And\\nnothing in the WPATH Standards of Care or the law\\nsupports excluding an entire class of gender dysphoric\\nindividuals from eligibility for GCS.\\n\\n Both Dr. Garvey and Dr. Andrade also relied on Edmo’s\\nfailure to attend psychotherapy sessions as an indication that\\nher mental health concerns are not well controlled. But\\npsychotherapy is not a precondition for surgery under the\\nWPATH Standards of Care. WPATH SOC at 28–29.\\n\\n We acknowledge that the WPATH Standards of Care are\\nflexible, and a simple deviation from those standards does\\nnot alone establish an Eighth Amendment claim. But the\\n 17\\n In concluding that Edmo does not meet the sixth WPATH\\ncriterion, Dr. Garvey expressed concern that there is a lack of evidence\\nregarding GCS in prison settings. That rationale acts as self-fulfilling\\nprophecy. If prisons and prison officials deny GCS to prisoners because\\nof a lack of data, the data will never be generated, and the cycle will\\ncontinue.\\n\\f EDMO V. CORIZON 57\\n\\nState’s experts purported to be applying those standards and\\nyet did so in a way that directly contradicted them. These\\nunsupported and unexplained deviations offer a further\\nreason why the district court did not clearly err in\\ndiscounting the testimony of the State’s experts. See Caro,\\n280 F.3d at 1253.\\n\\n Finally, the district court did not err in discrediting the\\nState’s experts because aspects of their opinions were\\nillogical and unpersuasive. For example, Dr. Garvey and\\nDr. Andrade expressed the view that Edmo does not meet the\\nfirst WPATH criterion—“persistent, well documented\\ngender dysphoria,” WPATH SOC at 60—because of a lack\\nof evidence from pre-incarceration records of Edmo\\npresenting as female. But both experts acknowledged that\\nEdmo has been diagnosed with and treated for gender\\ndysphoria since 2012—i.e., for six years as of the evidentiary\\nhearing. Neither Dr. Garvey nor Dr. Andrade questioned\\nEdmo’s diagnosis, and both agree that she currently suffers\\ngender dysphoria. There can be no doubt that Edmo has\\n“persistent, well documented gender dysphoria,” so their\\nopinion is inexplicable.\\n\\n Dr. Garvey’s and Dr. Andrade’s opinions on this point\\nalso ignore that individuals with gender dysphoria do not\\nalways experience symptoms early in life or throughout their\\nlife, or do not identify them as such. As Dr. Ettner testified,\\n“gender dysphoria intensifies with age.” And as with\\ntreatment for any other medical condition, treatment for\\ngender dysphoria must be based on a patient’s current\\nsituation.\\n\\n The opinions of Edmo’s experts are notably devoid of\\nthese flaws. Dr. Ettner and Dr. Gorton cogently and\\npersuasively explained why GCS is medically necessary for\\nEdmo and why Edmo meets the WPATH criteria for GCS.\\n\\f58 EDMO V. CORIZON\\n\\n For example, consistent with the WPATH Standards of\\nCare, Dr. Ettner explained that Edmo has lived for\\n“12 continuous months . . . in a gender role that is congruent\\nwith gender identity” (the sixth WPATH criterion) because\\nshe has lived “as a woman to the best of her ability in a male\\nprison.” In support of her opinion, Dr. Ettner cited Edmo’s\\n“appearance . . . , her disciplinary records, which indicated\\nthat she had attempted to wear her hair in a feminine\\nhairstyle and to wear makeup even though that was against\\nthe rules and she was – received some sort of disciplinary\\naction for that, and her – the way that she was receiving\\nfemale undergarments and had developed the stigma of\\nfemininity, the secondary sex characteristics, breast\\ndevelopment, et cetera.” Dr. Gorton similarly explained that\\nEdmo satisfies the sixth WPATH criterion because she has\\nlived for years in her “target gender role . . . despite an\\nenvironment that’s very hostile to that and some negative\\nconsequences that she has experienced because of that.”\\n\\n Moreover, both Dr. Ettner and Dr. Gorton offered\\nreasoned explanations tying Edmo’s self-castration attempts\\nto her severe gender dysphoria. Dr. Ettner explained that\\ndoctors regard “surgical self-treatment . . . as an intentional\\nattempt to remove the target organ that produces\\ntestosterone, which, in fact, is the cure for gender\\ndysphoria.” As Dr. Gorton elaborated, Edmo’s self-\\ncastration attempts demonstrate deficient treatment for\\n“severe genital-focused gender dysphoria.” He rejected the\\nnotion that Edmo’s depression and anxiety drove her self-\\ncastration attempts: “there [are] a lot of people with\\ndepression and anxiety who don’t remove their testicles.”\\n\\n In light of the experts’ backgrounds and experience, and\\nthe reasonableness, consistency, and persuasiveness of their\\nopinions, the district court did not err in crediting the\\n\\f EDMO V. CORIZON 59\\n\\nopinions of Edmo’s experts and giving little weight to those\\nof the State’s experts. The district court carefully examined\\nthe voluminous record, extensive testimony, and conflicting\\nexpert opinions in this case and set forth clear reasons,\\nsupported by the record, for relying on the testimony of\\nEdmo’s experts. See La Quinta Worldwide, 762 F.3d at 879\\n(a factual finding is clear error if it is “illogical, implausible,\\nor without support in inferences that may be drawn from the\\nfacts in the record”); Caro, 280 F.3d at 1253; Beech Aircraft\\nCorp. v. United States, 51 F.3d 834, 838 (9th Cir. 1995) (per\\ncuriam). The credited expert testimony established that GCS\\nis medically necessary to alleviate Edmo’s gender dysphoria.\\n\\n b. Dr. Eliason’s Assessment\\n\\n Turning from the expert testimony offered, the State\\ncontends that Edmo’s experts, at most, created a dispute of\\nprofessional judgment with Edmo’s treating psychiatrist, Dr.\\nEliason, who it urges reasonably concluded that GCS is\\ninappropriate for Edmo. If that is the case, the argument\\ngoes, then Edmo’s Eighth Amendment claim fails because\\nthe dispute is merely a “difference of opinion . . . between\\nmedical professionals” about “what medical care is\\nappropriate.” Snow, 681 F.3d at 987. The problem for the\\nState is that Dr. Eliason’s decision “was medically\\nunacceptable under the circumstances.” Toguchi, 391 F.3d\\nat 1058 (quoting Jackson, 90 F.3d at 332).\\n\\n In particular, as the district court found, Dr. Eliason did\\nnot follow accepted standards of care in the area of\\ntransgender health care. See Edmo, 358 F. Supp. 3d at 1126.\\nDr. Eliason explained in his notes that, in his view, GCS is\\nmedically necessary in three situations: “congenital\\nmalformation or ambiguous genitalia,” “severe and\\ndevastating dysphoria that is primarily due to genitals,” or\\n“some type of medical problem in which endogenous sexual\\n\\f60 EDMO V. CORIZON\\n\\nhormones were causing severe physiological damage.” The\\nconclusion of his notes—“[t]his inmate does not meet any of\\nthose [three] criteria”—suggests that he views those as the\\nonly three scenarios in which GCS would be medically\\nnecessary, an impression he did not dispel during his\\ntestimony. Those “criteria” (Dr. Eliason’s term), however,\\nbear little resemblance to the widely accepted, evidence-\\nbased criteria set out in the WPATH’s Standards of Care. As\\nDr. Eliason acknowledged, the NCCHC endorses the\\nWPATH Standards of Care as the accepted standards for the\\ntreatment of transgender prisoners. And as the district court\\nfound and the State does not contest, “[t]here are no other\\ncompeting, evidence-based standards that are accepted by\\nany nationally or internationally recognized medical\\nprofessional groups.” Id. at 1125. Dr. Eliason did not follow\\nthese standards in rendering his decision.\\n\\n The State challenges the district court’s finding that\\nDr. Eliason “did not apply the WPATH Criteria,” id. at 1126,\\non two grounds. First, citing Dr. Eliason’s testimony at the\\nevidentiary hearing, it urges that Dr. Eliason concluded that\\nGCS was not medically necessary for Edmo because Edmo’s\\nmental health issues were not well controlled (the fourth\\nWPATH criterion) and she had not consistently presented as\\nfemale outside of prison (the sixth).\\n\\n The district court’s rejection of this post hoc explanation\\nwas not clear error. Neither of the explanations offered by\\nDr. Eliason during the evidentiary hearing appears in\\nDr. Eliason’s notes. Nor did he give these reasons during his\\ndeposition. Their absence is conspicuous, given that\\nDr. Eliason took the time to indicate instances where, in his\\nopinion, GCS is appropriate and to explain that Edmo did\\nnot satisfy his “criteria.”\\n\\f EDMO V. CORIZON 61\\n\\n Second, the State highlights that Dr. Eliason’s notes\\nrecommend further “supportive counseling” for Edmo and\\nindicate that Edmo was up for parole. The State construes\\nthese notes as shorthand for the fourth and sixth WPATH\\ncriteria, respectively. The State’s proposed reading of\\nDr. Eliason’s notes is unreasonable. His notes are clear that\\nGCS is not needed because Edmo did not meet his three\\n“criteria,” and the district court was well within its\\nfactfinding discretion in rejecting the State’s strained\\nreading. We therefore conclude that the district court\\nreasonably found that Dr. Eliason “did not rely upon any\\nfinding that Ms. Edmo did not meet the WPATH criteria in\\nconcluding in his April 2016 assessment that she did not\\nmeet the criteria for gender confirmation surgery.” Id.\\nat 1120.\\n\\n Notably, neither Dr. Eliason nor the State has offered any\\nexplanation or support for Dr. Eliason’s “criteria.”\\nDr. Eliason testified that he could not recall where he came\\nup with them.\\n\\n Nor has Dr. Eliason or the State contended that\\nDr. Eliason’s criteria were a reasonable deviation or\\nmodification of the WPATH Standards of Care. In any\\nevent, we could not accept that argument. Dr. Eliason’s\\ncriteria—apparently invented out of whole cloth—are so far\\nafield from the WPATH standards that we cannot\\ncharacterize his decision as a flexible application of or\\ndeviation from those standards. Indeed, as Dr. Gorton\\nexplained, two of Dr. Eliason’s criteria are inapplicable to\\nthe care of transgender individuals. Dr. Eliason’s criterion\\nof “congenital malformation or ambiguous genitalia” “isn’t\\n. . . germane to transgender people.” His statement that GCS\\ncould be needed when “endogenous sexual hormones were\\ncausing severe physiological damage,” is, in Dr. Gorton’s\\n\\f62 EDMO V. CORIZON\\n\\nwords, “bizarre. I can’t think of a clinical circumstance\\nwhere . . . your hormones that your body produces are\\nattacking you . . . . I just don’t understand what [Dr. Eliason]\\nis talking about there.”\\n\\n Dr. Eliason, in short, did not follow the accepted\\nstandards of care in the area of transgender health care, nor\\ndid he reasonably deviate from or flexibly apply them.\\nDr. Eliason did not apply the established standards, even as\\na starting point, in his evaluation.\\n\\n Putting to the side Dr. Eliason’s failure to follow or\\nreasonably deviate from the accepted standards of care, his\\ndecision was internally contradictory in an important way.\\nHis notes reflect that GCS would be medically necessary if\\na person is suffering “severe and devastating gender\\ndysphoria that is primarily due to genitals.” At his\\ndeposition, Dr. Eliason conceded that self-castration could\\nshow gender dysphoria sufficiently severe to satisfy that\\ncriterion. And at the evidentiary hearing, he acknowledged\\nthat Edmo “does primarily meet that criteri[on].” Thus, even\\nunder Dr. Eliason’s own criteria, Edmo should have been\\nprovided GCS. Neither Dr. Eliason nor the State has\\nreconciled this important contradiction between\\nDr. Eliason’s criteria and his determination.\\n\\n In sum, Dr. Eliason’s evaluation was not an exercise of\\nmedically acceptable professional judgment. Dr. Eliason’s\\ndecision was based on inexplicable criteria far afield from\\nthe recognized standards of care and, even applying\\nDr. Eliason’s criteria, Edmo qualifies for GCS. Given the\\ncredited expert testimony that GCS is necessary to treat\\nEdmo’s gender dysphoria, Dr. Eliason’s contrary\\n\\f EDMO V. CORIZON 63\\n\\ndetermination was “medically unacceptable under the\\ncircumstances.” 18 Snow, 681 F.3d at 988.\\n\\n 2. Deliberate Indifference\\n\\n The State next contends that even if the treatment\\nprovided Edmo was medically unacceptable, no defendant\\nacted “in conscious disregard of an excessive risk to\\n[Edmo’s] health.” Hamby, 821 F.3d at 1092 (quoting Snow,\\n681 F.3d at 988). We disagree.\\n\\n The record demonstrates that Dr. Eliason acted with\\ndeliberate indifference to Edmo’s serious medical needs.\\nDr. Eliason knew, as of the time of his evaluation, that Edmo\\nhad attempted to castrate herself. He also knew that Edmo\\nsuffers from gender dysphoria; he knew she experiences\\n“clinically significant” distress that impairs her ability to\\nfunction. He acknowledged that Edmo’s self-castration\\nattempt was evidence that Edmo’s gender dysphoria, in his\\nwords, “had risen to another level.” Dr. Eliason nonetheless\\ncontinued with Edmo’s ineffective treatment plan.\\n\\n Edmo then tried to castrate herself a second time, in\\nDecember 2016. Dr. Eliason knew of that nearly\\n\\n 18\\n Dr. Eliason was not alone in his decision. Dr. Stoddart,\\nDr. Young, and Jeremy Clark agreed with his assessment, as did the\\nMTC. The State contends that such general agreement demonstrates that\\nDr. Eliason’s decision was reasonable. But general agreement in a\\nmedically unacceptable form of treatment does not somehow make it\\nreasonable. This is especially so in light of the limited review those\\nindividuals performed: Dr. Stoddard, Dr. Young, and Jeremy Clark\\nagreed with Dr. Eliason’s recommended treatment as he presented it to\\nthem and without personally evaluating Edmo, and the MTC “does not\\nmake any individual treatment decisions regarding [gender dysphoric]\\ninmates. Those determinations are made by the individual clinicians or\\nthe medical staff employed by Corizon,” like Dr. Eliason.\\n\\f64 EDMO V. CORIZON\\n\\ncatastrophic event, but he did not reevaluate or recommend\\na change to Edmo’s treatment plan, despite indicating in his\\nApril 2016 evaluation that he would continue to monitor and\\nassess Edmo’s condition. Dr. Eliason continued to see Edmo\\nafter that time, and he considered Edmo’s treatment as a\\nmember of the MTC. At no point did Dr. Eliason change his\\nmind or the treatment plan regarding surgery. Under these\\ncircumstances, we conclude that Dr. Eliason knew of and\\ndisregarded the substantial risk of severe harm to Edmo.\\nFarmer, 511 U.S. at 837.\\n\\n The State urges that neither Dr. Eliason nor any other\\ndefendant acted with deliberate indifference because none\\nacted with “malice, intent to inflict pain, or knowledge that\\n[the] recommended course of treatment was medically\\ninappropriate.” The State misstates the standard. A prisoner\\n“must show that prison officials ‘kn[e]w [ ] of and\\ndisregard[ed]’ the substantial risk of harm,’ but the officials\\nneed not have intended any harm to befall the inmate; ‘it is\\nenough that the official acted or failed to act despite his\\nknowledge of a substantial risk of serious harm.’” Lemire v.\\nCal. Dep’t of Corr. & Rehab., 726 F.3d 1062, 1074 (9th Cir.\\n2013) (alterations in original) (quoting Farmer, 511 U.S.\\nat 837, 842). Neither the Supreme Court nor this court has\\never required a plaintiff to show a “sinister [prison official]\\nwith improper motives,” as the State would require. It is\\nenough that Dr. Eliason knew of and disregarded an\\nexcessive risk to Edmo’s health by rejecting her request for\\nGCS and then never re-evaluating his decision despite\\nongoing harm to Edmo.\\n\\n The State also contends that because the defendants\\nprovided some care to Edmo, no defendant could have been\\ndeliberately indifferent. The provision of some medical\\ntreatment, even extensive treatment over a period of years,\\n\\f EDMO V. CORIZON 65\\n\\ndoes not immunize officials from the Eighth Amendment’s\\nrequirements. See Lopez v. Smith, 203 F.3d 1122, 1132 (9th\\nCir. 2000) (en banc) (explaining that “[a] prisoner need not\\nprove that he was completely denied medical care” to make\\nout an Eighth Amendment claim); see also De’lonta,\\n708 F.3d at 526 (“[J]ust because [officials] have provided\\nDe’lonta with some treatment consistent with the GID\\nStandards of Care, it does not follow that they have\\nnecessarily provided her with constitutionally adequate\\ntreatment.”). As the Fourth Circuit has aptly analogized,\\n\\n imagine that prison officials prescribe a\\n painkiller to an inmate who has suffered a\\n serious injury from a fall, but that the\\n inmate’s symptoms, despite the medication,\\n persist to the point that he now, by all\\n objective measure, requires evaluation for\\n surgery. Would prison officials then be free\\n to deny him consideration for surgery,\\n immunized from constitutional suit by the\\n fact they were giving him a painkiller? We\\n think not.\\n\\nDe’lonta, 708 F.3d at 526. Here, although the treatment\\nprovided Edmo was important, it stopped short of what was\\nmedically necessary.\\n\\n 3. Out-of-Circuit Precedent\\n\\n Our decision cleaves to settled Eighth Amendment\\njurisprudence, which requires a fact-specific analysis of the\\nrecord (as construed by the district court) in each case. See\\nPatel v. Kent Sch. Dist., 648 F.3d 965, 975 (9th Cir. 2011)\\n(“Deliberate-indifference cases are by their nature highly\\nfact-specific . . . .”); see also Rachel v. Troutt, 820 F.3d 390,\\n394 (10th Cir. 2016) (“Each step of this [deliberate\\n\\f66 EDMO V. CORIZON\\n\\nindifference] inquiry is fact-intensive.” (quoting Hartsfield\\nv. Colburn, 491 F.3d 394, 397 (8th Cir. 2007))); Roe v.\\nElyea, 631 F.3d 843, 859 (7th Cir. 2011) (“[I]nmate medical\\ncare decisions must be fact-based with respect to the\\nparticular inmate, the severity and stage of his condition, the\\nlikelihood and imminence of further harm and the efficacy\\nof available treatments.”); Youmans v. Gagnon, 626 F.3d\\n557, 564 (11th Cir. 2010) (“Judicial decisions addressing\\ndeliberate indifference to a serious medical need, like\\ndecisions in the Fourth Amendment search-and-seizure\\nrealm, are very fact specific.”); Chance v. Armstrong,\\n143 F.3d 698, 703 (2d Cir. 1998) (“Whether a course of\\ntreatment was the product of sound medical judgment,\\nnegligence, or deliberate indifference depends on the facts\\nof the case.”).\\n\\n Several years ago, the First Circuit, sitting en banc,\\nemployed that fact-based approach to evaluate a gender\\ndysphoric prisoner’s Eighth Amendment claim seeking\\nGCS. The First Circuit confronted the following record:\\ncredited expert testimony disagreed as to whether GCS was\\nmedically necessary; the prisoner’s active treatment plan,\\nwhich did not include GCS, had “led to a significant\\nstabilization in her mental state”; and a report and testimony\\nfrom correctional officials detailed significant security\\nconcerns that would arise if the prisoner underwent GCS.\\nKosilek, 774 F.3d at 86–96. “After carefully considering the\\ncommunity standard of medical care, the adequacy of the\\nprovided treatment, and the valid security concerns\\narticulated by the DOC,” a 3–2 majority of the en banc court\\nconcluded that the plaintiff had not demonstrated GCS was\\nmedically necessary treatment for her gender dysphoria. Id.\\nat 68.\\n\\f EDMO V. CORIZON 67\\n\\n Our approach mirrors the First Circuit’s, but the\\nimportant factual differences between cases yield different\\noutcomes. Notably, the security concerns in Kosilek, which\\nthe First Circuit afforded “wide-ranging deference,” are\\ncompletely absent here. Id. at 92. The State does not so\\nmuch as allude to them. The medical evidence also differs.\\nIn Kosilek, qualified and credited experts disagreed about\\nwhether GCS was necessary. Id. at 90. As explained above,\\nthe district court’s careful factual findings admit of no such\\ndisagreement here. Rather, they unequivocally establish that\\nGCS is the safe, effective, and medically necessary treatment\\nfor Edmo’s severe gender dysphoria.\\n\\n We recognize, however, that our decision is in tension\\nwith Gibson v. Collier. In that case, the Fifth Circuit held,\\nin a split decision, that “[a] state does not inflict cruel and\\nunusual punishment by declining to provide [GCS] to a\\ntransgender inmate.” 920 F.3d at 215. It did so on a “sparse\\nrecord”—which included only the WPATH Standards of\\nCare and was notably devoid of “witness testimony or\\nevidence from professionals in the field”—compiled by a\\npro se plaintiff. Id. at 220. Despite the sparse record, a 2–1\\nmajority of the Gibson panel concluded that “there is no\\nconsensus in the medical community about the necessity and\\nefficacy of [GCS] as a treatment for gender dysphoria. . . .\\nThis on-going medical debate dooms Gibson’s claim.” Id.\\nat 221.\\n\\n We respectfully disagree with the categorical nature of\\nour sister circuit’s holding. Most fundamentally, Gibson\\nrelies on an incorrect, or at best outdated, premise: that\\n“[t]here is no medical consensus that [GCS] is a necessary\\nor even effective treatment for gender dysphoria.” Id. at 223.\\n\\n As the record here demonstrates and the State does not\\nseriously dispute, the medical consensus is that GCS is\\n\\f68 EDMO V. CORIZON\\n\\neffective and medically necessary in appropriate\\ncircumstances. The WPATH Standards of Care—which are\\nendorsed by the American Medical Association, the\\nAmerican Medical Student Association, the American\\nPsychiatric Association, the American Psychological\\nAssociation, the American Family Practice Association, the\\nEndocrine Society, the National Association of Social\\nWorkers, the American Academy of Plastic Surgeons, the\\nAmerican College of Surgeons, Health Professionals\\nAdvancing LGBTQ Equality, the HIV Medicine\\nAssociation, the Lesbian, Bisexual, Gay and Transgender\\nPhysician Assistant Caucus, and Mental Health America—\\nrecognize this fact. WPATH SOC at 54–55. Each expert in\\nthis case agrees. As do others in the medical community.\\nSee, e.g., U.S. Dep’t of Health & Human Servs., No. A-13-\\n87, Decision No. 2576; Bao Ngoc N. Tran, et al., Gender\\nAffirmation Surgery: A Synopsis Using American College of\\nSurgeons National Surgery Quality Improvement Program\\nand National Inpatient Sample Databases, 80 Annals Plastic\\nSurgery S229, S234 (2018); Frey, A Historical Review of\\nGender-Affirming Medicine, 14 J. Sexual Med. at 991; see\\nalso What We Know Project, Ctr. for the Study of\\nInequality, Cornell Univ., What Does the Scholarly\\nResearch Say About the Effect of Gender Transition on\\nTransgender Well-Being?, https://whatweknow.inequality.c\\nornell.edu/topics/lgbt-equality/what-does-the-scholarly-rese\\narch-say-about-the-well-being-of-transgender-people/ (last\\nvisited July 10, 2019) (reviewing the available literature and\\nfinding “a robust international consensus in the peer-\\nreviewed literature that gender transition, including medical\\ntreatments such as hormone therapy and surgeries, improves\\nthe overall well-being of transgender individuals”). The\\nFifth Circuit is the outlier.\\n\\f EDMO V. CORIZON 69\\n\\n Gibson’s broad holding stemmed from a dismaying\\ndisregard for procedure. As noted, the “sparse” summary\\njudgment record that the pro se plaintiff developed included\\n“only the WPATH Standards of Care.” Gibson, 920 F.3d at\\n221. Perhaps that factual deficiency doomed Gibson’s\\nEighth Amendment claim. See id. at 223–24. But to reach\\nits broader holding that denying GCS cannot, as a matter of\\nlaw, violate the Eighth Amendment—in other words, to\\nreject every conceivable Eighth Amendment claim based on\\nthe denial of GCS—the Fifth Circuit coopted the record from\\nKosilek, a First Circuit decision that predates Gibson by four\\nyears. Id. at 221–23. We doubt the analytical value of such\\nan anomalous procedural approach.\\n\\n Worse yet, the medical opinions from Kosilek do not\\nsupport the Fifth Circuit’s categorical holding. Dr. Chester\\nSchmidt’s and Dr. Stephen Levine’s testimony in Kosilek,\\nwhich the Fifth Circuit relied on, do not support the\\nproposition that GCS is never medically necessary. Dr.\\nSchmidt and Dr. Levine testified that GCS was not necessary\\nin the factual circumstances of that case, that is, based on the\\nunique medical needs of the prisoner at issue. See Kosilek,\\n774 F.3d at 76–79.\\n\\n The only suggestion in Kosilek that GCS is never\\nmedically necessary is in the First Circuit’s recitation of the\\ntestimony of Dr. Cynthia Osborne. See Gibson, 920 F.3d\\nat 221. The First Circuit recounted that Dr. Osborne testified\\nthat she “did not view [GCS] as medically necessary in light\\nof the ‘whole continuum from noninvasive to invasive’\\ntreatment options available to individuals with” gender\\ndysphoria. Kosilek, 774 F.3d at 77. To the extent this vague\\nportrait of Dr. Osborne’s testimony conveys her belief that\\nGCS is never medically necessary, she has apparently\\nchanged her view in the more than ten years since she\\n\\f70 EDMO V. CORIZON\\n\\ntestified in Kosilek. Like both sides and all four medical\\nexperts who testified here, Dr. Osborne now agrees that GCS\\n“can be medically necessary for some, though not all,\\npersons with [gender dysphoria], including some prison\\ninmates.” Osborne & Lawrence, Male Prison Inmates With\\nGender Dysphoria, 45 Archives of Sexual Behav. at 1651.\\nIn her and her co-author’s words, “[GCS] is a safe, effective,\\nand widely accepted treatment for [gender dysphoria];\\ndisputing the medical necessity of [GCS] based on assertions\\nto the contrary is unsupportable.” Id. The predicate medical\\nopinions that Gibson is premised upon, then, do not support\\nthe Fifth Circuit’s view that GCS is never medically\\nnecessary. The consensus is that GCS is effective and\\nmedically necessary in appropriate circumstances. 19\\n\\n Gibson is unpersuasive for several additional reasons. It\\ndirectly conflicts with decisions of this circuit, the Fourth\\n\\n 19\\n We do not suggest that every member of the medical and mental\\nhealth communities agrees that GCS may be medically necessary. There\\nare outliers. But when the medical consensus is that a treatment is\\neffective and medically necessary under the circumstances, prison\\nofficials render unacceptable care by following the views of outliers\\nwithout offering a credible medical basis for deviating from the accepted\\nview. See Kosilek, 774 F.3d at 90 n.12 (explaining that it is not enough\\nfor “correctional administrators wishing to avoid treatment . . . simply to\\nfind a single practitioner willing to attest that some well-accepted\\ntreatment is not necessary”); Hamilton v. Endell, 981 F.2d 1062, 1067\\n(9th Cir. 1992) (“By choosing to rely upon a medical opinion which a\\nreasonable person would likely determine to be inferior, the prison\\nofficials took actions which may have amounted to the denial of medical\\ntreatment, and the unnecessary and wanton infliction of pain.” (quotation\\nomitted)), overruled in part on other grounds as recognized in Snow,\\n681 F.3d at 986; cf. also Bragdon v. Abbott, 524 U.S. 624, 650 (1998)\\n(“A health care professional who disagrees with the prevailing medical\\nconsensus may refute it by citing a credible scientific basis for deviating\\nfrom the accepted norm.”).\\n\\f EDMO V. CORIZON 71\\n\\nCircuit, and the Seventh Circuit, all of which have held that\\ndenying surgical treatment for gender dysphoria can pose a\\ncognizable Eighth Amendment claim. Rosati, 791 F.3d at\\n1040 (alleged blanket ban on GCS and denial of GCS to\\nplaintiff with severe symptoms, including repeated self-\\ncastration attempts, states an Eighth Amendment claim);\\nFields v. Smith, 653 F.3d 550, 552–53, 558–59 (7th Cir.\\n2011) (law banning hormone treatment and GCS, even if\\nmedically necessary, violates the Eighth Amendment);\\nDe’lonta, 708 F.3d at 525 (alleged denial of an evaluation\\nfor GCS states an Eighth Amendment claim). 20 Relatedly,\\nGibson eschews Eighth Amendment precedent requiring a\\ncase-by-case determination of the medical necessity of a\\nparticular treatment. See, e.g., Colwell v. Bannister,\\n763 F.3d 1060, 1068 (9th Cir. 2014) (holding that the\\n“blanket, categorical denial of medically indicated surgery\\nsolely on the basis of an administrative policy . . . is the\\nparadigm of deliberate indifference” (quotation omitted));\\nRoe, 631 F.3d at 859.\\n\\n In this latter respect, Gibson also contradicts and\\nmisconstrues the precedent it purports to follow: Kosilek.\\nAccording to the Gibson majority, “the majority in Kosilek\\neffectively allowed a blanket ban on sex reassignment\\nsurgery.” 920 F.3d at 216. Not so. The First Circuit did\\nprecisely what we do here: assess whether the record before\\nit demonstrated deliberate indifference to the plaintiff’s\\n\\n 20\\n The Fifth Circuit unpersuasively attempted to reconcile its\\ndecision with Rosati and De’lonta, pointing out that those decisions\\n“allowed Eighth Amendment claims for [GCS] to survive motions to\\ndismiss, without addressing the merits.” Gibson, 920 F.3d at 223 n.8.\\nBut if Gibson is correct that failing to provide GCS cannot amount to\\ndeliberate indifference, then a plaintiff cannot state an Eighth\\nAmendment claim based on the denial of GCS. Rosati and De’lonta\\nwould necessarily have been decided differently under Gibson’s holding.\\n\\f72 EDMO V. CORIZON\\n\\ngender dysphoria. On the record before it, the First Circuit\\ndetermined that either of two courses of treatment (one\\nincluded GCS and one did not) were medically acceptable.\\nKosilek, 774 F.3d at 90. In light of those medically\\nacceptable alternatives, the First Circuit explained that it was\\nnot its place to “second guess medical judgments or to\\nrequire that the DOC adopt the more compassionate of two\\nadequate options.” Id. (quotation omitted). It expressly\\ncautioned that the opinion should not be read to “create a de\\nfacto ban against [GCS] as a medical treatment for any\\nincarcerated individual,” as “any such policy would conflict\\nwith the requirement that medical care be individualized\\nbased on a particular prisoner’s serious medical needs.” Id.\\nat 91 (citing Roe, 631 F.3d at 862–63). The Fifth Circuit\\ndisregarded these words of warning. 21\\n\\n ***\\n\\n In summary, Edmo has established that she suffers from\\na “serious medical need,” Jett, 439 F.3d at 1096, and that the\\ntreatment provided was “medically unacceptable under the\\ncircumstances” and chosen “in conscious disregard of an\\nexcessive risk” to her health, Hamby, 821 F.3d at 1092. She\\nestablished her Eighth Amendment claim of deliberate\\nindifference as to Defendant-Appellant Dr. Eliason.\\n\\n\\n\\n 21\\n Gibson’s final, originalist rationale—that it cannot be cruel and\\nunusual to deny a surgery that has only once been provided to an inmate,\\n920 F.3d at 226–28—warrants little discussion. Gibson’s originalist\\nunderstanding of the Eighth Amendment does not control; Estelle does,\\nand under Estelle a plaintiff establishes an Eighth Amendment claim by\\ndemonstrating that prison officials were deliberately indifferent to a\\nserious medical need. 429 U.S. at 106. This standard protects the\\nevolving standards of decency enshrined in the Eighth Amendment.\\n\\f EDMO V. CORIZON 73\\n\\n B. Irreparable Harm\\n\\n The State next contends that the district court erred in\\nfinding that Edmo would be irreparably harmed absent an\\ninjunction.\\n\\n In reaching its conclusion, the district court found that\\nEdmo experiences ongoing “clinically significant distress,”\\nmeaning “the distress impairs or severely limits [her] ability\\nto function in a meaningful way.” Edmo, 358 F. Supp. 3d\\nat 1110–11. This finding is supported by Edmo’s testimony\\nthat her gender dysphoria causes her to feel “depressed,”\\n“disgusting,” “tormented,” and “hopeless”; that she actively\\nexperiences thoughts of self-castration; and that she “self-\\nmedicat[es]” by cutting her arms with a razor to avoid acting\\non those thoughts and impulses. The district court also\\nfound that in the absence of surgery, Edmo “will suffer\\nserious psychological harm and will be at high risk of self-\\ncastration and suicide.” Id. at 1128. This finding is\\nsupported by the credited expert testimony of Dr. Ettner and\\nDr. Gorton, who detailed the escalating risks of self-surgery,\\nsuicide, and emotional decompensation should Edmo be\\ndenied surgery.\\n\\n It is no leap to conclude that Edmo’s severe, ongoing\\npsychological distress and the high risk of self-castration and\\nsuicide she faces absent surgery constitute irreparable harm.\\nSee Stanley v. Univ. of S. Cal., 13 F.3d 1313, 1324 n.5 (9th\\nCir. 1994); Thomas v. County of Los Angeles, 978 F.2d 504,\\n511 (9th Cir. 1992); Chalk v. U.S. Dist. Ct. Cent. Dist. of\\nCal., 840 F.2d 701, 709 (9th Cir. 1988). Moreover, the\\ndeprivation of Edmo’s constitutional right to adequate\\nmedical care is sufficient to establish irreparable harm. See\\nNelson v. NASA, 530 F.3d 865, 882 (9th Cir. 2008) (“Unlike\\nmonetary injuries, constitutional violations cannot be\\nadequately remedied through damages and therefore\\n\\f74 EDMO V. CORIZON\\n\\ngenerally constitute irreparable harm.”), rev’d and\\nremanded on other grounds, 562 U.S. 134 (2011).\\n\\n The State offers three contentions as to why the district\\ncourt erred in finding that Edmo would be irreparably\\ninjured in the absence of an injunction. None is persuasive.\\n\\n First, the State argues that the “long delay” of “nearly a\\nyear” between Edmo filing her Amended Complaint and her\\npreliminary injunction motion “implies a lack of urgency\\nand irreparable harm.” We disagree. The procedural history\\ndemonstrates that Edmo did not sit on her rights. Proceeding\\npro se, Edmo moved for preliminary injunctive relief when\\nshe filed her original complaint. The court then appointed\\ncounsel for Edmo, and shortly after appearing, appointed\\ncounsel withdrew Edmo’s motion and filed an amended\\ncomplaint. To assess the urgency of surgery, Edmo’s\\ncounsel promptly sought access to Edmo’s medical records,\\nwhich the State did not produce until more than six months\\nlater. Edmo moved for injunctive relief shortly thereafter.\\nDuring that time, Edmo and her counsel diligently\\ninvestigated and compiled the necessary record to move for\\ninjunctive relief. That it took them months to do their\\ndiligence does not suggest that Edmo will not be harmed\\nabsent an injunction.\\n\\n Second, the State contends that Edmo has not established\\nirreparable injury because both she and her expert,\\nDr. Gorton, agree that GCS is not an emergency surgery and\\nthat the State should have six months to provide such\\nsurgery. The State’s argument would preclude courts from\\nordering non-emergent medical care, even if the Eighth\\nAmendment demands it. That is untenable. The State also\\nignores the rationale for the six-month time period. As Dr.\\nGorton explained, all patients who receive GCS “are seen,\\nthey are evaluated, there is a process you have to go\\n\\f EDMO V. CORIZON 75\\n\\nthrough.” In his experience, that process typically concludes\\nwithin six months. That Edmo requested relief on a\\nreasonable timeline, based on the medical evidence, does not\\nundermine the strong evidence of irreparable injury.\\n\\n Third, the State contends that Edmo has not established\\nirreparable harm because she “has not attempted suicide or\\nself-castration for years.” That argument overlooks the\\nprofound, persistent distress Edmo’s gender dysphoria\\ncauses, as well as the credited expert testimony that absent\\nGCS, Edmo is at risk of further attempts at self-castration,\\nand possibly suicide. The district court did not err in finding\\nthat Edmo would be irreparably harmed in the absence of an\\ninjunction.\\n\\n IV. Challenges to the Scope of the Injunction\\n\\n We turn to the State’s contentions that the district court’s\\ninjunction was overbroad.\\n\\n A. Individual Defendants\\n\\n The State contends that the injunction should not apply\\nto Atencio, Zmuda, Yordy, Siegert, Dr. Young, Dr. Craig,\\nDr. Eliason, or Dr. Whinnery because the district court did\\nnot find that they, individually, were deliberately indifferent\\nto Edmo’s medical needs.\\n\\n As explained in Section III.A, Edmo has established that\\nDr. Eliason was deliberately indifferent to her serious\\nmedical needs. The injunction was properly entered against\\nhim because he personally participated in the deprivation of\\nEdmo’s constitutional rights. See Colwell, 763 F.3d at 1070.\\n\\n Edmo sued Attencio, Zmuda, and Yordy in their official\\ncapacities. An official-capacity suit for injunctive relief is\\n\\f76 EDMO V. CORIZON\\n\\nproperly brought against any persons who “would be\\nresponsible for implementing any injunctive relief.” Pouncil\\nv. Tilton, 704 F.3d 568, 576 (9th Cir. 2012). The State does\\nnot contest that Attencio, as Director of IDOC, and Zmuda,\\nas Deputy Director of IDOC, would be responsible for\\nimplementing any injunctive relief ordered. Edmo properly\\nnamed them as defendants to her Eighth Amendment claim\\nfor injunctive relief, regardless of their personal\\ninvolvement. See Colwell, 763 F.3d at 1070–71 (director of\\na state correctional system is a proper defendant in an\\nofficial-capacity suit seeking injunctive relief for Eighth\\nAmendment violations). Yordy is no longer the Warden of\\nISCI, but, by operation of the Federal Rules, his successor,\\nAl Ramirez, is “automatically substituted as party” in his\\nofficial capacity. Fed. R. Civ. P. 25(d). Ramirez is properly\\na defendant to Edmo’s Eighth Amendment claim for\\ninjunctive relief, regardless of his personal involvement. See\\nColwell, 763 F.3d at 1070–71 (warden is a proper defendant\\nin an official-capacity suit seeking injunctive relief for\\nEighth Amendment violations). Because Edmo may\\nproperly pursue her Eighth Amendment claim for injunctive\\nrelief against Attencio, Zmuda, and Ramirez in their official\\ncapacities, they are properly included within the scope of the\\ndistrict court’s injunction. On remand, the district court shall\\namend the injunction to substitute Al Ramirez (or the then-\\ncurrent Warden of ISCI) as a party for Yordy.\\n\\n Edmo also named Yordy as a defendant in his individual\\ncapacity. She likewise named Siegert, Dr. Young, Dr. Craig,\\nand Dr. Whinnery as defendants in their individual\\ncapacities (though she does not argue on appeal that the\\ninjunction properly included them). We hold that the\\nevidence in the current record is insufficient to conclude that\\nthey were deliberately indifferent to Edmo’s serious medical\\nneeds. In particular, the record does not show what they\\n\\f EDMO V. CORIZON 77\\n\\nknew about Edmo’s condition and what role they played in\\nher treatment or lack thereof. Edmo has not established their\\nliability, and the district court improperly included them\\nwithin the scope of the injunction. We vacate the district\\ncourt’s injunction to the extent it applies to Yordy, Siegert,\\nDr. Young, Dr. Craig, and Dr. Whinnery in their individual\\ncapacities. See California v. Azar, 911 F.3d 558, 585 (9th\\nCir. 2018) (vacating in part an overbroad injunction and\\nremanding to the district court). On remand, the district\\ncourt shall modify the injunction to exclude those defendants\\nfrom its scope.\\n\\n B. Corizon\\n\\n The State also contends that the injunction should not\\napply to Corizon. It urges that Corizon does not have a\\npolicy barring GCS and argues that such a policy is a\\nprerequisite to liability under Monell v. Department of Social\\nServices, 436 U.S. 658 (1978). We have not yet determined\\nwhether Monell applies “to private entities acting on behalf\\nof state governments,” such as Corizon. Oyenik v. Corizon\\nHealth Inc., 696 F. App’x 792, 794 n.1 (9th Cir. 2017). We\\nleave that issue for another day. Instead, we vacate the\\ninjunction as to Corizon and remand with instructions to the\\ndistrict court to modify the injunction to exclude Corizon.\\nSee Azar, 911 F.3d at 585. Doing so still provides Edmo the\\nrelief she seeks at this stage. 22\\n\\n\\n\\n\\n 22\\n For similar reasons, we need not reach Edmo’s contention and the\\ndistrict court’s finding that “Corizon and IDOC have a de facto policy or\\npractice of refusing” GCS to prisoners. Edmo, 358 F. Supp. 3d at 1127.\\n\\f78 EDMO V. CORIZON\\n\\n C. Relief Ordered\\n\\n The State next contends that the injunctive relief ordered\\nis overbroad because it requires the State to provide Edmo\\nall “adequate medical care.” The State misconstrues the\\ndistrict court’s order. The order, read in context, requires\\ndefendants to provide GCS, as well as “adequate medical\\ncare” that is “reasonably necessary” to accomplish that\\nend—not every conceivable form of adequate medical care.\\nEdmo, 358 F. Supp. 3d at 1129; see also id. at 1109\\n(“Plaintiff Adree Edmo alleges that prison authorities\\nviolated her Eighth Amendment rights by refusing to provide\\nher with gender confirmation surgery. For the reasons\\nexplained below, the Court agrees and will order defendants\\nto provide her with this procedure, a surgery which is\\nconsidered medically necessary under generally accepted\\nstandards of care.”); id. at 1110 (“[F]or the reasons explained\\nin detail below, IDOC and Corizon will be ordered to\\nprovide Ms. Edmo with gender confirmation surgery.”).\\n\\n The State similarly contends that the injunctive relief\\nordered is overbroad because it requires the State to provide\\nEdmo surgery even though the defendants are not surgeons\\nand no surgeon has evaluated Edmo. We reject this obtuse\\nreading of the district court’s order. The district court\\nordered the State to “take all actions reasonably necessary to\\nprovide Ms. Edmo gender confirmation surgery.” Edmo,\\n358 F. Supp. 3d at 1129. That means that the State must take\\nsteps within its power to provide GCS to Edmo, such as\\nfinding a surgeon and scheduling a surgical evaluation.\\nIndeed, we modified our stay of the district court’s order to\\npermit a surgical consultation, which went forward in April\\n2019. Oral Arg. at 12:00–12:10. The State cannot\\nreasonably understand the district court’s December 13,\\n2018 order to require that the defendants themselves provide\\n\\f EDMO V. CORIZON 79\\n\\nsurgery. To the extent there are issues arising from a surgical\\nevaluation, the State can raise those issues with the district\\ncourt. 23\\n\\n V. Challenges to the Procedure Used by the District\\n Court\\n\\n Finally, the State contends that the district court\\nimproperly converted an evidentiary hearing on a\\npreliminary injunction into a final trial on the merits of\\nEdmo’s Eighth Amendment claim for GCS without giving\\nthem adequate notice and in violation of their Seventh\\nAmendment right to a jury trial. We address and reject each\\ncontention.\\n\\n A. Notice\\n\\n We first address the State’s contention that the district\\ncourt erroneously converted the evidentiary hearing into a\\nfinal trial on the merits without giving the State “clear and\\nunambiguous notice.” Under Federal Rule of Civil\\nProcedure 65(a)(2), “[a] district court may consolidate a\\npreliminary injunction hearing with a trial on the merits, but\\nonly when it provides the parties with clear and\\nunambiguous notice [of the intended consolidation] either\\nbefore the hearing commences or at a time which will afford\\n\\n 23\\n The State contends for the first time in its reply brief that the\\ninjunctive relief ordered was inappropriate because the WPATH\\nStandards of Care require two referrals from qualified mental health\\nprofessionals who have independently assessed the patient before GCS\\nmay be provided. It similarly contends for the first time in its reply in\\nsupport of its motion to dismiss that the order is overbroad because it\\ndoes not specify the type of GCS ordered. Because the State did not\\npresent these arguments in its opening brief, we do not consider them.\\nSee Smith v. Marsh, 194 F.3d 1045, 1052 (9th Cir. 1999).\\n\\f80 EDMO V. CORIZON\\n\\nthe parties a full opportunity to present their respective\\ncases.” Isaacson v. Horne, 716 F.3d 1213, 1220 (9th Cir.\\n2013) (second alteration in original) (quotation omitted).\\n“What constitutes adequate notice depends upon the facts of\\nthe case.” Michenfelder v. Sumner, 860 F.2d 328, 337 (9th\\nCir. 1988).\\n\\n A party challenging consolidation must show not only\\ninadequate notice, but also “substantial prejudice in the\\nsense that [it] was not allowed to present material evidence.”\\nMichenfelder, 860 F.2d at 337; see also 11A Charles Alan\\nWright et al., Federal Practice and Procedure § 2950 (3d\\ned. Apr. 2019 update). “We have on occasion upheld a\\ndistrict court’s failure to give any notice whatsoever before\\nfinally determining the merits after only a preliminary\\ninjunction hearing, where the complaining party has failed\\nto show how additional evidence could have altered the\\noutcome.” Michenfelder, 860 F.2d at 337.\\n\\n At the outset, we note that the State was provided notice,\\ntwice, that the district court considered the evidentiary\\nhearing a final trial on the merits of Edmo’s request for GCS.\\nAt the beginning of the hearing, the district court explained\\n“it’s hard for me to envision this hearing being anything but\\na hearing on a final injunction at least as to that part of the\\nrelief requested [GCS],” and it asked the parties to address\\nby the end of the hearing whether it was for a permanent\\ninjunction. At the close of the hearing, the district court\\nagain questioned whether it could order GCS in a\\npreliminary injunction. It explained that it had, in effect,\\n“kind of treated this hearing as the final hearing” on Edmo’s\\nrequest for GCS, and it again asked the parties to address in\\ntheir oral closings or written briefs whether the hearing was\\none for a permanent injunction. The State never answered\\nthe court’s question or objected to consolidation, despite the\\n\\f EDMO V. CORIZON 81\\n\\ndistrict court specifically noting it had treated the hearing as\\nfinal. Cf. Reilly v. United States, 863 F.2d 149, 160 (1st Cir.\\n1988) (“[W]hen a trial judge announces a proposed course\\nof action which litigants believe to be erroneous, the parties\\ndetrimentally affected must act expeditiously to call the error\\nto the judge’s attention or to cure the defect, not lurk in the\\nbushes waiting to ask for another trial when their litigatory\\nmilk curdles.”). This is not a case where the district court\\ngave no notice whatsoever.\\n\\n Regardless, the State has not shown any prejudice. With\\nfull awareness of the stakes, the district court permitted the\\nparties four months of discovery and held a three-day\\nevidentiary hearing. The parties called seven witnesses,\\nsubmitted declarations in lieu of live testimony for other\\nwitnesses, and submitted thousands of pages of exhibits and\\nextensive pre- and post-trial briefing. Most importantly,\\nboth parties put on extensive evidence concerning the\\ntreatment provided to and withheld from Edmo and why it\\nwas or was not appropriate—the key issue at the hearing.\\n\\n When it comes to identifying prejudice, the State is\\ntellingly short on specifics. It indicates that it “would have\\nobjected” to consolidation, but it failed to do so despite\\nrepeated invitations—indeed, directives—to address the\\nissue. The State also urges that it would have requested that\\nthe named defendants be able to testify live, but it\\nstipulated—knowing full well the stakes of the hearing—to\\nsubmit certain testimony via declaration “[i]n lieu of and/or\\nin addition to live testimony.” Moreover, the State fails to\\nidentify what testimony those witnesses would have offered\\nor explain how presenting that testimony live, instead of via\\ndeclaration, “could have altered the outcome.”\\nMichenfelder, 860 F.2d at 337. The district court did not\\n\\f82 EDMO V. CORIZON\\n\\ncommit reversible error in consolidating the evidentiary\\nhearing with a trial on the merits of Edmo’s request for GCS.\\n\\n B. Seventh Amendment\\n\\n We turn to the State’s related contention that the district\\ncourt violated the defendants’ Seventh Amendment right to\\na jury trial by converting the evidentiary hearing into a trial\\non the merits. We review that contention de novo. Palmer\\nv. Valdez, 560 F.3d 965, 968 (9th Cir. 2009).\\n\\n The Seventh Amendment guarantees the right to a trial\\nby jury “[i]n Suits at common law, where the value in\\ncontroversy shall exceed twenty dollars.” U.S. Const.\\namend. VII. In a case such as this, where legal claims are\\njoined with equitable claims, a party “has a right to jury\\nconsideration of all legal claims, as well as all issues\\ncommon to both claims.” Plummer v. W. Int’l Hotels Co.,\\n656 F.2d 502, 504 n.6 (9th Cir. 1981) (citing Curtis v.\\nLoether, 415 U.S. 189, 196 n.11 (1974)). “Otherwise, the\\ncourt might limit the parties’ opportunity to try to a jury\\nevery issue underlying the legal claims by affording\\npreclusive effect to its own findings of fact on questions that\\nare common to both the legal and equitable claims.” Lacy v.\\nCook County, 897 F.3d 847, 858 (7th Cir. 2018).\\n\\n Like other constitutional rights, the right to a jury trial in\\ncivil suits can be waived. See United States v. Moore,\\n340 U.S. 616, 621 (1951). It is well established that “[a]\\nfailure to object to a proceeding in which the court sits as the\\nfinder of fact waives a valid jury demand as to any claims\\ndecided in that proceeding, at least where it was clear that\\nthe court intended to make fact determinations.” Fillmore v.\\nPage, 358 F.3d 496, 503 (7th Cir. 2004) (quotation omitted);\\nsee also 9 Wright & Miller, Federal Practice and Procedure\\n§ 2321 (“The right to jury trial also may be waived as it has\\n\\f EDMO V. CORIZON 83\\n\\nin many, many cases, by conduct, such as failing to object to\\nor actually participating in a bench trial . . . .”).\\n\\n For example, in White v. McGinnis, we held that “[a]\\nparty’s vigorous participation in a bench trial, without so\\nmuch as a mention of a jury, . . . can only be ascribed to\\nknowledgeable relinquishment of the prior jury demand.”\\n903 F.2d 699, 703 (9th Cir. 1990) (en banc). We explained\\nthat where a party chooses “to argue his case fully before the\\ndistrict judge[,] it is not unjust to hold him to that\\ncommitment.” Id. By contrast, we have held that “[w]hen a\\nparty participates in [a] bench trial ordered by the trial court\\nwhile continuing to demand a jury trial, his ‘continuing\\nobjection’ is ‘sufficient to preserve his right to appeal the\\ndenial of his request for a jury.’” Solis v. County of Los\\nAngeles, 514 F.3d 946, 957 (9th Cir. 2008) (quoting United\\nStates v. Nordbrock, 941 F.2d 947, 950 (9th Cir. 1991)).\\n“This is because the party in such a case is not seeking ‘two\\nbites at the procedural apple’ . . . . Rather, when a trial court\\ndenies a party a jury trial despite the party’s continuing\\ndemand, the party has little choice but to accede to the trial\\ncourt’s ruling and participate in the bench trial.” Id. (citation\\nomitted); see also Lovelace v. Dall, 820 F.2d 223, 228 (7th\\nCir. 1987) (“Another policy justifying the jury demand\\nwaiver rule is the view that it is unfair to permit a party to\\nhave a trial, discover that it has lost, and then raise the jury\\nissue because it is unsatisfied with the result of the trial.”).\\n\\n The State seeks a second bite at the apple. It vigorously\\nparticipated in the evidentiary hearing without ever raising\\nthe right to a jury trial. The State remained silent in the face\\nof statements from the district court that it was considering\\ntreating, and then that it had treated, the hearing as a final\\ntrial on the merits, which made it clear that the court\\n“intended to make fact determinations.” Fillmore, 358 F.3d\\n\\f84 EDMO V. CORIZON\\n\\nat 503. It also remained silent despite the district court\\nasking twice whether the hearing was one for a permanent\\ninjunction—as clear a time as any to raise any concerns\\nabout a jury trial.\\n\\n The State raised the issue of a jury trial for the first time\\non appeal, after the district court ruled against it. Even after\\nthe district court’s ruling, the State made no objection or\\nclaim to a jury trial. This conduct waived the State’s right to\\na jury trial with respect to issues common to Edmo’s request\\nfor an injunction ordering GCS and her legal claims.\\n\\n VI. Conclusion\\n\\n We apply the dictates of the Eighth Amendment today in\\nan area of increased social awareness: transgender health\\ncare. We are not the first to speak on the subject, nor will\\nwe be the last. Our court and others have been considering\\nEighth Amendment claims brought by transgender prisoners\\nfor decades. During that time, the medical community’s\\nunderstanding of what treatments are safe and medically\\nnecessary to treat gender dysphoria has changed as more\\ninformation becomes available, research is undertaken, and\\nexperience is gained. The Eighth-Amendment inquiry takes\\naccount of that developing understanding. See Estelle,\\n429 U.S. at 102–03.\\n\\n We hold that where, as here, the record shows that the\\nmedically necessary treatment for a prisoner’s gender\\ndysphoria is gender confirmation surgery, and responsible\\nprison officials deny such treatment with full awareness of\\nthe prisoner’s suffering, those officials violate the Eighth\\nAmendment’s prohibition on cruel and unusual punishment.\\n\\f EDMO V. CORIZON 85\\n\\n ***\\n\\n We affirm the district court’s entry of an injunction for\\nEdmo. However, we vacate the injunction to the extent it\\napplies to Corizon, Yordy, Siegert, Dr. Young, Dr. Craig,\\nand Dr. Whinnery, in their individual capacities, and remand\\nto the district court to modify the injunction accordingly.\\nThe district court shall also modify the injunction to\\nsubstitute Al Ramirez in his official capacity as Warden of\\nISCI for Yordy.\\n\\n Although we addressed this appeal on an expedited\\nbasis, it has been more than a year since doctors concluded\\nthat GCS is medically necessary for Edmo. We urge the\\nState to move forward. We emphatically do not speak to\\nother cases, but the facts of this case call for expeditious\\neffectuation of the injunction.\\n\\n In light of the nature and urgency of the relief at issue,\\nwe will disfavor any motion, absent extraordinary\\ncircumstances or consent from all parties, to extend the\\nperiod to petition for rehearing or rehearing en banc. Our\\nstay of the district court’s December 13, 2018 order shall\\nautomatically terminate upon issuance of the mandate.\\n\\n Costs on appeal are awarded to Edmo.\\n\\n AFFIRMED IN PART, VACATED IN PART, AND\\nREMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4432772/\", \"author_raw\": \"B. Lynn Winmill, Chief District Judge, Presiding\"}]}"},"judge1":{"kind":"string","value":"M MARGARET MCKEOWN"},"judge2":{"kind":"string","value":"RONALD M GOULD"},"judge3":{"kind":"string","value":"ROBERT S LASNIK"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4655519/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: Adree Edmo v. Corizon, Inc.\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-23\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n ADREE EDMO, AKA Mason Edmo, No. 19-35017\n Plaintiff-Appellee,\n D.C. No.\n v. 1:17-cv-00151-\n BLW\n CORIZON, INC.; SCOTT ELIASON;\n MURRAY YOUNG; CATHERINE\n WHINNERY,\n Defendants-Appellants,\n\n and\n\n IDAHO DEPARTMENT OF\n CORRECTIONS; AL RAMIREZ, in his\n official capacity as warden of Idaho\n State Correctional Institution; *\n HENRY ATENCIO; JEFF ZMUDA;\n HOWARD KEITH YORDY; RICHARD\n CRAIG; RONA SIEGERT,\n Defendants.\n\n\n ADREE EDMO, AKA Mason Edmo, No. 19-35019\n Plaintiff-Appellee,\n\n\n\n\n *\n Al Ramirez is substituted in his official capacity for his\npredecessor, Howard Keith Yordy, pursuant to Rule 43(c)(2) of the\nFederal Rules of Appellate Procedure.\n\f2 EDMO V. CORIZON\n\n\n v. D.C. No.\n 1:17-cv-00151-\n IDAHO DEPARTMENT OF BLW\n CORRECTIONS; AL RAMIREZ, in his\n official capacity as warden of Idaho\n State Correctional Institution; HENRY OPINION\n ATENCIO; JEFF ZMUDA; HOWARD\n KEITH YORDY; RICHARD CRAIG;\n RONA SIEGERT,\n Defendants-Appellants,\n\n and\n\n CORIZON, INC.; SCOTT ELIASON;\n MURRAY YOUNG; CATHERINE\n WHINNERY,\n Defendants.\n\n Appeal from the United States District Court\n for the District of Idaho\n B. Lynn Winmill, Chief District Judge, Presiding\n\n Argued and Submitted May 16, 2019\n San Francisco, California\n\n Filed August 23, 2019\n\n Before: M. Margaret McKeown and Ronald M. Gould,\n Circuit Judges, and Robert S. Lasnik, ** District Judge.\n\n Per Curiam Opinion\n\n **\n The Honorable Robert S. Lasnik, United States District Judge for\nthe Western District of Washington, sitting by designation.\n\f EDMO V. CORIZON 3\n\n SUMMARY ***\n\n\n Eighth Amendment / Prisoner Rights\n\n The panel affirmed the district court’s entry of a\npermanent injunction in favor of Idaho state prisoner Adree\nEdmo, but vacated the injunction to the extent it applied to\ndefendants Corizon, Howard Yordy, Rona Siegert, Dr.\nYoung, Dr. Craig, and Dr. Whinnery, in their individual\ncapacities, in Edmo’s action seeking medical treatment for\ngender dysphoria.\n\n The district court concluded that Edmo had established\nher Eighth Amendment claim. The district court further\nconcluded that gender confirmation surgery (“GCS”) was\nmedically necessary for Edmo, and ordered the State to\nprovide the surgery.\n\n The panel credited the district court’s factual findings as\nlogical and well-supported, and held that the responsible\nprison authorities were deliberately indifferent to Edmo’s\ngender dysphoria, in violation of the Eighth Amendment.\nThe panel held that the record, as construed by the district\ncourt, established that Edmo had a serious medical need, that\nthe appropriate medical treatment was GCS, and that prison\nauthorities had not provided that treatment despite full\nknowledge of Edmo’s ongoing and extreme suffering and\nmedical needs. The panel rejected the State’s position that\nthere was a reasoned disagreement between qualified\nmedical professionals. The panel emphasized that its\n\n\n ***\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\f4 EDMO V. CORIZON\n\nanalysis was individual to Edmo, and rested on the record of\nthis case.\n\n Addressing further aspects of the appeal, the panel\nrejected the State’s contention that the district court did not\nmake the Prison Litigation Reform Act’s requisite “need-\nnarrowness-intrusiveness” findings, causing the injunction\nto automatically expire and mooting the appeal. The panel\nheld that the district court’s order, considered as a whole,\nmade all the findings required by 18 U.S.C. § 3626(a)(1)(A),\nand Ninth Circuit precedent. The panel also held that the\npermanent injunction that the district court entered had not\nexpired, and remained in place, albeit stayed. The panel\naccordingly denied the State’s motion to dismiss.\n\n The panel held that the district court did not err in\ngranting a permanent injunction. Specifically, the panel\nheld, based on the district court’s factual findings, that Edmo\nestablished her Eighth Amendment claim and that she will\nsuffer irreparable harm – in the form of ongoing mental\nanguish and possible physical harm – if GCS is not provided.\nThe State did not dispute that Edmo’s gender dysphoria was\na sufficiently serious medical need to trigger the State’s\nobligations under the Eighth Amendment. The panel held\nthat the district court did not err in crediting the testimony of\nEdmo’s experts that GCS was medically necessary to treat\nEdmo’s gender dysphoria and that the State’s failure to\nprovide that treatment was medically unacceptable. The\npanel further held that the district court did not err in\ndiscrediting the State’s experts because aspects of their\nopinions were illogical and unpersuasive. Also, the panel\nheld that the record demonstrated that Dr. Eliason acted with\ndeliberate indifference to Edmo’s serious medical needs.\nThe panel noted that its decision was in tension with the Fifth\nCircuit’s decision in Gibson v. Collier, 920 F.3d 212 (5th\n\f EDMO V. CORIZON 5\n\nCir. 2019), and the panel rejected that decision’s categorical\nholding that denying GCS cannot, as a matter of law, violate\nthe Eighth Amendment.\n\n The panel held that the district court did not err in finding\nthat Edmo would be irreparably harmed absent an\ninjunction. The panel rejected the State’s contentions as to\nwhy the district court erred in this finding.\n\n The panel next considered the State’s challenges to the\nscope of the injunction. The panel held that the injunction\nwas properly entered against Dr. Eliason because he\npersonally participated in the deprivation of Edmo’s\nconstitutional rights. The panel also held that because Edmo\nmay properly pursue her Eighth Amendment claim for\ninjunctive relief against Attencio, Zmuda and Ramirez in\ntheir official capacities, they were properly included within\nthe scope of the district court’s injunction. On remand, the\ndistrict court shall amend the injunction to substitute the\ncurrent warden as a party for Yordy. The panel vacated the\ndistrict court’s injunction to the extent it applied to Yordy,\nSiegert, Dr. Young, Dr. Craig, and Dr. Whinnery in their\nindividual capacities because the evidence in the record was\ninsufficient to conclude that they were deliberately\nindifferent to Edmo’s serious medical needs. The panel\nvacated the injunction as to Corizon, and remanded with\ninstructions to the district court to modify the injunction to\nexclude Corizon. Finally, the panel held that the injunctive\nrelief ordered was not overbroad.\n\n The panel considered the State’s challenges to the\nprocedure used by the district court. The panel rejected the\nState’s contention that the district court erroneously\nconverted the evidentiary hearing into a final trial on the\nmerits without giving proper notice. The panel held that the\nState did receive notice, and in any event, the State had not\n\f6 EDMO V. CORIZON\n\nshown any prejudice. The panel also rejected the State’s\ncontention that the district court violated defendants’\nSeventh Amendment right to a jury trial by converting the\nevidentiary hearing into a trial on the merits. The panel held\nthat the State’s conduct waived its right to a jury trial with\nrespect to issues common to Edmo’s request for an\ninjunction ordering GCS and her legal claims.\n\n\n COUNSEL\n\nBrady J. Hall (argued), Special Deputy Attorney General;\nLawrence G. Wasden, Attorney General; Office of the\nAttorney General, Boise, Idaho; Marisa S. Crecelius, Moore\nElia Kraft & Hall LLP, Boise, Idaho; for Defendants-\nAppellants Idaho Department of Corrections, Henry\nAtencio, Jeff Zmuda, Howard Keith Yordy, Richard Craig,\nand Rona Siegert.\n\nDylan A. Eaton (argued), J. Kevin West, and Bryce Jensen,\nParsons Behle & Latimer, Boise, Idaho, for Defendants-\nAppellants Corizon, Inc.; Scott Eliason; Murray Young; and\nCatherine Whinnery.\n\nLori Rifkin (argued), Hadsell Stormer & Renick LLP,\nEmeryville, California; Dan Stormer and Shaleen Shanbhag,\nHadsell Stormer & Renick LLP, Pasadena, California; Craig\nDurham and Deborah Ferguson, Ferguson Durham PLLC,\nBoise, Idaho; Amy Whelan and Julie Wilensky, National\nCenter for Lesbian Rights, San Francisco, California; for\nPlaintiff-Appellee.\n\nDavid M. Shapiro, Sheila A. Bedi, and Vanessa del Valle,\nRoderick & Solange MacArthur Justice Center, Chicago,\nIllinois; Molly E. Whitman, Akin Gump Strauss Hauer &\n\f EDMO V. CORIZON 7\n\nFeld LLP, Dallas, Texas; for Amici Curiae Andrea\nArmstrong, Sharon Dolovich, Betsy Ginsberg, Michael B.\nMushlin, Alexander A. Reinert, Laura Rovner, and Margo\nSchlanger.\n\nMolly Kafka and Richard Alan Eppink, ACLU of Idaho\nFoundation, Boise, Idaho; Devon A. Little and Derek\nBorchardt, Walden Macht & Haran LLP, New York, New\nYork; Amy Fettig and Jennifer Wedekind, ACLU National\nPrison Project, Washington, D.C.; Gabriel Arkles and Rose\nSaxe, ACLU LGBT & HIV Project/ACLU Foundation, New\nYork, New York; for Amici Curiae Former Corrections\nOfficials.\n\nDevi M. Rao and Jason T. Perkins, Jenner & Block LLP,\nWashington, D.C., for Amici Curiae Medical and Mental\nHealth Professional Organizations.\n\nSharif E. Jacob, Ryan K. M. Wong, Kristin E. Hucek, and\nPatrick E. Murray, Keker Van Nest & Peters LLP, San\nFrancisco, California, for Amicus Curiae Jody L. Herman.\n\nAlan E. Schoenfeld, Wilmer Cutler Pickering Hale and Dorr\nLLP, New York, New York; Michael Posada, Wilmer Cutler\nPickering Hale and Dorr LLP, Washington, D.C.; Richard\nSaenz, Lambda Legal Defense & Education Fund Inc., New\nYork, New York; A. Chinyere Ezie, Center for\nConstitutional Rights, New York, New York; for Amici\nCuriae Civil Rights & Non-Profit Organizations.\n\f8 EDMO V. CORIZON\n\n OPINION\n\nPER CURIAM:\n\n The Eighth Amendment prohibits “cruel and unusual\npunishments.” U.S. Const. amend. VIII. “The Amendment\nembodies broad and idealistic concepts of dignity, civilized\nstandards, humanity, and decency . . . .” Estelle v. Gamble,\n429 U.S. 97, 102 (1976) (quotation omitted). Our society\nrecognizes that prisoners “retain the essence of human\ndignity inherent in all persons.” Brown v. Plata, 563 U.S.\n493, 510 (2011).\n\n Consistent with the values embodied by the Eighth\nAmendment, for more than 40 years the Supreme Court has\nheld that “deliberate indifference to serious medical needs”\nof prisoners constitutes cruel and unusual punishment.\nEstelle, 429 U.S. at 106. When prison authorities do not\nabide by their Eighth Amendment duty, “the courts have a\nresponsibility to remedy the resulting . . . violation.” Brown,\n563 U.S. at 511. We do so here.\n\n Adree Edmo (formerly Mason Dean Edmo) is a male-to-\nfemale transgender prisoner in the custody of the Idaho\nDepartment of Correction (“IDOC”). Edmo’s sex assigned\nat birth (male) differs from her gender identity (female). The\nincongruity causes Edmo to experience persistent distress so\nsevere it limits her ability to function. She has twice\nattempted self-castration to remove her male genitalia,\nwhich cause her profound anguish.\n\n Both sides and their medical experts agree: Edmo suffers\nfrom gender dysphoria, a serious medical condition. They\nalso agree that the appropriate benchmark regarding\ntreatment for gender dysphoria is the World Professional\nAssociation of Transgender Health Standards of Care for the\n\f EDMO V. CORIZON 9\n\nHealth of Transsexual, Transgender, and Gender\nNonconforming People (“WPATH Standards of Care”).\nAnd the State 1 does not seriously dispute that in certain\ncircumstances, gender confirmation surgery (“GCS”) can be\na medically necessary treatment for gender dysphoria. The\nparties’ dispute centers around whether GCS is medically\nnecessary for Edmo—a question we analyze with deference\nto the district court’s factual findings.\n\n Following four months of intensive discovery and a\nthree-day evidentiary hearing, the district court concluded\nthat GCS is medically necessary for Edmo and ordered the\nState to provide the surgery. Its ruling hinged on findings\nindividual to Edmo’s medical condition. The ruling also\nrested on the finding that Edmo’s medical experts testified\npersuasively that GCS was medically necessary, whereas\ntestimony from the State’s medical experts deserved little\nweight. In contrast to Edmo’s experts, the State’s witnesses\nlacked relevant experience, could not explain their\ndeviations from generally accepted guidelines, and testified\nillogically and inconsistently in important ways.\n\n The district court’s detailed factual findings were amply\nsupported by its careful review of the extensive evidence and\ntestimony. Indeed, they are essentially unchallenged. The\nappeal boils down to a disagreement about the implications\nof the factual findings.\n\n Crediting, as we must, the district court’s logical, well-\nsupported factual findings, we hold that the responsible\n 1\n In addition to IDOC, Edmo sued Corizon, Inc. (a private for-profit\ncorporation that provides health care to inmates in IDOC custody) and\nvarious employees of IDOC and Corizon. The defendants briefed the\ncase jointly, and for ease of reference we refer to them collectively as\n“the State.”\n\f10 EDMO V. CORIZON\n\nprison authorities have been deliberately indifferent to\nEdmo’s gender dysphoria, in violation of the Eighth\nAmendment. The record before us, as construed by the\ndistrict court, establishes that Edmo has a serious medical\nneed, that the appropriate medical treatment is GCS, and that\nprison authorities have not provided that treatment despite\nfull knowledge of Edmo’s ongoing and extreme suffering\nand medical needs. In so holding, we reject the State’s\nportrait of a reasoned disagreement between qualified\nmedical professionals. We also emphasize that the analysis\nhere is individual to Edmo and rests on the record in this\ncase. We do not endeavor to project whether individuals in\nother cases will meet the threshold to establish an Eighth\nAmendment violation. The district court’s order entering\ninjunctive relief for Edmo is affirmed, with minor\nmodifications noted below.\n\n Our opinion proceeds as follows. In Part I, we provide\nbackground on gender dysphoria, the standard of care, and\nthe evidence considered and factual findings made by the\ndistrict court. Part II explains why this appeal complies with\nthe Prison Litigation Reform Act (“PLRA”) and is not moot.\nIn Part III, we turn to the gravamen of the appeal: Edmo’s\nEighth Amendment claim and showing of irreparable injury.\nPart IV addresses the State’s challenges to the injunction’s\nscope and narrows the injunction as to certain defendants.\nPart V rejects the State’s objections to the procedure\nemployed by the district court. We conclude in Part VI.\n\f EDMO V. CORIZON 11\n\n I. Background 2\n\n A. Gender Dysphoria and its Treatment\n\n Transgender individuals have a “[g]ender identity”—a\n“deeply felt, inherent sense” of their gender—that does not\nalign with their sex assigned at birth. 3 Am. Psychol. Ass’n,\nGuidelines for Psychological Practice with Transgender\nand Gender Nonconforming People, 70 Am. Psychologist\n832, 834 (2015). Recent estimates suggest that\napproximately 1.4 million transgender adults live in the\nUnited States, or 0.6 percent of the adult population.\nAndrew R. Flores et al., The Williams Inst., How Many\nAdults Identify as Transgender in the United States?, at\n2 (2016), http://williamsinstitute.law.ucla.edu/wp-content/\nuploads/How-Many-Adults-Identify-as-Transgender-in-the-\nUnited-States.pdf.\n\n Gender dysphoria 4 is “[d]istress that is caused by a\ndiscrepancy between a person’s gender identity and that\nperson’s sex assigned at birth (and the associated gender role\nand/or primary and secondary sex characteristics).” World\nProf’l Ass’n for Transgender Health, Standards of Care for\nthe Health of Transsexual, Transgender, and Gender-\n\n 2\n The following sections are derived from the district court’s factual\nfindings and the record on appeal.\n 3\n At birth, infants are classified as male or female based on visual\nobservation of their external genitalia. This is a person’s “sex assigned\nat birth,” but it may not be the person’s gender identity.\n 4\n Until recently, the medical community commonly referred to\ngender dysphoria as “gender identity disorder.” See Kosilek v. Spencer,\n774 F.3d 63, 68 n.1 (1st Cir. 2014).\n\f12 EDMO V. CORIZON\n\nNonconforming People 2 (7th ed. 2011) (hereinafter\n“WPATH SOC”). The Fifth Edition of the American\nPsychiatric Association’s Diagnostic and Statistical Manual\nof Mental Disorders (“DSM-5”) sets forth two conditions\nthat must be met for a person to be diagnosed with gender\ndysphoria. 5\n\n First, there must be “[a] marked incongruence between\none’s experienced/expressed gender and assigned gender, of\nat least 6 months’ duration, as manifested by at least two of\nthe following”:\n\n (1) “a marked incongruence between one’s\n experienced/expressed gender and primary\n and/or secondary sex characteristics”;\n\n (2) “a strong desire to be rid of one’s primary\n and/or secondary sex characteristics because\n of a marked incongruence with one’s\n experienced/expressed gender”;\n\n (3) “a strong desire for the primary and/or\n secondary sex characteristics of the other\n gender”;\n\n (4) “a strong desire to be of the other gender”;\n\n (5) “a strong desire to be treated as the other\n gender”; or\n\n\n\n\n 5\n Each expert in the case used these criteria to determine whether\nEdmo has gender dysphoria.\n\f EDMO V. CORIZON 13\n\n (6) “a strong conviction that one has the\n typical feelings and reactions of the other\n gender.”\n\nAm. Psychiatric Ass’n, Diagnostic and Statistical Manual of\nMental Disorders 452 (5th ed. 2013) (hereinafter “DSM-5”).\nSecond, the person’s condition must be associated with\n“clinically significant distress”—i.e., distress that impairs or\nseverely limits the person’s ability to function in a\nmeaningful way and has reached a threshold that requires\nmedical or surgical intervention, or both. Id. at 453, 458.\nNot every transgender person has gender dysphoria, and not\nevery gender dysphoric person has the same medical needs.\n\n Gender dysphoria is a serious but treatable medical\ncondition. Left untreated, however, it can lead to debilitating\ndistress, depression, impairment of function, substance use,\nself-surgery to alter one’s genitals or secondary sex\ncharacteristics, self-injurious behaviors, and even suicide.\n\n The district court found that the World Professional\nAssociation of Transgender Health Standards of Care for the\nHealth of Transsexual, Transgender, and Gender\nNonconforming People (“WPATH Standards of Care”) 6\n“are the internationally recognized guidelines for the\ntreatment of individuals with gender dysphoria.” Edmo v.\nIdaho Dep’t of Corr., 358 F. Supp. 3d 1103, 1111 (D. Idaho\n2018). Most courts agree. See, e.g., De’lonta v. Johnson,\n708 F.3d 520, 522–23 (4th Cir. 2013); Keohane v. Jones,\n328 F. Supp. 3d 1288, 1294 (N.D. Fla. 2018), appeal filed,\n\n 6\n The WPATH Standards of Care were formerly referred to as the\n“Harry Benjamin Standards of Care” and were promulgated by WPATH\nunder its former name, the “Harry Benjamin International Gender\nDysphoria Association.” Kosilek, 774 F.3d at 70 & n.3.\n\f14 EDMO V. CORIZON\n\nNo. 18-14096 (11th Cir. 2018); Norsworthy v. Beard, 87 F.\nSupp. 3d 1164, 1170 (N.D. Cal.), appeal dismissed &\nremanded, 802 F.3d 1090 (9th Cir. 2015); Soneeya v.\nSpencer, 851 F. Supp. 2d 228, 231–32 (D. Mass. 2012). But\nsee Gibson v. Collier, 920 F.3d 212, 221 (5th Cir. 2019)\n(“[T]he WPATH Standards of Care reflect not consensus,\nbut merely one side in a sharply contested medical debate\nover [GCS].”); cf. Kosilek, 774 F.3d at 76–79 (recounting\ntestimony questioning the WPATH Standards of Care). And\nmany of the major medical and mental health groups in the\nUnited States—including the American Medical\nAssociation, the American Medical Student Association, the\nAmerican Psychiatric Association, the American\nPsychological Association, the American Family Practice\nAssociation, the Endocrine Society, the National\nAssociation of Social Workers, the American Academy of\nPlastic Surgeons, the American College of Surgeons, Health\nProfessionals Advancing LGBTQ Equality, the HIV\nMedicine Association, the Lesbian, Bisexual, Gay and\nTransgender Physician Assistant Caucus, and Mental Health\nAmerica—recognize the WPATH Standards of Care as\nrepresenting the consensus of the medical and mental health\ncommunities regarding the appropriate treatment for\ntransgender and gender dysphoric individuals.\n\n Each expert in this case relied on the WPATH Standards\nof Care in rendering an opinion. As the State acknowledged\nto the district court, the WPATH Standards of Care “provide\nthe best guidance,” and “are the best standards out there.”\n“There are no other competing, evidence-based standards\nthat are accepted by any nationally or internationally\nrecognized medical professional groups.” Edmo, 358 F.\nSupp. 3d at 1125.\n\f EDMO V. CORIZON 15\n\n “[B]ased on the best available science and expert\nprofessional consensus,” the WPATH Standards of Care\nprovide “flexible clinical guidelines” “to meet the diverse\nhealth care needs of transsexual, transgender, and gender\nnonconforming people.” WPATH SOC at 1–2. Treatment\nunder the WPATH Standards of Care must be\nindividualized: “[w]hat helps one person alleviate gender\ndysphoria might be very different from what helps another\nperson.” Id. at 5. “Clinical departures from the [WPATH\nStandards of Care] may come about because of a patient’s\nunique anatomic, social, or psychological situation; an\nexperienced health professional’s evolving method of\nhandling a common situation; a research protocol; lack of\nresources in various parts of the world; or the need for\nspecific harm reduction strategies.” Id. at 2.\n\n The WPATH Standards of Care identify the following\nevidence-based treatment options for individuals with\ngender dysphoria:\n\n (1) “changes in gender expression and role\n (which may involve living part time or full\n time in another gender role, consistent with\n one’s gender identity)”;\n\n (2) “psychotherapy (individual, couple,\n family, or group) for purposes such as\n exploring gender identity, role, and\n expression[,] addressing the negative impact\n of gender dysphoria and stigma on mental\n health[,] alleviating internalized\n transphobia[,] enhancing social and peer\n support[,] improving body image[,] or\n promoting resilience”;\n\f16 EDMO V. CORIZON\n\n (3) “hormone therapy to feminize or\n masculinize the body”; and\n\n (4) “surgery to change primary and/or\n secondary sex characteristics (e.g.,\n breasts/chest, external and/or internal\n genitalia, facial features, body contouring).”\n\nId. at 10. The WPATH Standards of Care state that many\nindividuals “find comfort with their gender identity, role,\nand expression without surgery.” Id. at 54. For others,\nhowever, “surgery is essential and medically necessary to\nalleviate their gender dysphoria.” Id. That group cannot\nachieve “relief from gender dysphoria . . . without\nmodification of their primary and/or secondary sex\ncharacteristics to establish greater congruence with their\ngender identity.” Id. at 55; see also Jae Sevelius & Valerie\nJenness, Challenges and Opportunities for Gender-\nAffirming Healthcare for Transgender Women in Prison,\n13 Int’l J. Prisoner Health 32, 36 (2017) (“Negative\noutcomes such as genital self-harm, including autocastration\nand/or autopenectomy, can arise when gender-affirming\nsurgeries are delayed or denied.”); George R. Brown &\nEverett McDuffie, Health Care Policies Addressing\nTransgender Inmates in Prison Systems in the United States,\n15 J. Corr. Health Care 280, 287–88 (2009) (describing the\nauthors’ “firsthand knowledge of completed autocastration\nand/or autopenectomy in six facilities in four states”).\n\n The weight of opinion in the medical and mental health\ncommunities agrees that GCS is safe, effective, and\nmedically necessary in appropriate circumstances. See, e.g.,\nU.S. Dep’t of Health & Human Servs., No. A-13-87,\nDecision No. 2576, (Dep’t Appeals Bd. May 30, 2014);\nRandi Ettner, et al., Principles of Transgender Medicine and\n\f EDMO V. CORIZON 17\n\nSurgery 109–11 (2d ed. 2016); Jordan D. Frey, et al., A\nHistorical Review of Gender-Affirming Medicine: Focus on\nGenital Reconstruction Surgery, 14 J. Sexual Med. 991, 991\n(2017); Cynthia S. Osborne & Anne A. Lawrence, Male\nPrison Inmates With Gender Dysphoria: When Is Sex\nReassignment Surgery Appropriate?, 45 Archives of Sexual\nBehav. 1649, 1651–53 (2016); see also De’lonta, 708 F.3d\nat 523 (“Pursuant to the Standards of Care, after at least one\nyear of hormone therapy and living in the patient’s identified\ngender role, sex reassignment surgery may be necessary for\nsome individuals for whom serious symptoms persist. In\nthese cases, the surgery is not considered experimental or\ncosmetic; it is an accepted, effective, medically indicated\ntreatment for [gender dysphoria].”).\n\n The WPATH criteria for genital reconstruction surgery\nin male-to-female patients include the following:\n\n (1) “persistent, well documented gender\n dysphoria”;\n\n (2) “capacity to make a fully informed\n decision and to consent for treatment”;\n\n (3) “age of majority in a given country”;\n\n (4) “if significant medical or mental health\n concerns are present, they must be well\n controlled”;\n\n (5) “12 continuous months of hormone\n therapy as appropriate to the patient’s gender\n goals”; and\n\f18 EDMO V. CORIZON\n\n (6) “12 continuous months of living in a\n gender role that is congruent with their\n gender identity.”\n\nWPATH SOC at 60. The parties’ dispute focuses on\nwhether Edmo satisfied the fourth and sixth criteria.\n\n With respect to the fourth criterion, the WPATH\nStandards of Care provide that coexisting medical or mental\nhealth concerns unrelated to the person’s gender dysphoria\ndo not necessarily preclude surgery. Id. at 25. But those\nconcerns need to be managed prior to, or concurrent with,\ntreatment of a person’s gender dysphoria. Id. Coexisting\nmedical or mental health issues resulting from a person’s\ngender dysphoria are not an impediment under the fourth\ncriterion. It may be difficult to determine, however, whether\nmental or medical health concerns result from the gender\ndysphoria or are unrelated.\n\n The WPATH Standards of Care explain that the sixth\ncriterion—living for 12 months in an identity-congruent\nrole—is intended to ensure that the person experiences the\nfull range of “different life experiences and events that may\noccur throughout the year.” Id. at 61. During that time, the\npatient should present consistently in her desired gender\nrole. Id.\n\n Scientific studies show that the regret rate for individuals\nwho undergo GCS is low, in the range of one to two percent.\nSee, e.g., Osborne & Lawrence, Male Prison Inmates With\nGender Dysphoria, 45 Archives of Sexual Behav. at 1660;\nWilliam Byne, et al., Report of the American Psychiatric\nAssociation Task Force on Treatment of Gender Identity\nDisorder, 41 Archives of Sexual Behav. 759, 780–81\n(2012). The district court found, and the State does not\ndispute on appeal, that Edmo does not have any of the risk\n\f EDMO V. CORIZON 19\n\nfactors that would make her likely to regret GCS. See Edmo,\n358 F. Supp. 3d at 1121.\n\n The WPATH Standards of Care apply equally to all\nindividuals “irrespective of their housing situation” and\nexplicitly state that health care for transgender individuals\n“living in an institutional environment should mirror that\nwhich would be available to them if they were living in a\nnon-institutional setting within the same community.”\nWPATH SOC at 67. The next update to the WPATH\nStandards of Care will likewise apply equally to incarcerated\npersons. The National Commission on Correctional Health\nCare (“NCCHC”), a leading professional organization in\nhealth care delivery in the correctional context, endorses the\nWPATH Standards of Care as the accepted standards for the\ntreatment of transgender prisoners.\n\n In summary, the broad medical consensus in the area of\ntransgender health care requires providers to individually\ndiagnose, assess, and treat individuals’ gender dysphoria,\nincluding for those individuals in institutionalized\nenvironments. Treatment can and should include GCS when\nmedically necessary. Failure to follow an appropriate\ntreatment plan can expose transgender individuals to a\nserious risk of psychological and physical harm. The State\ndoes not dispute these points; it contends that GCS is not\nmedically necessary for Edmo.\n\n B. Edmo’s Treatment\n\n Edmo is a transgender woman in IDOC custody. Her sex\nassigned at birth was male, but she identifies as female. In\nher words, “my brain typically operates female, even though\nmy body hasn’t corresponded with my brain.”\n\f20 EDMO V. CORIZON\n\n Edmo has been incarcerated since pleading guilty in\n2012 to sexual abuse of a 15-year-old male at a house party.\nEdmo was 21 years old at the time of the criminal offense.\nEdmo is currently incarcerated at the Idaho State\nCorrectional Institution (“ISCI”). At the time of the\nevidentiary hearing, she was 30 years old and due to be\nreleased from prison in 2021.\n\n Edmo has viewed herself as female since age 5 or 6. She\nstruggled with her gender identity as a child and teenager,\npresenting herself intermittently as female, but around age\n20 or 21 she began living fulltime as a woman.\n\n Although she identified as female from an early age,\nEdmo first learned the term “gender dysphoria” and the\ncontours of that diagnosis around the time of her\nincarceration. Shortly thereafter, Corizon psychiatrist\nDr. Scott Eliason diagnosed her with “gender identity\ndisorder,” now referred to as gender dysphoria. Corizon\npsychologist Dr. Claudia Lake confirmed that diagnosis.\n\n While incarcerated, Edmo has changed her legal name to\nAdree Edmo and the sex on her birth certificate to “female”\nto affirm her gender identity. Throughout her incarceration,\nEdmo has consistently presented as female, despite receiving\nmany disciplinary offense reports for doing so. For example,\nwhen able to do so, Edmo has worn her hair in feminine\nhairstyles and worn makeup, for which she has received\nmultiple disciplinary offense reports. 7 Medical providers\nhave documented Edmo’s feminine presentation since 2012.\n\n\n 7\n Before the evidentiary hearing, Edmo tried to receive access to\nfemale commissary items, such as women’s underwear. Most of her\nrequests were denied. On the eve of the evidentiary hearing, IDOC\n\f EDMO V. CORIZON 21\n\n Neither the parties nor their experts dispute that Edmo\nsuffers from gender dysphoria. That dysphoria causes Edmo\nto feel “depressed,” “disgusting,” “tormented,” and\n“hopeless.”\n\n To alleviate Edmo’s gender dysphoria, prison officials\nhave, since 2012, provided hormone therapy. Edmo has\nfollowed and complied with her hormone therapy regimen,\nwhich helps alleviate her gender dysphoria to some extent.\nThe hormones “clear[] [her] mind” and have resulted in\nbreast growth, body fat redistribution, and changes in her\nskin. Today, Edmo is hormonally confirmed, which means\nthat she has the hormones and secondary sex characteristics\n(characteristics, such as women’s breasts, that appear during\npuberty but are not part of the reproductive system) of an\nadult female. Edmo has gained the maximum physical\nchanges associated with hormone treatment.\n\n Hormone therapy has not completely alleviated Edmo’s\ngender dysphoria. Edmo continues to experience significant\ndistress related to gender incongruence. Much of that\ndistress is caused by her male genitalia. Edmo testified that\nshe feels “depressed, embarrassed, [and] disgusted” by her\nmale genitalia and that this is an “everyday reoccurring\nthought.” Her medical records confirm her disgust, noting\nrepeated efforts by Edmo to purchase underwear to keep, in\nEdmo’s words, her “disgusting penis” out of sight.\n\n In addition to her gender dysphoria, Edmo suffers from\nmajor depressive disorder with anxiety and drug and alcohol\naddiction, although her addiction has been in remission\n\n\namended its policy concerning the treatment of gender dysphoric\nprisoners to increase transgender women’s access to female commissary\nitems.\n\f22 EDMO V. CORIZON\n\nwhile incarcerated. Edmo has taken her prescribed\nmedications for depression and anxiety. Prison officials\nhave also provided Edmo mental health treatment to help her\nwork through her serious underlying mental health issues\nand a pre-incarceration history of trauma, abuse, and suicide\nattempts. Edmo sees her psychiatrist when scheduled. But\nEdmo does not see her treating clinician, Krina Stewart,\nbecause Edmo does not believe Stewart is qualified to treat\nher gender dysphoria. Edmo has attended group therapy\nsessions inconsistently.\n\n In September 2015, Edmo attempted to castrate herself\nfor the first time using a disposable razor blade. 8 Before\ndoing so, she left a note to alert officials that she was not\n“trying to commit suicide,” and was instead “only trying to\nhelp [her]self.” Edmo did not complete the castration,\nthough she continued to report thoughts of self-castration in\nthe following months.\n\n On April 20, 2016, Dr. Eliason evaluated Edmo for GCS.\nAt the time, IDOC’s policy concerning the treatment of\ngender dysphoric prisoners provided that GCS “will not be\nconsidered for individuals within [IDOC], unless determined\nmedically necessary by” the treating physician. 9 Corizon’s\npolicy does not mention GCS.\n\n In his evaluation, Dr. Eliason noted that Edmo reported\nshe was “doing alright.” He also noted that Edmo had been\non hormone replacement therapy for the last year and a half,\nbut that she felt she needed more. He reported that Edmo\n 8\n She had previously reported thoughts of self-castration to\nclinicians.\n 9\n IDOC revised its policy shortly before the evidentiary hearing, but\nits revised policy contains functionally identical language.\n\f EDMO V. CORIZON 23\n\nhad stated that hormone replacement therapy helped\nalleviate her gender dysphoria, but she remained frustrated\nwith her male anatomy.\n\n Dr. Eliason indicated that Edmo appeared feminine in\ndemeanor and interaction style. He also indicated that Edmo\nhad previously attempted to “mutilate her genitalia” because\nof the severity of her distress. Dr. Eliason later testified that,\nat the time of his evaluation, he felt that Edmo’s gender\ndysphoria “had risen to another level,” as evidenced by her\nself-castration attempt.\n\n But Dr. Eliason also flagged that he had spoken to prison\nstaff about Edmo’s behavior and they explained it was\n“notable for animated affect and no observed distress.” He\nsimilarly noted that he had personally observed Edmo and\ndid not see significant dysphoria; instead, she “looked\npleasant and had a good mood.”\n\n As to GCS, Dr. Eliason explained in his notes that while\nmedical necessity for GCS is “not very well defined and is\nconstantly shifting,” in his view, GCS would be medically\nnecessary in at least three situations: (1) “congenital\nmalformations or ambiguous genitalia,” (2) “severe and\ndevastating dysphoria that is primarily due to genitals,” or\n(3) “some type of medical problem in which endogenous\nsexual hormones were causing severe physiological\ndamage.” Dr. Eliason concluded that Edmo “does not meet\nany of those . . . criteria” and, for that reason, GCS is not\nmedically necessary for her.\n\n Dr. Eliason instead concluded that hormone therapy and\nsupportive counseling suffice to treat Edmo’s gender\ndysphoria for the time being, despite recognizing that Edmo\nhad attempted self-castration on that regimen. Dr. Eliason\n\f24 EDMO V. CORIZON\n\nindicated that he would continue to monitor and assess\nEdmo.\n\n Dr. Eliason staffed Edmo’s evaluation with Dr. Jeremy\nStoddart, Dr. Murray Young, and Jeremy Clark, who all\nagreed with his assessment. They did not observe Edmo;\nrather, they agreed with Dr. Eliason’s recommended\ntreatment as he presented it to them. The record is sparse on\nthe qualifications of Dr. Stoddart and Dr. Young, but Clark\nhas never personally treated anyone with gender dysphoria\nand was not qualified under IDOC policy to assess whether\nGCS would be appropriate for Edmo.\n\n Dr. Eliason also discussed his evaluation with IDOC’s\nManagement and Treatment Committee (“MTC”), a multi-\ndisciplinary team composed of medical providers, mental\nhealth clinicians, IDOC’s Chief Psychologist, and prison\nleadership. The MTC meets periodically to evaluate and\naddress the unique medical, mental health, and housing\nneeds of prisoners with gender dysphoria. The committee\n“does not make any individual treatment decisions\nregarding” treatment for inmates with gender dysphoria.\n“Those determinations are made by the individual clinicians\nor the medical staff employed by Corizon.” The MTC\nagreed with Dr. Eliason’s assessment.\n\n Although not mentioned in his April 20, 2016 notes,\nDr. Eliason testified at the evidentiary hearing that he\nconsidered the WPATH Standards of Care when\ndetermining Edmo’s treatment. Citing those standards,\nDr. Eliason testified that he did not believe GCS was\nappropriate for two reasons: (1) because mental health\nissues separate from Edmo’s gender dysphoria were not\n“fully in adequate control” and (2) because Edmo had not\nlived in her identified gender role for 12 months outside of\nprison. He explained that Edmo needed to experience\n\f EDMO V. CORIZON 25\n\n“living as a woman” around “her real social network – her\nfamily and friends on the outside” so that she could\n“determine whether or not she felt like that was her real\nidentity.”\n\n Edmo was never evaluated for GCS again, but the MTC\nconsidered her gender dysphoria and treatment plan during\nlater meetings. The MTC continues to believe that GCS is\nnot medically necessary or appropriate for Edmo.\n\n In December 2016, Edmo tried to castrate herself for the\nsecond time. A medical note from the incident reports that\nEdmo said she no longer wanted her testicles. Edmo\nreported to medical providers that she was “feeling\nangry/frustrated that [she] was not receiving the help desired\nrelated to [her] gender dysphoria. Inmate Edmo’s actions\nwere reported as a method to stop/cease testosterone\nproduction in Edmo’s body. Edmo denied suicidal ideation\n. . . .”\n\n Edmo’s second attempt was more successful than the\nfirst. She was able to open her testicle sac with a razor blade\nand remove one testicle. She abandoned her attempt,\nhowever, when there was too much blood to continue. She\nthen sought medical assistance and was transported to a\nhospital, where her testicle was repaired. Edmo was\nreceiving hormone therapy both times she attempted self-\ncastration.\n\n Edmo testified that she was disappointed in herself for\ncoming so close but failing to complete her self-castration\nattempts. She also testified that she continues to actively\nthink about self-castration. To avoid acting on those\nthoughts and impulses, Edmo “self-medicat[es]” by cutting\nher arms with a razor. She says that the physical pain helps\n\f26 EDMO V. CORIZON\n\nto ease the “emotional torment” and mental anguish her\ngender dysphoria causes her.\n\n Edmo further testified that she expects GCS to help\nalleviate some of her gender dysphoria. In particular, she\ntestified that she expects GCS to help her avoid having “as\nmuch depression about myself and my physical body. I\ndon’t think I will be so anxious that people are always\nknowing I’m different . . . .” Edmo recognizes, however,\nthat GCS “is not a fix-all”: “[i]t’s not a magic operation. . . .\nI’m still going to have to face the same stressors that we all\nface in everyday life . . . .”\n\n C. Initiation of this Action\n\n Edmo filed a pro se complaint on April 6, 2017. She also\nmoved for a temporary restraining order, a preliminary\ninjunction, and the appointment of counsel.\n\n Edmo’s motion for appointment of counsel was granted\nin part, and counsel for Edmo appeared in June and August\n2017. Counsel withdrew Edmo’s pro se motion for\npreliminary injunction shortly thereafter.\n\n On September 1, 2017, Edmo filed an amended\ncomplaint asserting claims under 42 U.S.C. § 1983, the\nEighth Amendment, the Fourteenth Amendment, the\nAmericans with Disabilities Act, the Affordable Care Act,\nand for common law negligence. She named as defendants\nIDOC, Henry Atencio (Director of IDOC), Jeff Zmuda\n(Deputy Director of IDOC), Howard Keith Yordy (former\nWarden of ISCI), Dr. Richard Craig (Chief Psychologist at\nISCI), Rona Siegert (Health Services Director at ISCI),\nCorizon, Dr. Eliason, Dr. Young, and Dr. Catherine\nWhinnery (Corizon employee).\n\f EDMO V. CORIZON 27\n\n Through counsel, Edmo filed a renewed motion for a\npreliminary injunction on June 1, 2018. Among other relief,\nEdmo sought an order requiring the State to provide her with\na referral to a qualified surgeon and access to GCS.\n\n The State moved to extend the time to respond to Edmo’s\nmotion. After a status conference, the district court set an\nevidentiary hearing for October 10, 11, and 12, 2018. The\ncourt permitted the parties to undertake four months of\nextensive fact and expert discovery in preparation for the\nhearing.\n\n D. The Evidentiary Hearing\n\n At the evidentiary hearing, each side had eight hours to\npresent its case. The district court heard live testimony from\nseven witnesses over three days. It also considered\nthousands of pages of exhibits, including Edmo’s medical\nrecords. With the parties’ agreement, the court also\npermitted the State to submit declarations in lieu of live\ntestimony and permitted Edmo to impeach the declarations\nwith deposition testimony.\n\n At the outset of the hearing, the district court noted that\n“[w]e’re here on a hearing for a temporary injunction,” but\nit explained that “it’s hard for me to envision this hearing\nbeing anything but a hearing on a final injunction[,] at least\nas to” the injunctive relief ordering GCS. The court stated\nthat it was unsure whether that made a difference, and it\nasked the parties to address at some point whether the\nhearing was for a preliminary injunction or a permanent\ninjunction. Notably, the State did not do so.\n\n The district court heard testimony from three percipient\nwitnesses: Edmo, Dr. Eliason (the Corizon physician), and\nJeremy Clark (an IDOC clinician who did not meet IDOC’s\n\f28 EDMO V. CORIZON\n\ncriteria to assess Edmo for GCS). Their relevant testimony\nis largely recounted above.\n\n It also heard testimony from four expert witnesses, two\neach for Edmo and the State. Dr. Randi Ettner, Ph.D. in\npsychology, testified first for Edmo. Dr. Ettner is one of the\nauthors of the current (seventh) version of the WPATH\nStandards of Care. She has been a WPATH member since\n1993 and chairs its Institutionalized Persons Committee. Dr.\nEttner has authored or edited many peer-reviewed\npublications on the treatment of gender dysphoria and\ntransgender health care more broadly, including the leading\ntextbook used in medical schools on the subject. She also\ntrains medical and mental health providers on treating people\nwith gender dysphoria. Dr. Ettner has been retained as an\nexpert witness on gender dysphoria and its treatment in\nmany court cases, and she has been appointed as an\nindependent expert by one federal court to evaluate an\nincarcerated person for GCS.\n\n Dr. Ettner has evaluated, diagnosed, and treated between\n2,500 and 3,000 individuals with gender dysphoria. She has\nreferred about 300 people for GCS. She has also refused to\nrecommend surgery for some patients who have requested it.\nShe believes that not everyone who has gender dysphoria\nneeds GCS. Dr. Ettner also has “[e]xtensive experience”\ntreating and providing post-operative care for patients who\nhave undergone GCS.\n\n Dr. Ettner has assessed approximately 30 incarcerated\nindividuals with gender dysphoria for GCS and other\nmedical care, but she has not treated incarcerated patients.\nShe has not worked in a prison and she is not a Certified\nCorrectional Healthcare Professional.\n\f EDMO V. CORIZON 29\n\n Based on her evaluation of Edmo and a review of\nEdmo’s medical records, Dr. Ettner diagnosed Edmo with\ngender dysphoria, depressive disorder, anxiety, and suicidal\nideation. In Dr. Ettner’s opinion, GCS is medically\nnecessary for Edmo and should be immediately performed.\nShe explained that most patients with gender dysphoria do\nnot require GCS, but Edmo requires it because hormone\ntherapy has been inadequate for her and Edmo has attempted\nto remove her own testicles. Dr. Ettner further explained that\nGCS would give Edmo congruent genitalia, eliminating the\nsevere distress Edmo experiences due to her male anatomy.\n\n Dr. Ettner further opined that Edmo meets the WPATH\ncriteria for GCS. She explained that Edmo has “persistent\nand well-documented long-standing gender dysphoria”;\nEdmo “has no thought disorders and no impaired reality\ntesting”; Edmo is the age of majority in this country;\nalthough Edmo has depression and anxiety, those conditions\ndo not “impair her ability to undergo surgery” because they\nare “as controlled as [they] can be”; Edmo has had six years\nof hormone therapy; and Edmo has lived for more than one\nyear “as a woman to the best of her ability in a male prison.”\n\n More specifically, as to the fourth criterion, Dr. Ettner\nopined that Edmo does not have mental health concerns that\nwould preclude GCS. She explained that Edmo’s depression\nand anxiety are as “controlled as can be” because Edmo “is\ntaking the maximum amount of medication that controls\ndepression.” Dr. Ettner noted that Edmo has complied with\ntaking her prescribed medications and that psychotherapy is\nnot “a precondition for surgery” under the WPATH\nStandards of Care. She also flagged that Edmo has the\ncapacity to comply with her postsurgical treatment, as\nevidenced by her compliance with her hormone therapy to\ndate.\n\f30 EDMO V. CORIZON\n\n As to the clinical significance of Edmo’s self-castration\nattempts and cutting behaviors, Dr. Ettner explained that\nneither behavior indicates that Edmo has inadequately\ncontrolled mental health concerns. Rather, those behaviors\nindicate “the need for treatment for gender dysphoria.”\nDr. Ettner explained that\n\n when an individual who is not psychotic or\n delusional attempts what we call surgical\n self-treatment – because we don’t regard\n removal of the testicles or attempted removal\n of the testicles as either mutilation or self-\n harm – we regard it as an intentional attempt\n to remove the target organ that produces\n testosterone, which, in fact, is the cure for\n gender dysphoria.\n\nIn Dr. Ettner’s opinion, Edmo’s depression and anxiety “will\nbe attenuated post surgery.”\n\n Dr. Ettner opined that Edmo satisfies the sixth criterion\nbecause she has lived “as a woman to the best of her ability\nin a male prison.” Dr. Ettner based her opinion on Edmo’s\n“appearance . . . , her disciplinary records, which indicated\nthat she had attempted to wear her hair in a feminine\nhairstyle and to wear makeup even though that was against\nthe rules and she was – received some sort of disciplinary\naction for that, and her – the way that she was receiving\nfemale undergarments and had developed the stigma of\nfemininity, the secondary sex characteristics, breast\ndevelopment, et cetera.”\n\n Dr. Ettner opined that if Edmo does not receive GCS,\n“[t]he risks would be, as typical in inadequately treated or\nuntreated gender dysphoria, either surgical self-treatment,\nemotional decompensation, or suicide.” Dr. Ettner\n\f EDMO V. CORIZON 31\n\nexplained that Edmo “is at particular risk of suicide given\nthat she has a high degree of suicide ideation.” If, on the\nother hand, Edmo receives surgery, Dr. Ettner opined that\n\n [i]t would eliminate the gender dysphoria. It\n would provide a level of wellbeing that she\n hasn’t had previously. It would eliminate\n 80 percent of the testosterone in her body,\n necessitating a lower dose of hormones going\n forward, which would be particularly helpful\n given that she has elevated liver enzymes.\n And it would, I believe, eliminate much of the\n depression and the attendant symptoms that\n she is experiencing.\n\n Dr. Ryan Gorton, M.D., also testified for Edmo.\nDr. Gorton is an emergency medicine physician. He also\nworks pro bono at a clinic serving uninsured patients or those\nwith Medicare or Medicaid. Many of those patients have\nmental health conditions or have been in prison. He has\npublished peer-reviewed articles on the treatment of gender\ndysphoria, and he has been qualified as an expert witness in\ncases involving transgender health care. Dr. Gorton also\nprovides training on transgender health care issues to many\ngroups, is a member of WPATH, and serves on WPATH’s\nTransgender Medicine and Research Committee and its\nInstitutionalized Persons Committee.\n\n Dr. Gorton has been the primary care physician for about\n400 patients with gender dysphoria. At the time of the\nevidentiary hearing, Dr. Gorton was treating approximately\n100 patients with gender dysphoria. Dr. Gorton has assessed\npatients for gender dysphoria, initiated and monitored\nhormone treatment, referred patients for mental health\ntreatment, and determined the appropriateness of GCS. At\n\f32 EDMO V. CORIZON\n\nthe time of the evidentiary hearing, Dr. Gorton was\nproviding follow-up care for about 30 patients who had\nvaginoplasty. Dr. Gorton has no experience treating\ntransgender inmates and is not a Certified Correctional\nHealthcare Professional.\n\n Based on his review of Edmo’s medical records and his\nin-person evaluation of Edmo, Dr. Gorton opined that GCS\nis medically necessary for Edmo and that she meets the\nWPATH criteria for GCS. He explained that Edmo has\n“persistent well-documented gender dysphoria,” as shown in\nher prison medical records; she has the capacity “to make a\nfully informed decision and to consent for treatment”\nbecause “she didn’t seem at all impaired in her decision-\nmaking capacity”; she is the age of majority; she has\ndepression and anxiety, “but they are not to a level that\nwould preclude her getting [GCS]”; she had 12 consecutive\nmonths of hormone therapy; and she has been living in her\n“target gender role . . . despite an environment that’s very\nhostile to that and some negative consequences that she has\nexperienced because of that.”\n\n Dr. Gorton further opined that if Edmo “is not provided\nsurgery, there is a very substantial chance she will try to\nattempt self-surgery again. And that’s especially worrisome\ngiven her attempts have been progressive. . . . So I think she\nmight be successful” on her next attempt. He predicted that\nthere is little chance that Edmo’s gender dysphoria will\nimprove without surgery. Conversely, Dr. Gorton\nanticipated that Edmo is unlikely to regret surgery because\n“her gender dysphoria is very genital-focused” and regret\nrates among GCS patients are very low.\n\n Dr. Gorton also opined that Edmo’s self-castration\nattempts demonstrate “that she has severe genital-focused\ngender dysphoria and that she is not getting the medically\n\f EDMO V. CORIZON 33\n\nnecessary treatment to alleviate that.” He elaborated that\nEdmo’s depression and anxiety are not driving Edmo’s self-\ncastration attempts: “there [are] a lot of people with\ndepression and anxiety who don’t remove their testicles.”\n\n Finally, Dr. Gorton criticized Dr. Eliason’s evaluation of\nEdmo. He explained that he disagreed with Dr. Eliason’s\nconclusion that Edmo does not need GCS and he also\ndisagreed with the three “criteria” Dr. Eliason gave for when\nGCS would be necessary. Dr. Gorton criticized Dr.\nEliason’s first criterion—that GCS could be needed where\nthere is “congenital malformation or ambiguous genitalia”—\nbecause that situation “isn’t even germane to transgender\npeople”; rather, it relates to “people with intersex\nconditions.” As to the second criterion—that GCS could be\nneeded when a patient is suffering from “severe and\ndevastating gender dysphoria that is primarily due to\ngenitals”—Dr. Gorton pointed out that the WPATH\nStandards of Care for surgery require only “clear and\nsignificant dysphoria.” And even applying Dr. Eliason’s\nhigher bar, Dr. Gorton explained that Edmo would still\nqualify for GCS because she has twice attempted self-\ncastration, demonstrating “severe genital-focused\ndysphoria.” Finally, Dr. Gorton characterized Dr. Eliason’s\nthird criterion—that GCS could be needed in situations when\n“endogenous sexual hormones were causing severe\nphysiological damage”—as “bizarre.” Dr. Gorton could not\nconjure “a clinical circumstance where that would be the\ncase that your hormones that your body produces are\nattacking you . . . . I just don’t understand what [Dr. Eliason]\nis talking about there.”\n\n Dr. Keelin Garvey, M.D., testified for the State.\nDr. Garvey is a psychiatrist and Certified Correctional\nHealthcare Professional. As the former Chief Psychiatrist of\n\f34 EDMO V. CORIZON\n\nthe Massachusetts Department of Corrections, Dr. Garvey\nchaired the Gender Dysphoria Treatment Committee. She\ndirectly treated a “couple of patients” with gender dysphoria\nearlier in her career as Deputy Medical Director, but she has\nnot done so in recent years. Prior to evaluating Edmo, Dr.\nGarvey had never evaluated a patient in person to determine\nwhether that person needed GCS. Dr. Garvey has never\nrecommended a patient for GCS, and she has not done\nfollow-up care with a person who has received GCS.\n\n Based on her evaluation of Edmo and a review of\nEdmo’s medical records, Dr. Garvey diagnosed Edmo with\ngender dysphoria, major depressive disorder, alcohol use\ndisorder, stimulant use disorder, and opioid use disorder.\nShe explained that the latter three are in remission.\n\n Relying on the WPATH Standards of Care, Dr. Garvey\nopined that GCS is not medically necessary for Edmo. 10 Dr.\nGarvey first explained that Edmo does not meet the first\nWPATH Standards of Care criterion—“persistent, well\ndocumented gender dysphoria”—because of a lack of\nevidence in pre-incarceration medical records that Edmo\npresented as female before her time in prison. Dr. Garvey\nacknowledged, however, that Edmo has been presenting as\nfemale since 2012 and that she has been diagnosed with\ngender dysphoria since that time.\n\n Dr. Garvey then explained that Edmo does not meet the\nfourth criterion—“medical/mental health concerns must be\nwell controlled”—because Edmo “is actively self-injuring.”\nDr. Garvey elaborated that “self-injury in any form is never\n\n 10\n Dr. Garvey testified that she relies on the WPATH Standards of\nCare and the NCCHC guidelines adopting those standards when treating\ninmates with gender dysphoria.\n\f EDMO V. CORIZON 35\n\nconsidered a healthy or productive coping mechanism” and\nthat she would like to see Edmo “develop further coping\nskills that she would be able to use following surgery so that\nshe is not engaging in self-injury after surgery.” Dr.\nGarvey’s concern is that GCS is a “stressful undertaking”\nand Edmo lacks “effective coping strategies” to deal with the\nstress.\n\n Finally, Dr. Garvey testified that Edmo does not meet the\nsixth criterion—“12 continuous months of living in a gender\nrole that is congruent with gender identity”—because Edmo\nhas not presented as female outside of prison and “there [are]\nchallenges to using her time in a men’s prison as this real-\nlife experience because it doesn’t offer her the opportunity\nto actually experience all those things she is going to go\nthrough on the outside.”\n\n Dr. Joel Andrade, Ph.D. in social work, also testified for\nthe State. He is a licensed clinical social worker and is a\nCertified Correctional Healthcare Professional with an\nemphasis in mental health. Dr. Andrade has over a decade\nof experience providing and supervising the provision of\ncorrectional mental health care, including directing and\noverseeing the treatment of inmates diagnosed with gender\ndysphoria in the custody of the Massachusetts Department\nof Corrections in his roles as clinical director, chair of the\nGender Dysphoria Supervision Group, and member of the\nGender Dysphoria Treatment Committee.\n\n As a member of the Gender Dysphoria Treatment\nCommittee, Dr. Andrade recommended GCS for two\ninmates. But the recommendations were contingent on the\ninmates living in a women’s prison for approximately\n12 months before the surgery. The Massachusetts\nDepartment of Corrections, like IDOC, houses prisoners\n\f36 EDMO V. CORIZON\n\naccording to their genitals, so the inmates had not been\nmoved (nor had their surgery occurred).\n\n Dr. Andrade has never directly treated patients with\ngender dysphoria, nor has he been a treating clinician for a\npatient who has had GCS. His “experience with gender\ndysphoria comes almost exclusively from [his] participation\non the Massachusetts Department of Corrections[’] Gender\nDysphoria Treatment Committee and Supervision Group.”\nDr. Andrade did not qualify, under the IDOC gender\ndysphoria policy in effect at the time of his assessment of\nEdmo, to assess a person for GCS because he is neither a\npsychologist nor a physician.\n\n Based on his evaluation of Edmo and a review of her\nmedical records, Dr. Andrade diagnosed Edmo with “major\ndepressive disorder, recurrent, in partial remission,”\n“generalized anxiety disorder,” “alcohol use disorder,\nsevere,” and gender dysphoria. Dr. Andrade also diagnosed\nEdmo with borderline personality disorder. The district\ncourt did not credit this diagnosis, however, because no other\nperson (including the State’s other expert, Dr. Garvey) has\never diagnosed Edmo with borderline personality disorder\nand Dr. Andrade was unable to identify his criteria for this\ndiagnosis. Edmo, 358 F. Supp. 3d at 1120. The record\namply supports the district court’s finding in this respect.\n\n Dr. Andrade opined that Edmo does not meet the\nWPATH criteria for GCS. He explained that, based on his\nreview of Edmo’s pre-incarceration records, Edmo did not\npresent as female or discuss her gender dysphoria before\nincarceration. Dr. Andrade testified that he would like to see\nEdmo live as female outside of a correctional setting before\nreceiving GCS, or, at the least, live in a women’s prison first.\nIDOC, however, houses prisoners according to their\ngenitals. Dr. Andrade also explained that Edmo needs to\n\f EDMO V. CORIZON 37\n\nwork through some of her trauma, particularly sexual abuse\nthat she suffered, and other mental health concerns before\nreceiving surgery. Dr. Andrade opined that Edmo’s mental\nhealth issues will not be cured by GCS.\n\n At the close of the hearing, the district court reiterated\nthat it was unsure “how we can hear [Edmo’s request for\nGCS] on a preliminary injunction. . . . [I]f I order it, then it’s\ndone.” The court further suggested that the request for GCS\ncould “only be resolved in a final hearing” and noted that it\nhad, in effect, “treated this hearing as [a] final hearing on the\nissue.”\n\n The court, as it had done at the outset of the hearing,\nasked the parties to address whether the hearing was for a\npreliminary or permanent injunction. In response, Edmo\ncontended that the court could order GCS in a preliminary\ninjunction. The State did not address the court’s question. It\ninstead contended that the standard for a mandatory\ninjunction—which can be preliminary or permanent—\nshould apply.\n\n E. The District Court’s Decision\n\n The district court rendered its decision on December 13,\n2018. After recounting the evidence and making extensive\nfactual findings, the district court began its analysis by\nnoting that it was unsure whether the standard for a\npreliminary injunction or the standard for a permanent\ninjunction applied. The court noted that “the nature of the\nrelief requested in this case, coupled with the extensive\nevidence presented by the parties over a 3-day evidentiary\nhearing, [may have] effectively converted these proceedings\ninto a final trial on the merits of the plaintiff’s request for\npermanent injunctive relief.” Edmo, 358 F. Supp. 3d at 1122\nn.1. It also indicated that “both parties appear to have treated\n\f38 EDMO V. CORIZON\n\nthe evidentiary hearing” as a final trial on the merits. Id. The\ndistrict court explained that the difference was immaterial,\nhowever, because Edmo was entitled to relief under either\nstandard. Id.\n\n On the merits, the district court concluded that Edmo had\nestablished her Eighth Amendment claim. The district court\nfirst held that Edmo suffers from gender dysphoria, which is\nundisputedly “a serious medical condition.” Id. at 1124.\n\n It then concluded that GCS is medically necessary to\ntreat Edmo’s gender dysphoria. See id. at 1124–26. In a\ncarefully considered, 45-page opinion, the district court\nspecifically found “credible the testimony of Plaintiff’s\nexperts Drs. Ettner and Gorton, who have extensive personal\nexperience treating individuals with gender dysphoria both\nbefore and after receiving gender confirmation surgery,” and\nwho opined that GCS was medically necessary. Id. at 1125.\nThe court rejected the contrary opinions of the State’s\nexperts because “neither Dr. Garvey nor Dr. Andrade has\nany direct experience with patients receiving gender\nconfirmation surgery or assessing patients for the medical\nnecessity of gender confirmation surgery,” and neither of the\nState’s experts had meaningful “experience treating patients\nwith gender dysphoria other than assessing them for the\nexistence of the condition.” Id. The district court also noted\nthat the State’s “experts appear to misrepresent the WPATH\nStandards of Care by concluding that Ms. Edmo, despite\npresenting as female since her incarceration in 2012, cannot\nsatisfy the WPATH criteria because she has not presented as\nfemale outside of the prison setting.” Id. As the district\ncourt noted, “there is no requirement in the WPATH\nStandards of Care that a patient live for twelve months in his\nor her gender role outside of prison before becoming eligible\nfor” GCS. Id. (quotation omitted).\n\f EDMO V. CORIZON 39\n\n Finally, the district court explained that the State was\ndeliberately indifferent to Edmo’s gender dysphoria because\nit “fail[ed] to provide her with available treatment that is\ngenerally accepted in the field as safe and effective, despite\nher actual harm and ongoing risk of future harm including\nself-castration attempts, cutting, and suicidal ideation.” Id.\nat 1126–27. The district court also stated that the evidence\n“suggest[ed] that Ms. Edmo has not been provided gender\nconfirmation surgery because Corizon and IDOC have a de\nfacto policy or practice of refusing this treatment for gender\ndysphoria to prisoners,” which amounts to deliberate\nindifference. Id. at 1127.\n\n After analyzing the merits, the district court concluded\nthat Edmo satisfied the other prerequisites to injunctive\nrelief. Id. at 1127–28. The district court found that, given\nEdmo’s continuing emotional distress and self-castration\nattempts, “Edmo is at serious risk of life-threatening self-\nharm” if she does not receive GCS. Id. at 1128. The State,\non the other hand, had not shown that it would be harmed if\nordered to provide GCS, so the equities favored Edmo. Id.\n\n Having concluded that Edmo was entitled to an\ninjunction, the court ordered the State “to provide Plaintiff\nwith adequate medical care, including gender confirmation\nsurgery.” Id. at 1129. It ordered the State to “take all actions\nreasonably necessary to provide Ms. Edmo gender\nconfirmation surgery as promptly as possible and no later\nthan six months from the date of this order.” Id.\n\n F. Appellate Proceedings\n\n The State filed timely notices of appeal on January 9,\n2019. It also asked the district court to stay its order pending\nappeal. The district court denied the State’s motion on\nMarch 4.\n\f40 EDMO V. CORIZON\n\n The State then filed in this court a motion to stay pending\nappeal. A motions panel granted that motion. Edmo\nsubsequently moved to amend the stay to allow her to\nundergo a previously scheduled pre-surgery consultation.\nThe motions panel granted that motion and amended the\nstay.\n\n On April 3, the State filed an “urgent motion” to dismiss\nthis appeal as moot. We indicated on April 5 that our court\nwould consider that motion with the merits, not on an urgent\nbasis.\n\n After hearing oral argument on May 16, we ordered a\nlimited remand to the district court to clarify three points.\nRelevant here, we asked the district court to clarify whether\nit granted Edmo a permanent injunction in its December 13,\n2018 order. The district court clarified that it “granted\npermanent injunctive relief.” Edmo v. Idaho Dep’t of Corr.,\nNo. 1:17-CV-00151-BLW, 2019 WL 2319527, at *2 (D.\nIdaho May 31, 2019). We also asked the district court to\nclarify whether it had concluded that Edmo had succeeded\non the merits of her Eighth Amendment claim. The district\ncourt responded that it had. Id.\n\n Having received the district court’s response to our\nlimited remand order, we proceed to the issues on appeal.\nThe State challenges the district court’s grant of injunctive\nrelief to Edmo on multiple grounds. It contends that this\nappeal is moot because the injunction did not comply with\nthe PLRA and has, for that reason, automatically expired. It\ncontends that the decision not to provide GCS to Edmo\nreflects a difference of prudent medical opinion and cannot\nsupport an Eighth Amendment claim. It contends that Edmo\nwill not be irreparably harmed absent an injunction. It\ncontends that the injunction is overbroad. Finally, it\ncontends that, to the extent the district court converted the\n\f EDMO V. CORIZON 41\n\nevidentiary hearing into a final trial on the merits of Edmo’s\nrequest for GCS, it was provided inadequate notice and the\ncourt violated its right to a jury trial.\n\n II. Mootness\n\n “We first address, as we must, the question of mootness\n. . . .” Shell Offshore Inc. v. Greenpeace, Inc., 815 F.3d 623,\n628 (9th Cir. 2016). An appeal is moot “[w]hen events\nchange such that the appellate court can no longer grant ‘any\neffectual relief whatever to the prevailing party.’” Id.\n(quoting City of Erie v. Pap’s A.M., 529 U.S. 277, 287\n(2000)). In those circumstances, we “lack[] jurisdiction and\nmust dismiss the appeal.” Id.\n\n The State contends that the injunction does not comply\nwith provisions of the PLRA and, for that reason, has\nautomatically expired under the terms of the statute.\nRelevant here, the PLRA provides that a\n\n court shall not grant or approve any\n prospective relief unless the court finds that\n such relief is narrowly drawn, extends no\n further than necessary to correct the violation\n of the Federal right, and is the least intrusive\n means necessary to correct the violation of\n the Federal right. The court shall give\n substantial weight to any adverse impact on\n public safety or the operation of a criminal\n justice system caused by the relief.\n\n18 U.S.C. § 3626(a)(1)(A). Courts often refer to this\nprovision as the “need-narrowness-intrusiveness” inquiry.\nGraves v. Arpaio, 623 F.3d 1043, 1048 n.1 (9th Cir. 2010)\n(per curiam) (quoting Pierce v. County of Orange, 526 F.3d\n1190, 1205 (9th Cir. 2008)). The PLRA further provides that\n\f42 EDMO V. CORIZON\n\nany “[p]reliminary injunctive relief shall automatically\nexpire on the date that is 90 days after its entry, unless the\ncourt makes the findings required under subsection (a)(1)\n[quoted above] for the entry of prospective relief and makes\nthe order final before the expiration of the 90-day period.”\n18 U.S.C. § 3626(a)(2).\n\n The State contends that the district court did not make\nthe PLRA’s requisite need-narrowness-intrusiveness\nfindings or make its order final within 90 days, causing the\ninjunction to expire under 18 U.S.C. § 3626(a)(2).\nGenerally, the expiration of an injunction challenged on\nappeal moots the appeal. See Kitlutsisti v. ARCO Alaska,\nInc., 782 F.2d 800, 801 (9th Cir. 1986); see also United\nStates v. Sec’y, Fla. Dep’t of Corr., 778 F.3d 1223, 1228–29\n(11th Cir. 2015). The State asserts separate, albeit\noverlapping, contentions in their motion to dismiss this\nappeal and in their briefing. We reject those arguments.\n\n A. Need-Narrowness-Intrusiveness Findings\n\n The State first contends that the district court did not\nmake the PLRA’s need-narrowness-intrusiveness findings,\ncausing the injunction to automatically expire and mooting\nthis appeal. 11 As we have explained in prior decisions, the\nPLRA “has not substantially changed the threshold findings\nand standards required to justify an injunction.” Gomez v.\nVernon, 255 F.3d 1118, 1129 (9th Cir. 2001). When\n“determining the appropriateness of the relief ordered,”\nappellate “courts must do what they have always done”:\n\n 11\n We question whether the State’s need-narrowness-intrusiveness\nchallenge, properly understood, implicates mootness. But because the\nresult is the same, we accept the State’s framing for purposes of our\nanalysis.\n\f EDMO V. CORIZON 43\n\n“consider the order as a whole.” Armstrong v.\nSchwarzenegger, 622 F.3d 1058, 1070 (9th Cir. 2010).\nDistrict courts must make need-narrowness-intrusiveness\n“findings sufficient to allow a ‘clear understanding’ of the\nruling,” but they need not “make such findings on a\nparagraph by paragraph, or even sentence by sentence,\nbasis.” Id. (quotation omitted). “What is important, and\nwhat the PLRA requires, is a finding that the set of reforms\nbeing ordered—the ‘relief’—corrects the violations of\nprisoners’ rights with the minimal impact possible on\ndefendants’ discretion over their policies and procedures.”\nId.\n\n Here, the district court made the necessary need-\nnarrowness-intrusiveness findings. At the start of its\nDecember 13, 2018 order, the district court explained that\nany injunction must meet the PLRA’s need-narrowness-\nintrusiveness requirement. See Edmo, 358 F. Supp. 3d at\n1122. The district court then explained how the relief being\nordered, GCS, “corrects the violations of” Edmo’s rights.\nSee Armstrong, 622 F.3d at 1071. Specifically, the district\ncourt explained that GCS is medically necessary to alleviate\nEdmo’s gender dysphoria and that the State’s denial of GCS\namounts to deliberate indifference in violation of the Eighth\nAmendment. See Edmo, 358 F. Supp. 3d at 1116–21, 1123–\n27, 1129. The district court limited the relief ordered to have\n“the minimal impact possible on [the State’s] discretion over\ntheir policies and procedures.” See Armstrong, 622 F.3d at\n1071. Specifically, the district court limited the relief to\n“actions reasonably necessary” to provide GCS, cautioned\nthat its conclusion is based on “the unique facts and\ncircumstances presented” by Edmo, and noted that its\n“decision is not intended, and should not be construed, as a\ngeneral finding that all inmates suffering from gender\ndysphoria are entitled to [GCS].” Edmo, 358 F. Supp. 3d at\n\f44 EDMO V. CORIZON\n\n1110, 1129. Finally, the district court rejected the notion that\ninjunctive relief would have “any adverse impact on public\nsafety or the operation of a criminal justice system.”\n18 U.S.C. § 3626(a)(1)(A). It explained that the State had\n“made no showing that an order requiring them to provide”\nGCS to Edmo “causes them injury.” Edmo, 358 F. Supp. 3d\nat 1128. The district court’s order, considered as a whole,\nmade all the findings required by 18 U.S.C. § 3626(a)(1)(A)\nand our precedent. See Armstrong, 622 F.3d at 1070.\n\n B. Finality\n\n The State next argues that the injunction has\nautomatically expired under the PLRA because the district\ncourt did not make its order “final” within 90 days of\nentering injunctive relief. See 18 U.S.C. § 3626(a)(2); see\nalso Sec’y, Fla. Dep’t of Corr., 778 F.3d at 1228–29\n(holding that an appeal of a preliminary injunction was moot\nbecause the district court “did not issue an order finalizing\nits [preliminary-injunction] order,” and “[a]s a result, the\npreliminary injunction expired by operation of law” 90 days\nlater). The PLRA provision cited by the State applies to\npreliminary injunctive relief, not permanent injunctive\nrelief. See 18 U.S.C. § 3626(a)(2). The permanent\ninjunction that the district court entered has not expired. See\nEdmo, 358 F. Supp. 3d at 1122 n.1 (concluding that Edmo is\n“entitled to relief” under the permanent injunction standard);\nsee also Edmo, 2019 WL 2319527, at *2 (clarifying on\nlimited remand that the district court granted Edmo a\npermanent injunction). It remains in place, albeit stayed.\n\f EDMO V. CORIZON 45\n\n There is a live controversy on appeal. 12 We accordingly\nDENY the State’s motion to dismiss and proceed to the\nmerits of the appeal.\n\n III. Challenges to the District Court’s Grant of\n Injunctive Relief\n\n An injunction is an “extraordinary remedy never\nawarded as of right.” Winter v. Nat. Res. Def. Council, Inc.,\n555 U.S. 7, 24 (2008). “To be entitled to a permanent\ninjunction, a plaintiff must demonstrate: (1) actual success\non the merits; (2) that it has suffered an irreparable injury;\n(3) that remedies available at law are inadequate; (4) that the\nbalance of hardships justify a remedy in equity; and (5) that\nthe public interest would not be disserved by a permanent\ninjunction.” 13 Indep. Training & Apprenticeship Program\n\n\n\n 12\n Even construed as a preliminary injunction, the district court’s\nDecember 13, 2018 order is not moot. On May 31, 2019, the district\ncourt, incorporating its previous findings, renewed the injunction. See\nEdmo, 2019 WL 2319527, at *2. Because the district court renewed the\ninjunction, we can consider its merits. See Mayweathers v. Newland,\n258 F.3d 930, 935–36 (9th Cir. 2001) (holding that district courts may\nrenew preliminary injunctions under the PLRA while an appeal is\npending, and considering the merits of the renewed injunction). And we\nhave jurisdiction under 28 U.S.C. § 1292(a)(1) regardless of whether the\ndistrict court’s order is considered a preliminary or permanent\ninjunction. See Hendricks v. Bank of Am., N.A., 408 F.3d 1127, 1131\n(9th Cir. 2005) (preliminary injunction); TransWorld Airlines, Inc. v.\nAm. Coupon Exch., Inc., 913 F.2d 676, 680–81 (9th Cir. 1990)\n(permanent injunction where the “district court retained jurisdiction to\ndetermine damages” and to adjudicate a separate claim).\n 13\n We agree with the State that the injunction is mandatory, as\nopposed to prohibitory, because it requires the State to act. Based on that\ndistinction, the State argues that Edmo must satisfy a higher burden of\n\f46 EDMO V. CORIZON\n\nv. Cal. Dep’t of Indus. Relations, 730 F.3d 1024, 1032 (9th\nCir. 2013) (citing eBay Inc. v. MercExch., L.L.C., 547 U.S.\n388, 391 (2006)).\n\n We review for abuse of discretion the district court’s\ndecision to grant a permanent injunction. Ariz. Dream Act\nCoal. v. Brewer, 855 F.3d 957, 965 (9th Cir. 2017). We\n\n\nproof to be entitled to injunctive relief, and that the district court failed\nto hold Edmo to that burden. On that point, we disagree.\n\n The State errs by relying on cases that concern mandatory\npreliminary injunctions. Because mandatory preliminary injunctions go\n“well beyond simply maintaining the status quo [p]endente lite,” they are\n“particularly disfavored” and “are not issued in doubtful cases.” Marlyn\nNutraceuticals, Inc. v. Mucos Pharma GmbH & Co., 571 F.3d 873, 879\n(9th Cir. 2009) (alteration in original) (quoting Anderson v. United\nStates, 612 F.2d 1112, 1114–15 (9th Cir. 1980)). The calculus is\ndifferent in the context of permanent injunctions. A plaintiff must show\nactual success on the merits, see Amoco Prod. Co. v. Village of Gambell,\n480 U.S. 531, 546 n.12 (1987), so there is no concern that a mandatory\npermanent injunction will upset the status quo only for a later trial on the\nmerits to show that the plaintiff was not entitled to equitable relief. As a\nresult, a plaintiff need not show that “extreme or very serious damage\nwill result,” as is required for mandatory preliminary injunctions.\n\n As we have explained, the district court granted Edmo injunctive\nrelief under both the preliminary and permanent injunction standards.\nSee Edmo, 358 F. Supp. 3d at 1122 n.1; see also Edmo, 2019 WL\n2319527, at *2. Because the standard for granting permanent injunctive\nrelief is higher (in that it requires actual success on the merits) and the\nState contends in its opening brief that we should review the injunction\nas a permanent injunction, we consider whether the district court erred\nin granting Edmo permanent injunctive relief. But we would also affirm\nunder the mandatory preliminary injunction standard, because the district\ncourt correctly applied the proper standard for mandatory preliminary\ninjunctive relief, and not the lower standard for prohibitory preliminary\ninjunctions. See Edmo, 358 F. Supp. 3d at 1122, 1128.\n\f EDMO V. CORIZON 47\n\nreview “any determination underlying the grant of an\ninjunction by the standard that applies to that\ndetermination.” Ting v. AT&T, 319 F.3d 1126, 1134–35 (9th\nCir. 2003). Accordingly, the district court’s factual findings\non Edmo’s Eighth Amendment claim are reviewed for clear\nerror. See Graves, 623 F.3d at 1048. Clear error exists if the\nfinding is “illogical, implausible, or without support in\ninferences that may be drawn from the facts in the record.”\nLa Quinta Worldwide LLC v. Q.R.T.M., S.A. de C.V.,\n762 F.3d 867, 879 (9th Cir. 2014) (quoting Herb Reed\nEnters., LLC v. Florida Entm’t Mgmt., Inc., 736 F.3d 1239,\n1247 (9th Cir. 2013)). We review de novo the district court’s\n“conclusion that the facts . . . demonstrate an Eighth\nAmendment violation.” Hallett v. Morgan, 296 F.3d 732,\n744 (9th Cir. 2002).\n\n The State contends that the district court erred in\ngranting an injunction because (1) Edmo’s Eighth\nAmendment claim fails and (2) Edmo has not shown that she\nwill suffer irreparable injury in the absence of an\ninjunction. 14 We disagree. We hold, based on the district\ncourt’s factual findings, that Edmo established her Eighth\nAmendment claim and that she will suffer irreparable\nharm—in the form of ongoing mental anguish and possible\nphysical harm—if GCS is not provided.\n\n A. The Merits of Edmo’s Eighth Amendment Claim\n\n “[D]eliberate indifference to serious medical needs of\nprisoners” violates the Eighth Amendment. Estelle,\n429 U.S. at 104. Because “society takes from prisoners the\nmeans to provide for their own needs,” Brown, 563 U.S.\n\n 14\n Because the State does not contest the other injunction factors, we\ndo not address them.\n\f48 EDMO V. CORIZON\n\nat 510, the government has an “obligation to provide medical\ncare for those whom it is punishing by incarceration,”\nEstelle, 429 U.S. at 103.\n\n To establish a claim of inadequate medical care, a\nprisoner must first “show a ‘serious medical need’ by\ndemonstrating that ‘failure to treat a prisoner’s condition\ncould result in further significant injury or the ‘unnecessary\nand wanton infliction of pain.’” Jett v. Penner, 439 F.3d\n1091, 1096 (9th Cir. 2006) (quoting McGuckin v. Smith,\n974 F.2d 1050, 1059 (9th Cir. 1991), overruled on other\ngrounds by WMX Techs., Inc. v. Miller, 104 F.3d 1133 (9th\nCir. 1997) (en banc)). Serious medical needs can relate to\n“physical, dental and mental health.” Hoptowit v. Ray,\n682 F.2d 1237, 1253 (9th Cir. 1982), abrogated on other\ngrounds by Sandin v. Conner, 515 U.S. 472 (1995).\n\n The State does not dispute that Edmo’s gender dysphoria\nis a sufficiently serious medical need to trigger the State’s\nobligations under the Eighth Amendment. Nor could it.\nGender dysphoria is a “serious . . . medical condition” that\ncauses “clinically significant distress”—distress that impairs\nor severely limits an individual’s ability to function in a\nmeaningful way. DSM-5 at 453, 458. As Edmo testified,\nher gender dysphoria causes her to feel “depressed,”\n“disgusting,” “tormented,” and “hopeless,” and it has caused\npast efforts and active thoughts of self-castration. As this\nand many other courts have recognized, Edmo’s gender\ndysphoria is a sufficiently serious medical need to implicate\nthe Eighth Amendment. See Rosati v. Igbinoso, 791 F.3d\n1037, 1039–40 (9th Cir. 2015); Kosilek, 774 F.3d at 86;\nDe’lonta, 708 F.3d at 525; Battista v. Clarke, 645 F.3d 449,\n452 (1st Cir. 2011); Allard v. Gomez, 9 F. App’x 793, 794\n(9th Cir. 2001); White v. Farrier, 849 F.2d 322, 325 (8th Cir.\n1988); Meriwether v. Faulkner, 821 F.2d 408, 412 (7th Cir.\n\f EDMO V. CORIZON 49\n\n1987) (and cases cited therein); Norsworthy, 87 F. Supp. 3d\nat 1187; Konitzer v. Frank, 711 F. Supp. 2d 874, 905 (E.D.\nWis. 2010).\n\n If, as here, a prisoner establishes a sufficiently serious\nmedical need, that prisoner must then “show the [official’s]\nresponse to the need was deliberately indifferent.” Jett,\n439 F.3d at 1096. An inadvertent or negligent failure to\nprovide adequate medical care is insufficient to establish a\nclaim under the Eighth Amendment. Estelle, 429 U.S.\nat 105–06; see also Farmer v. Brennan, 511 U.S. 825, 835\n(1994) (“ordinary lack of due care” is insufficient to\nestablish an Eighth Amendment claim). In other words,\n“[m]edical malpractice does not become a constitutional\nviolation merely because the victim is a prisoner.” Estelle,\n429 U.S. at 106. To “show deliberate indifference, the\nplaintiff must show that the course of treatment the [official]\nchose was medically unacceptable under the circumstances\nand that the [official] chose this course in conscious\ndisregard of an excessive risk to the plaintiff’s health.”\nHamby v. Hammond, 821 F.3d 1085, 1092 (9th Cir. 2016)\n(quoting Snow v. McDaniel, 681 F.3d 978, 988 (9th Cir.\n2012), overruled in part on other grounds by Peralta v.\nDillard, 744 F.3d 1076 (9th Cir. 2014) (en banc)).\n\n 1. The Medical Necessity of GCS for Edmo\n\n The crux of the State’s appeal is that it provided adequate\nand medically acceptable care to Edmo.\n\n Accepted standards of care and practice within the\nmedical community are highly relevant in determining what\ncare is medically acceptable and unacceptable. See Allard v.\nBaldwin, 779 F.3d 768, 772 (8th Cir. 2015); Henderson v.\nGhosh, 755 F.3d 559, 566 (7th Cir. 2014) (per curiam).\nTypically, “[a] difference of opinion between a physician\n\f50 EDMO V. CORIZON\n\nand the prisoner—or between medical professionals—\nconcerning what medical care is appropriate does not\namount to deliberate indifference.” Snow, 681 F.3d at 987;\nsee also Gibson, 920 F.3d at 220. But that is true only if the\ndueling opinions are medically acceptable under the\ncircumstances. See Toguchi v. Chung, 391 F.3d 1051, 1058\n(9th Cir. 2004) (a mere “difference of medical opinion . . .\n[is] insufficient, as a matter of law, to establish deliberate\nindifference,” but not if the “chosen course of treatment ‘was\nmedically unacceptable under the circumstances’”\n(alterations in original) (quoting Jackson v. McIntosh,\n90 F.3d 330, 332 (9th Cir. 1996))).\n\n “In deciding whether there has been deliberate\nindifference to an inmate’s serious medical needs, we need\nnot defer to the judgment of prison doctors or\nadministrators.” Hunt v. Dental Dep’t, 865 F.2d 198, 200\n(9th Cir. 1989). Nor does it suffice for “correctional\nadministrators wishing to avoid treatment . . . simply to find\na single practitioner willing to attest that some well-accepted\ntreatment is not necessary.” Kosilek, 774 F.3d at 90 n.12. In\nthe final analysis under the Eighth Amendment, we must\ndetermine, considering the record, the judgments of prison\nmedical officials, and the views of prudent professionals in\nthe field, whether the treatment decision of responsible\nprison authorities was medically acceptable.\n\n Reviewing the record and the district court’s extensive\nfactual findings, we conclude that Edmo has established that\nthe “course of treatment” chosen to alleviate her gender\ndysphoria “was medically unacceptable under the\ncircumstances.” Hamby, 821 F.3d at 1092 (quoting Snow,\n681 F.3d at 988). This conclusion derives from the district\ncourt’s factual findings, which are not “illogical,\nimplausible, or without support in inferences that may be\n\f EDMO V. CORIZON 51\n\ndrawn from the facts in the record.” La Quinta Worldwide\nLLC, 762 F.3d at 879 (quotation omitted).\n\n In particular, and as we will explain, this is not a case of\ndueling experts, as the State paints it. The district court\npermissibly credited the opinions of Edmo’s experts that\nGCS is medically necessary to treat Edmo’s gender\ndysphoria and that the State’s failure to provide that\ntreatment is medically unacceptable. Edmo’s experts are\nwell-qualified to render such opinions, and they logically\nand persuasively explained the necessity of GCS and applied\nthe WPATH Standards of Care—the undisputed starting\npoint in determining the appropriate treatment for gender\ndysphoric individuals. On the other side of the coin, the\ndistrict court permissibly discredited the contrary opinions\nof the State’s treating physician and medical experts. Those\nindividuals lacked expertise and incredibly applied (or did\nnot apply, in the case of the State’s treating physician) the\nWPATH Standards of Care. In other words, the district court\ndid not clearly err in making its credibility determinations,\nso it is not our role to reevaluate them. The credited\ntestimony establishes that GCS is medically necessary.\n\n a. Expert Testimony\n\n Turning first to the expert testimony offered, the district\ncourt credited the testimony of Edmo’s experts that GCS is\nmedically necessary to treat Edmo’s gender dysphoria and\nthat the State’s failure to provide that treatment is medically\nunacceptable. See Edmo, 358 F. Supp. 3d at 1120–21, 1125.\nDr. Ettner and Dr. Gorton opined that GCS is medically\nnecessary because Edmo’s current treatment has been\ninadequate, as evidenced by her self-castration attempts.\nThey also opined that if Edmo does not receive GCS, there\nis little chance that her gender dysphoria will improve and\nshe is at risk of committing self-surgery again, suicide, and\n\f52 EDMO V. CORIZON\n\nfurther emotional decompensation. On the other hand,\nproviding GCS to Edmo would, in the opinions of Dr. Ettner\nand Dr. Gorton, align Edmo’s genitalia with her gender\nidentity, thereby eliminating the severe distress Edmo\nexperiences from her male genitalia.\n\n In sharp contrast, the district court gave “virtually no\nweight” to the opinions of the State’s experts. Edmo, 358 F.\nSupp. 3d at 1126. Dr. Garvey and Dr. Andrade, who\npurported to rely on the WPATH Standards of Care, opined\nthat GCS is not medically necessary for Edmo.\n\n The district court did not err in crediting the testimony of\nEdmo’s experts and discounting the testimony of the State’s\nexperts. Dr. Ettner and Dr. Gorton are well-qualified to\nopine on the medical necessity of GCS. Both have\nsubstantial experience treating individuals with gender\ndysphoria. Dr. Ettner has evaluated, diagnosed, and treated\nbetween 2,500 and 3,000 individuals with gender dysphoria,\nwhile Dr. Gorton has been the primary care physician for\napproximately 400 patients with gender dysphoria. Both\nhave substantial experience evaluating whether GCS is\nmedically necessary for patients. Dr. Ettner has evaluated\nhundreds of people for GCS, referring approximately 300\nwhile refusing others, and Dr. Gorton routinely determines\nthe appropriateness of GCS for patients. They also have\nexperience providing follow-up care for patients who have\nundergone GCS. And both have published peer-reviewed\narticles concerning the treatment of gender dysphoria.\n\n The State’s experts, by contrast, have substantial\nexperience providing health care in institutional settings, but\nlack meaningful experience directly treating people with\ngender dysphoria. Dr. Garvey directly treated a “couple of\npatients” with gender dysphoria early in her career, while\nDr. Andrade has never provided direct treatment for patients\n\f EDMO V. CORIZON 53\n\nwith gender dysphoria. Moreover, prior to evaluating Edmo,\nneither had ever evaluated someone in person to determine\nthe medical necessity of GCS. Relatedly, Dr. Garvey and\nDr. Andrade have never provided follow-up care for a person\nwho has received GCS. Indeed, Dr. Andrade did not even\nqualify under IDOC policy to assess a person for GCS. And\nneither Dr. Garvey nor Dr. Andrade has published a peer-\nreviewed article concerning the treatment of gender\ndysphoria.\n\n Neither Dr. Ettner nor Dr. Gorton have treated prisoners\nwith gender dysphoria, nor are they Certified Correctional\nHealthcare Professionals. But both serve on WPATH’s\nInstitutionalized Persons Committee, which “looks at the\ncare and the assessment of individuals who are incarcerated\nand develops standards for treatment” of such individuals.\nThey are thus familiar with medical treatment in prison\nsettings. Moreover, Dr. Ettner has assessed approximately\n30 incarcerated persons with gender dysphoria for GCS and\nother medical care.\n\n More to the point, the more relevant experience for\ndetermining the medical necessity of GCS is having treated\nindividuals with gender dysphoria, having evaluated\nindividuals for GCS, and having treated them post-\noperatively. Such experience lends itself to fundamental\nknowledge of whether GCS is necessary and the potential\nrisks of providing or foregoing the surgery. Edmo’s experts\nhave the requisite experience; the State’s experts do not. For\nthat reason alone, the district court did not clearly err in\ncrediting the opinions of Edmo’s experts over those of the\nState. 15 See Caro v. Woodford, 280 F.3d 1247, 1253 (9th\n\n 15\n The State contends that neither Dr. Ettner nor Dr. Gorton was\nqualified to offer expert opinions as to the appropriate medical care for\n\f54 EDMO V. CORIZON\n\nCir. 2002) (explaining that we “must afford the District\nCourt considerable deference in its determination that the\nwitnesses were qualified to draw [their] conclusions”).\n\n Independent of the experts’ qualifications, the district\ncourt did not err in crediting the opinions of Edmo’s experts\nover those of the State because aspects of Dr. Garvey’s and\nDr. Andrade’s opinions ran contrary to the established\nstandards of care in the area of transgender health care—the\nWPATH Standards of Care—which they purported to\napply. 16 See Edmo, 358 F. Supp. 3d at 1125.\n\n\nEdmo because neither is a psychiatrist. So far as we can discern, the\nargument is that because a psychiatrist (Dr. Eliason) evaluated Edmo for\nGCS, only other psychiatrists are qualified to opine as to the medical\nnecessity of GCS and to contradict his assessment. See Oral Arg.\nat 10:00–10:30. We reject that contention. Edmo’s experts, as\nexplained, have significant experience evaluating patients for GCS—\nprecisely what Dr. Eliason did. On the basis of their medical experience\ntreating persons with gender dysphoria, they are well-qualified to render\nan opinion on the medical necessity of GCS and whether failure to\nprovide the surgery is medically acceptable. See Fed. R. Evid. 702.\n 16\n The State contends that the district court erred in requiring strict\nadherence to the flexible WPATH Standards of Care and in concluding\nthat any deviation from those standards is medically unacceptable. But\nthe district court correctly recognized that the WPATH Standards of\nCare are flexible, see Edmo, 358 F. Supp. 3d at 1111, and it appropriately\nused them as a starting point to gauge the credibility of each expert’s\ntestimony, see id. at 1125–26. Tellingly, each expert for Edmo and the\nState likewise used the WPATH Standards of Care as a starting point.\nAs the district court recognized: “There are no other competing,\nevidence-based standards that are accepted by any nationally or\ninternationally recognized medical professional groups.” Id. at 1125.\nAnd as the State acknowledged at the evidentiary hearing, the “WPATH\nstandards of care in the seventh edition do provide the best guidance”\nand “are the best standards out there.” For these reasons, the WPATH\nStandards of Care establish a useful starting point for analyzing the\n\f EDMO V. CORIZON 55\n\n For example, both Dr. Garvey and Dr. Andrade\nexpressed the view that Edmo does not meet the sixth\nWPATH criterion, “12 continuous months of living in a\ngender role that is congruent with gender identity.” WPATH\nSOC at 60. They pointed out that Edmo has not presented\nas female outside of prison and urged that she needs real-life\nexperiences in the community before undergoing GCS.\n\n These opinions run head-on into the WPATH Standards\nof Care. The WPATH standards, which the NCCHC\nendorses as the accepted standards for the treatment of\ntransgender inmates, apply\n\n in their entirety . . . to all transsexual,\n transgender, and gender nonconforming\n people, irrespective of their housing\n situation. People should not be discriminated\n against in their access to appropriate health\n care based on where they live, including\n institutional environments such as prisons\n .... Health care for transsexual,\n transgender, and gender nonconforming\n people living in an institutional environment\n should mirror that which would be available\n to them if they were living in a non-\n institutional setting within the same\n community.\n\n\n\ncredibility and weight to be given to each expert’s opinion and whether\nthat opinion was consistent with established standards of care. The State\ndoes not contest the district court’s finding that the WPATH Standards\nof Care are the “internationally recognized guidelines for the treatment\nof individuals with gender dysphoria.” Id. at 1111. They are the gold\nstandard on this issue.\n\f56 EDMO V. CORIZON\n\n All elements of assessment and treatment as\n described in the [Standards of Care] can be\n provided to people living in institutions.\n Access to these medically necessary\n treatments should not be denied on the basis\n of institutionalization or housing\n arrangements.\n\nWPATH SOC at 67. Dr. Garvey and Dr. Andrade’s view—\nthat GCS cannot be medically indicated for transgender\ninmates who did not present in a gender-congruent manner\nbefore incarceration—contradicts these accepted standards.\nDr. Garvey and Dr. Andrade would deny GCS to a class of\npeople because of their “institutionalization,” which the\nWPATH Standards of Care explicitly disavow. They\nprovide no persuasive explanation for their deviation. 17 And\nnothing in the WPATH Standards of Care or the law\nsupports excluding an entire class of gender dysphoric\nindividuals from eligibility for GCS.\n\n Both Dr. Garvey and Dr. Andrade also relied on Edmo’s\nfailure to attend psychotherapy sessions as an indication that\nher mental health concerns are not well controlled. But\npsychotherapy is not a precondition for surgery under the\nWPATH Standards of Care. WPATH SOC at 28–29.\n\n We acknowledge that the WPATH Standards of Care are\nflexible, and a simple deviation from those standards does\nnot alone establish an Eighth Amendment claim. But the\n 17\n In concluding that Edmo does not meet the sixth WPATH\ncriterion, Dr. Garvey expressed concern that there is a lack of evidence\nregarding GCS in prison settings. That rationale acts as self-fulfilling\nprophecy. If prisons and prison officials deny GCS to prisoners because\nof a lack of data, the data will never be generated, and the cycle will\ncontinue.\n\f EDMO V. CORIZON 57\n\nState’s experts purported to be applying those standards and\nyet did so in a way that directly contradicted them. These\nunsupported and unexplained deviations offer a further\nreason why the district court did not clearly err in\ndiscounting the testimony of the State’s experts. See Caro,\n280 F.3d at 1253.\n\n Finally, the district court did not err in discrediting the\nState’s experts because aspects of their opinions were\nillogical and unpersuasive. For example, Dr. Garvey and\nDr. Andrade expressed the view that Edmo does not meet the\nfirst WPATH criterion—“persistent, well documented\ngender dysphoria,” WPATH SOC at 60—because of a lack\nof evidence from pre-incarceration records of Edmo\npresenting as female. But both experts acknowledged that\nEdmo has been diagnosed with and treated for gender\ndysphoria since 2012—i.e., for six years as of the evidentiary\nhearing. Neither Dr. Garvey nor Dr. Andrade questioned\nEdmo’s diagnosis, and both agree that she currently suffers\ngender dysphoria. There can be no doubt that Edmo has\n“persistent, well documented gender dysphoria,” so their\nopinion is inexplicable.\n\n Dr. Garvey’s and Dr. Andrade’s opinions on this point\nalso ignore that individuals with gender dysphoria do not\nalways experience symptoms early in life or throughout their\nlife, or do not identify them as such. As Dr. Ettner testified,\n“gender dysphoria intensifies with age.” And as with\ntreatment for any other medical condition, treatment for\ngender dysphoria must be based on a patient’s current\nsituation.\n\n The opinions of Edmo’s experts are notably devoid of\nthese flaws. Dr. Ettner and Dr. Gorton cogently and\npersuasively explained why GCS is medically necessary for\nEdmo and why Edmo meets the WPATH criteria for GCS.\n\f58 EDMO V. CORIZON\n\n For example, consistent with the WPATH Standards of\nCare, Dr. Ettner explained that Edmo has lived for\n“12 continuous months . . . in a gender role that is congruent\nwith gender identity” (the sixth WPATH criterion) because\nshe has lived “as a woman to the best of her ability in a male\nprison.” In support of her opinion, Dr. Ettner cited Edmo’s\n“appearance . . . , her disciplinary records, which indicated\nthat she had attempted to wear her hair in a feminine\nhairstyle and to wear makeup even though that was against\nthe rules and she was – received some sort of disciplinary\naction for that, and her – the way that she was receiving\nfemale undergarments and had developed the stigma of\nfemininity, the secondary sex characteristics, breast\ndevelopment, et cetera.” Dr. Gorton similarly explained that\nEdmo satisfies the sixth WPATH criterion because she has\nlived for years in her “target gender role . . . despite an\nenvironment that’s very hostile to that and some negative\nconsequences that she has experienced because of that.”\n\n Moreover, both Dr. Ettner and Dr. Gorton offered\nreasoned explanations tying Edmo’s self-castration attempts\nto her severe gender dysphoria. Dr. Ettner explained that\ndoctors regard “surgical self-treatment . . . as an intentional\nattempt to remove the target organ that produces\ntestosterone, which, in fact, is the cure for gender\ndysphoria.” As Dr. Gorton elaborated, Edmo’s self-\ncastration attempts demonstrate deficient treatment for\n“severe genital-focused gender dysphoria.” He rejected the\nnotion that Edmo’s depression and anxiety drove her self-\ncastration attempts: “there [are] a lot of people with\ndepression and anxiety who don’t remove their testicles.”\n\n In light of the experts’ backgrounds and experience, and\nthe reasonableness, consistency, and persuasiveness of their\nopinions, the district court did not err in crediting the\n\f EDMO V. CORIZON 59\n\nopinions of Edmo’s experts and giving little weight to those\nof the State’s experts. The district court carefully examined\nthe voluminous record, extensive testimony, and conflicting\nexpert opinions in this case and set forth clear reasons,\nsupported by the record, for relying on the testimony of\nEdmo’s experts. See La Quinta Worldwide, 762 F.3d at 879\n(a factual finding is clear error if it is “illogical, implausible,\nor without support in inferences that may be drawn from the\nfacts in the record”); Caro, 280 F.3d at 1253; Beech Aircraft\nCorp. v. United States, 51 F.3d 834, 838 (9th Cir. 1995) (per\ncuriam). The credited expert testimony established that GCS\nis medically necessary to alleviate Edmo’s gender dysphoria.\n\n b. Dr. Eliason’s Assessment\n\n Turning from the expert testimony offered, the State\ncontends that Edmo’s experts, at most, created a dispute of\nprofessional judgment with Edmo’s treating psychiatrist, Dr.\nEliason, who it urges reasonably concluded that GCS is\ninappropriate for Edmo. If that is the case, the argument\ngoes, then Edmo’s Eighth Amendment claim fails because\nthe dispute is merely a “difference of opinion . . . between\nmedical professionals” about “what medical care is\nappropriate.” Snow, 681 F.3d at 987. The problem for the\nState is that Dr. Eliason’s decision “was medically\nunacceptable under the circumstances.” Toguchi, 391 F.3d\nat 1058 (quoting Jackson, 90 F.3d at 332).\n\n In particular, as the district court found, Dr. Eliason did\nnot follow accepted standards of care in the area of\ntransgender health care. See Edmo, 358 F. Supp. 3d at 1126.\nDr. Eliason explained in his notes that, in his view, GCS is\nmedically necessary in three situations: “congenital\nmalformation or ambiguous genitalia,” “severe and\ndevastating dysphoria that is primarily due to genitals,” or\n“some type of medical problem in which endogenous sexual\n\f60 EDMO V. CORIZON\n\nhormones were causing severe physiological damage.” The\nconclusion of his notes—“[t]his inmate does not meet any of\nthose [three] criteria”—suggests that he views those as the\nonly three scenarios in which GCS would be medically\nnecessary, an impression he did not dispel during his\ntestimony. Those “criteria” (Dr. Eliason’s term), however,\nbear little resemblance to the widely accepted, evidence-\nbased criteria set out in the WPATH’s Standards of Care. As\nDr. Eliason acknowledged, the NCCHC endorses the\nWPATH Standards of Care as the accepted standards for the\ntreatment of transgender prisoners. And as the district court\nfound and the State does not contest, “[t]here are no other\ncompeting, evidence-based standards that are accepted by\nany nationally or internationally recognized medical\nprofessional groups.” Id. at 1125. Dr. Eliason did not follow\nthese standards in rendering his decision.\n\n The State challenges the district court’s finding that\nDr. Eliason “did not apply the WPATH Criteria,” id. at 1126,\non two grounds. First, citing Dr. Eliason’s testimony at the\nevidentiary hearing, it urges that Dr. Eliason concluded that\nGCS was not medically necessary for Edmo because Edmo’s\nmental health issues were not well controlled (the fourth\nWPATH criterion) and she had not consistently presented as\nfemale outside of prison (the sixth).\n\n The district court’s rejection of this post hoc explanation\nwas not clear error. Neither of the explanations offered by\nDr. Eliason during the evidentiary hearing appears in\nDr. Eliason’s notes. Nor did he give these reasons during his\ndeposition. Their absence is conspicuous, given that\nDr. Eliason took the time to indicate instances where, in his\nopinion, GCS is appropriate and to explain that Edmo did\nnot satisfy his “criteria.”\n\f EDMO V. CORIZON 61\n\n Second, the State highlights that Dr. Eliason’s notes\nrecommend further “supportive counseling” for Edmo and\nindicate that Edmo was up for parole. The State construes\nthese notes as shorthand for the fourth and sixth WPATH\ncriteria, respectively. The State’s proposed reading of\nDr. Eliason’s notes is unreasonable. His notes are clear that\nGCS is not needed because Edmo did not meet his three\n“criteria,” and the district court was well within its\nfactfinding discretion in rejecting the State’s strained\nreading. We therefore conclude that the district court\nreasonably found that Dr. Eliason “did not rely upon any\nfinding that Ms. Edmo did not meet the WPATH criteria in\nconcluding in his April 2016 assessment that she did not\nmeet the criteria for gender confirmation surgery.” Id.\nat 1120.\n\n Notably, neither Dr. Eliason nor the State has offered any\nexplanation or support for Dr. Eliason’s “criteria.”\nDr. Eliason testified that he could not recall where he came\nup with them.\n\n Nor has Dr. Eliason or the State contended that\nDr. Eliason’s criteria were a reasonable deviation or\nmodification of the WPATH Standards of Care. In any\nevent, we could not accept that argument. Dr. Eliason’s\ncriteria—apparently invented out of whole cloth—are so far\nafield from the WPATH standards that we cannot\ncharacterize his decision as a flexible application of or\ndeviation from those standards. Indeed, as Dr. Gorton\nexplained, two of Dr. Eliason’s criteria are inapplicable to\nthe care of transgender individuals. Dr. Eliason’s criterion\nof “congenital malformation or ambiguous genitalia” “isn’t\n. . . germane to transgender people.” His statement that GCS\ncould be needed when “endogenous sexual hormones were\ncausing severe physiological damage,” is, in Dr. Gorton’s\n\f62 EDMO V. CORIZON\n\nwords, “bizarre. I can’t think of a clinical circumstance\nwhere . . . your hormones that your body produces are\nattacking you . . . . I just don’t understand what [Dr. Eliason]\nis talking about there.”\n\n Dr. Eliason, in short, did not follow the accepted\nstandards of care in the area of transgender health care, nor\ndid he reasonably deviate from or flexibly apply them.\nDr. Eliason did not apply the established standards, even as\na starting point, in his evaluation.\n\n Putting to the side Dr. Eliason’s failure to follow or\nreasonably deviate from the accepted standards of care, his\ndecision was internally contradictory in an important way.\nHis notes reflect that GCS would be medically necessary if\na person is suffering “severe and devastating gender\ndysphoria that is primarily due to genitals.” At his\ndeposition, Dr. Eliason conceded that self-castration could\nshow gender dysphoria sufficiently severe to satisfy that\ncriterion. And at the evidentiary hearing, he acknowledged\nthat Edmo “does primarily meet that criteri[on].” Thus, even\nunder Dr. Eliason’s own criteria, Edmo should have been\nprovided GCS. Neither Dr. Eliason nor the State has\nreconciled this important contradiction between\nDr. Eliason’s criteria and his determination.\n\n In sum, Dr. Eliason’s evaluation was not an exercise of\nmedically acceptable professional judgment. Dr. Eliason’s\ndecision was based on inexplicable criteria far afield from\nthe recognized standards of care and, even applying\nDr. Eliason’s criteria, Edmo qualifies for GCS. Given the\ncredited expert testimony that GCS is necessary to treat\nEdmo’s gender dysphoria, Dr. Eliason’s contrary\n\f EDMO V. CORIZON 63\n\ndetermination was “medically unacceptable under the\ncircumstances.” 18 Snow, 681 F.3d at 988.\n\n 2. Deliberate Indifference\n\n The State next contends that even if the treatment\nprovided Edmo was medically unacceptable, no defendant\nacted “in conscious disregard of an excessive risk to\n[Edmo’s] health.” Hamby, 821 F.3d at 1092 (quoting Snow,\n681 F.3d at 988). We disagree.\n\n The record demonstrates that Dr. Eliason acted with\ndeliberate indifference to Edmo’s serious medical needs.\nDr. Eliason knew, as of the time of his evaluation, that Edmo\nhad attempted to castrate herself. He also knew that Edmo\nsuffers from gender dysphoria; he knew she experiences\n“clinically significant” distress that impairs her ability to\nfunction. He acknowledged that Edmo’s self-castration\nattempt was evidence that Edmo’s gender dysphoria, in his\nwords, “had risen to another level.” Dr. Eliason nonetheless\ncontinued with Edmo’s ineffective treatment plan.\n\n Edmo then tried to castrate herself a second time, in\nDecember 2016. Dr. Eliason knew of that nearly\n\n 18\n Dr. Eliason was not alone in his decision. Dr. Stoddart,\nDr. Young, and Jeremy Clark agreed with his assessment, as did the\nMTC. The State contends that such general agreement demonstrates that\nDr. Eliason’s decision was reasonable. But general agreement in a\nmedically unacceptable form of treatment does not somehow make it\nreasonable. This is especially so in light of the limited review those\nindividuals performed: Dr. Stoddard, Dr. Young, and Jeremy Clark\nagreed with Dr. Eliason’s recommended treatment as he presented it to\nthem and without personally evaluating Edmo, and the MTC “does not\nmake any individual treatment decisions regarding [gender dysphoric]\ninmates. Those determinations are made by the individual clinicians or\nthe medical staff employed by Corizon,” like Dr. Eliason.\n\f64 EDMO V. CORIZON\n\ncatastrophic event, but he did not reevaluate or recommend\na change to Edmo’s treatment plan, despite indicating in his\nApril 2016 evaluation that he would continue to monitor and\nassess Edmo’s condition. Dr. Eliason continued to see Edmo\nafter that time, and he considered Edmo’s treatment as a\nmember of the MTC. At no point did Dr. Eliason change his\nmind or the treatment plan regarding surgery. Under these\ncircumstances, we conclude that Dr. Eliason knew of and\ndisregarded the substantial risk of severe harm to Edmo.\nFarmer, 511 U.S. at 837.\n\n The State urges that neither Dr. Eliason nor any other\ndefendant acted with deliberate indifference because none\nacted with “malice, intent to inflict pain, or knowledge that\n[the] recommended course of treatment was medically\ninappropriate.” The State misstates the standard. A prisoner\n“must show that prison officials ‘kn[e]w [ ] of and\ndisregard[ed]’ the substantial risk of harm,’ but the officials\nneed not have intended any harm to befall the inmate; ‘it is\nenough that the official acted or failed to act despite his\nknowledge of a substantial risk of serious harm.’” Lemire v.\nCal. Dep’t of Corr. & Rehab., 726 F.3d 1062, 1074 (9th Cir.\n2013) (alterations in original) (quoting Farmer, 511 U.S.\nat 837, 842). Neither the Supreme Court nor this court has\never required a plaintiff to show a “sinister [prison official]\nwith improper motives,” as the State would require. It is\nenough that Dr. Eliason knew of and disregarded an\nexcessive risk to Edmo’s health by rejecting her request for\nGCS and then never re-evaluating his decision despite\nongoing harm to Edmo.\n\n The State also contends that because the defendants\nprovided some care to Edmo, no defendant could have been\ndeliberately indifferent. The provision of some medical\ntreatment, even extensive treatment over a period of years,\n\f EDMO V. CORIZON 65\n\ndoes not immunize officials from the Eighth Amendment’s\nrequirements. See Lopez v. Smith, 203 F.3d 1122, 1132 (9th\nCir. 2000) (en banc) (explaining that “[a] prisoner need not\nprove that he was completely denied medical care” to make\nout an Eighth Amendment claim); see also De’lonta,\n708 F.3d at 526 (“[J]ust because [officials] have provided\nDe’lonta with some treatment consistent with the GID\nStandards of Care, it does not follow that they have\nnecessarily provided her with constitutionally adequate\ntreatment.”). As the Fourth Circuit has aptly analogized,\n\n imagine that prison officials prescribe a\n painkiller to an inmate who has suffered a\n serious injury from a fall, but that the\n inmate’s symptoms, despite the medication,\n persist to the point that he now, by all\n objective measure, requires evaluation for\n surgery. Would prison officials then be free\n to deny him consideration for surgery,\n immunized from constitutional suit by the\n fact they were giving him a painkiller? We\n think not.\n\nDe’lonta, 708 F.3d at 526. Here, although the treatment\nprovided Edmo was important, it stopped short of what was\nmedically necessary.\n\n 3. Out-of-Circuit Precedent\n\n Our decision cleaves to settled Eighth Amendment\njurisprudence, which requires a fact-specific analysis of the\nrecord (as construed by the district court) in each case. See\nPatel v. Kent Sch. Dist., 648 F.3d 965, 975 (9th Cir. 2011)\n(“Deliberate-indifference cases are by their nature highly\nfact-specific . . . .”); see also Rachel v. Troutt, 820 F.3d 390,\n394 (10th Cir. 2016) (“Each step of this [deliberate\n\f66 EDMO V. CORIZON\n\nindifference] inquiry is fact-intensive.” (quoting Hartsfield\nv. Colburn, 491 F.3d 394, 397 (8th Cir. 2007))); Roe v.\nElyea, 631 F.3d 843, 859 (7th Cir. 2011) (“[I]nmate medical\ncare decisions must be fact-based with respect to the\nparticular inmate, the severity and stage of his condition, the\nlikelihood and imminence of further harm and the efficacy\nof available treatments.”); Youmans v. Gagnon, 626 F.3d\n557, 564 (11th Cir. 2010) (“Judicial decisions addressing\ndeliberate indifference to a serious medical need, like\ndecisions in the Fourth Amendment search-and-seizure\nrealm, are very fact specific.”); Chance v. Armstrong,\n143 F.3d 698, 703 (2d Cir. 1998) (“Whether a course of\ntreatment was the product of sound medical judgment,\nnegligence, or deliberate indifference depends on the facts\nof the case.”).\n\n Several years ago, the First Circuit, sitting en banc,\nemployed that fact-based approach to evaluate a gender\ndysphoric prisoner’s Eighth Amendment claim seeking\nGCS. The First Circuit confronted the following record:\ncredited expert testimony disagreed as to whether GCS was\nmedically necessary; the prisoner’s active treatment plan,\nwhich did not include GCS, had “led to a significant\nstabilization in her mental state”; and a report and testimony\nfrom correctional officials detailed significant security\nconcerns that would arise if the prisoner underwent GCS.\nKosilek, 774 F.3d at 86–96. “After carefully considering the\ncommunity standard of medical care, the adequacy of the\nprovided treatment, and the valid security concerns\narticulated by the DOC,” a 3–2 majority of the en banc court\nconcluded that the plaintiff had not demonstrated GCS was\nmedically necessary treatment for her gender dysphoria. Id.\nat 68.\n\f EDMO V. CORIZON 67\n\n Our approach mirrors the First Circuit’s, but the\nimportant factual differences between cases yield different\noutcomes. Notably, the security concerns in Kosilek, which\nthe First Circuit afforded “wide-ranging deference,” are\ncompletely absent here. Id. at 92. The State does not so\nmuch as allude to them. The medical evidence also differs.\nIn Kosilek, qualified and credited experts disagreed about\nwhether GCS was necessary. Id. at 90. As explained above,\nthe district court’s careful factual findings admit of no such\ndisagreement here. Rather, they unequivocally establish that\nGCS is the safe, effective, and medically necessary treatment\nfor Edmo’s severe gender dysphoria.\n\n We recognize, however, that our decision is in tension\nwith Gibson v. Collier. In that case, the Fifth Circuit held,\nin a split decision, that “[a] state does not inflict cruel and\nunusual punishment by declining to provide [GCS] to a\ntransgender inmate.” 920 F.3d at 215. It did so on a “sparse\nrecord”—which included only the WPATH Standards of\nCare and was notably devoid of “witness testimony or\nevidence from professionals in the field”—compiled by a\npro se plaintiff. Id. at 220. Despite the sparse record, a 2–1\nmajority of the Gibson panel concluded that “there is no\nconsensus in the medical community about the necessity and\nefficacy of [GCS] as a treatment for gender dysphoria. . . .\nThis on-going medical debate dooms Gibson’s claim.” Id.\nat 221.\n\n We respectfully disagree with the categorical nature of\nour sister circuit’s holding. Most fundamentally, Gibson\nrelies on an incorrect, or at best outdated, premise: that\n“[t]here is no medical consensus that [GCS] is a necessary\nor even effective treatment for gender dysphoria.” Id. at 223.\n\n As the record here demonstrates and the State does not\nseriously dispute, the medical consensus is that GCS is\n\f68 EDMO V. CORIZON\n\neffective and medically necessary in appropriate\ncircumstances. The WPATH Standards of Care—which are\nendorsed by the American Medical Association, the\nAmerican Medical Student Association, the American\nPsychiatric Association, the American Psychological\nAssociation, the American Family Practice Association, the\nEndocrine Society, the National Association of Social\nWorkers, the American Academy of Plastic Surgeons, the\nAmerican College of Surgeons, Health Professionals\nAdvancing LGBTQ Equality, the HIV Medicine\nAssociation, the Lesbian, Bisexual, Gay and Transgender\nPhysician Assistant Caucus, and Mental Health America—\nrecognize this fact. WPATH SOC at 54–55. Each expert in\nthis case agrees. As do others in the medical community.\nSee, e.g., U.S. Dep’t of Health & Human Servs., No. A-13-\n87, Decision No. 2576; Bao Ngoc N. Tran, et al., Gender\nAffirmation Surgery: A Synopsis Using American College of\nSurgeons National Surgery Quality Improvement Program\nand National Inpatient Sample Databases, 80 Annals Plastic\nSurgery S229, S234 (2018); Frey, A Historical Review of\nGender-Affirming Medicine, 14 J. Sexual Med. at 991; see\nalso What We Know Project, Ctr. for the Study of\nInequality, Cornell Univ., What Does the Scholarly\nResearch Say About the Effect of Gender Transition on\nTransgender Well-Being?, https://whatweknow.inequality.c\nornell.edu/topics/lgbt-equality/what-does-the-scholarly-rese\narch-say-about-the-well-being-of-transgender-people/ (last\nvisited July 10, 2019) (reviewing the available literature and\nfinding “a robust international consensus in the peer-\nreviewed literature that gender transition, including medical\ntreatments such as hormone therapy and surgeries, improves\nthe overall well-being of transgender individuals”). The\nFifth Circuit is the outlier.\n\f EDMO V. CORIZON 69\n\n Gibson’s broad holding stemmed from a dismaying\ndisregard for procedure. As noted, the “sparse” summary\njudgment record that the pro se plaintiff developed included\n“only the WPATH Standards of Care.” Gibson, 920 F.3d at\n221. Perhaps that factual deficiency doomed Gibson’s\nEighth Amendment claim. See id. at 223–24. But to reach\nits broader holding that denying GCS cannot, as a matter of\nlaw, violate the Eighth Amendment—in other words, to\nreject every conceivable Eighth Amendment claim based on\nthe denial of GCS—the Fifth Circuit coopted the record from\nKosilek, a First Circuit decision that predates Gibson by four\nyears. Id. at 221–23. We doubt the analytical value of such\nan anomalous procedural approach.\n\n Worse yet, the medical opinions from Kosilek do not\nsupport the Fifth Circuit’s categorical holding. Dr. Chester\nSchmidt’s and Dr. Stephen Levine’s testimony in Kosilek,\nwhich the Fifth Circuit relied on, do not support the\nproposition that GCS is never medically necessary. Dr.\nSchmidt and Dr. Levine testified that GCS was not necessary\nin the factual circumstances of that case, that is, based on the\nunique medical needs of the prisoner at issue. See Kosilek,\n774 F.3d at 76–79.\n\n The only suggestion in Kosilek that GCS is never\nmedically necessary is in the First Circuit’s recitation of the\ntestimony of Dr. Cynthia Osborne. See Gibson, 920 F.3d\nat 221. The First Circuit recounted that Dr. Osborne testified\nthat she “did not view [GCS] as medically necessary in light\nof the ‘whole continuum from noninvasive to invasive’\ntreatment options available to individuals with” gender\ndysphoria. Kosilek, 774 F.3d at 77. To the extent this vague\nportrait of Dr. Osborne’s testimony conveys her belief that\nGCS is never medically necessary, she has apparently\nchanged her view in the more than ten years since she\n\f70 EDMO V. CORIZON\n\ntestified in Kosilek. Like both sides and all four medical\nexperts who testified here, Dr. Osborne now agrees that GCS\n“can be medically necessary for some, though not all,\npersons with [gender dysphoria], including some prison\ninmates.” Osborne & Lawrence, Male Prison Inmates With\nGender Dysphoria, 45 Archives of Sexual Behav. at 1651.\nIn her and her co-author’s words, “[GCS] is a safe, effective,\nand widely accepted treatment for [gender dysphoria];\ndisputing the medical necessity of [GCS] based on assertions\nto the contrary is unsupportable.” Id. The predicate medical\nopinions that Gibson is premised upon, then, do not support\nthe Fifth Circuit’s view that GCS is never medically\nnecessary. The consensus is that GCS is effective and\nmedically necessary in appropriate circumstances. 19\n\n Gibson is unpersuasive for several additional reasons. It\ndirectly conflicts with decisions of this circuit, the Fourth\n\n 19\n We do not suggest that every member of the medical and mental\nhealth communities agrees that GCS may be medically necessary. There\nare outliers. But when the medical consensus is that a treatment is\neffective and medically necessary under the circumstances, prison\nofficials render unacceptable care by following the views of outliers\nwithout offering a credible medical basis for deviating from the accepted\nview. See Kosilek, 774 F.3d at 90 n.12 (explaining that it is not enough\nfor “correctional administrators wishing to avoid treatment . . . simply to\nfind a single practitioner willing to attest that some well-accepted\ntreatment is not necessary”); Hamilton v. Endell, 981 F.2d 1062, 1067\n(9th Cir. 1992) (“By choosing to rely upon a medical opinion which a\nreasonable person would likely determine to be inferior, the prison\nofficials took actions which may have amounted to the denial of medical\ntreatment, and the unnecessary and wanton infliction of pain.” (quotation\nomitted)), overruled in part on other grounds as recognized in Snow,\n681 F.3d at 986; cf. also Bragdon v. Abbott, 524 U.S. 624, 650 (1998)\n(“A health care professional who disagrees with the prevailing medical\nconsensus may refute it by citing a credible scientific basis for deviating\nfrom the accepted norm.”).\n\f EDMO V. CORIZON 71\n\nCircuit, and the Seventh Circuit, all of which have held that\ndenying surgical treatment for gender dysphoria can pose a\ncognizable Eighth Amendment claim. Rosati, 791 F.3d at\n1040 (alleged blanket ban on GCS and denial of GCS to\nplaintiff with severe symptoms, including repeated self-\ncastration attempts, states an Eighth Amendment claim);\nFields v. Smith, 653 F.3d 550, 552–53, 558–59 (7th Cir.\n2011) (law banning hormone treatment and GCS, even if\nmedically necessary, violates the Eighth Amendment);\nDe’lonta, 708 F.3d at 525 (alleged denial of an evaluation\nfor GCS states an Eighth Amendment claim). 20 Relatedly,\nGibson eschews Eighth Amendment precedent requiring a\ncase-by-case determination of the medical necessity of a\nparticular treatment. See, e.g., Colwell v. Bannister,\n763 F.3d 1060, 1068 (9th Cir. 2014) (holding that the\n“blanket, categorical denial of medically indicated surgery\nsolely on the basis of an administrative policy . . . is the\nparadigm of deliberate indifference” (quotation omitted));\nRoe, 631 F.3d at 859.\n\n In this latter respect, Gibson also contradicts and\nmisconstrues the precedent it purports to follow: Kosilek.\nAccording to the Gibson majority, “the majority in Kosilek\neffectively allowed a blanket ban on sex reassignment\nsurgery.” 920 F.3d at 216. Not so. The First Circuit did\nprecisely what we do here: assess whether the record before\nit demonstrated deliberate indifference to the plaintiff’s\n\n 20\n The Fifth Circuit unpersuasively attempted to reconcile its\ndecision with Rosati and De’lonta, pointing out that those decisions\n“allowed Eighth Amendment claims for [GCS] to survive motions to\ndismiss, without addressing the merits.” Gibson, 920 F.3d at 223 n.8.\nBut if Gibson is correct that failing to provide GCS cannot amount to\ndeliberate indifference, then a plaintiff cannot state an Eighth\nAmendment claim based on the denial of GCS. Rosati and De’lonta\nwould necessarily have been decided differently under Gibson’s holding.\n\f72 EDMO V. CORIZON\n\ngender dysphoria. On the record before it, the First Circuit\ndetermined that either of two courses of treatment (one\nincluded GCS and one did not) were medically acceptable.\nKosilek, 774 F.3d at 90. In light of those medically\nacceptable alternatives, the First Circuit explained that it was\nnot its place to “second guess medical judgments or to\nrequire that the DOC adopt the more compassionate of two\nadequate options.” Id. (quotation omitted). It expressly\ncautioned that the opinion should not be read to “create a de\nfacto ban against [GCS] as a medical treatment for any\nincarcerated individual,” as “any such policy would conflict\nwith the requirement that medical care be individualized\nbased on a particular prisoner’s serious medical needs.” Id.\nat 91 (citing Roe, 631 F.3d at 862–63). The Fifth Circuit\ndisregarded these words of warning. 21\n\n ***\n\n In summary, Edmo has established that she suffers from\na “serious medical need,” Jett, 439 F.3d at 1096, and that the\ntreatment provided was “medically unacceptable under the\ncircumstances” and chosen “in conscious disregard of an\nexcessive risk” to her health, Hamby, 821 F.3d at 1092. She\nestablished her Eighth Amendment claim of deliberate\nindifference as to Defendant-Appellant Dr. Eliason.\n\n\n\n 21\n Gibson’s final, originalist rationale—that it cannot be cruel and\nunusual to deny a surgery that has only once been provided to an inmate,\n920 F.3d at 226–28—warrants little discussion. Gibson’s originalist\nunderstanding of the Eighth Amendment does not control; Estelle does,\nand under Estelle a plaintiff establishes an Eighth Amendment claim by\ndemonstrating that prison officials were deliberately indifferent to a\nserious medical need. 429 U.S. at 106. This standard protects the\nevolving standards of decency enshrined in the Eighth Amendment.\n\f EDMO V. CORIZON 73\n\n B. Irreparable Harm\n\n The State next contends that the district court erred in\nfinding that Edmo would be irreparably harmed absent an\ninjunction.\n\n In reaching its conclusion, the district court found that\nEdmo experiences ongoing “clinically significant distress,”\nmeaning “the distress impairs or severely limits [her] ability\nto function in a meaningful way.” Edmo, 358 F. Supp. 3d\nat 1110–11. This finding is supported by Edmo’s testimony\nthat her gender dysphoria causes her to feel “depressed,”\n“disgusting,” “tormented,” and “hopeless”; that she actively\nexperiences thoughts of self-castration; and that she “self-\nmedicat[es]” by cutting her arms with a razor to avoid acting\non those thoughts and impulses. The district court also\nfound that in the absence of surgery, Edmo “will suffer\nserious psychological harm and will be at high risk of self-\ncastration and suicide.” Id. at 1128. This finding is\nsupported by the credited expert testimony of Dr. Ettner and\nDr. Gorton, who detailed the escalating risks of self-surgery,\nsuicide, and emotional decompensation should Edmo be\ndenied surgery.\n\n It is no leap to conclude that Edmo’s severe, ongoing\npsychological distress and the high risk of self-castration and\nsuicide she faces absent surgery constitute irreparable harm.\nSee Stanley v. Univ. of S. Cal., 13 F.3d 1313, 1324 n.5 (9th\nCir. 1994); Thomas v. County of Los Angeles, 978 F.2d 504,\n511 (9th Cir. 1992); Chalk v. U.S. Dist. Ct. Cent. Dist. of\nCal., 840 F.2d 701, 709 (9th Cir. 1988). Moreover, the\ndeprivation of Edmo’s constitutional right to adequate\nmedical care is sufficient to establish irreparable harm. See\nNelson v. NASA, 530 F.3d 865, 882 (9th Cir. 2008) (“Unlike\nmonetary injuries, constitutional violations cannot be\nadequately remedied through damages and therefore\n\f74 EDMO V. CORIZON\n\ngenerally constitute irreparable harm.”), rev’d and\nremanded on other grounds, 562 U.S. 134 (2011).\n\n The State offers three contentions as to why the district\ncourt erred in finding that Edmo would be irreparably\ninjured in the absence of an injunction. None is persuasive.\n\n First, the State argues that the “long delay” of “nearly a\nyear” between Edmo filing her Amended Complaint and her\npreliminary injunction motion “implies a lack of urgency\nand irreparable harm.” We disagree. The procedural history\ndemonstrates that Edmo did not sit on her rights. Proceeding\npro se, Edmo moved for preliminary injunctive relief when\nshe filed her original complaint. The court then appointed\ncounsel for Edmo, and shortly after appearing, appointed\ncounsel withdrew Edmo’s motion and filed an amended\ncomplaint. To assess the urgency of surgery, Edmo’s\ncounsel promptly sought access to Edmo’s medical records,\nwhich the State did not produce until more than six months\nlater. Edmo moved for injunctive relief shortly thereafter.\nDuring that time, Edmo and her counsel diligently\ninvestigated and compiled the necessary record to move for\ninjunctive relief. That it took them months to do their\ndiligence does not suggest that Edmo will not be harmed\nabsent an injunction.\n\n Second, the State contends that Edmo has not established\nirreparable injury because both she and her expert,\nDr. Gorton, agree that GCS is not an emergency surgery and\nthat the State should have six months to provide such\nsurgery. The State’s argument would preclude courts from\nordering non-emergent medical care, even if the Eighth\nAmendment demands it. That is untenable. The State also\nignores the rationale for the six-month time period. As Dr.\nGorton explained, all patients who receive GCS “are seen,\nthey are evaluated, there is a process you have to go\n\f EDMO V. CORIZON 75\n\nthrough.” In his experience, that process typically concludes\nwithin six months. That Edmo requested relief on a\nreasonable timeline, based on the medical evidence, does not\nundermine the strong evidence of irreparable injury.\n\n Third, the State contends that Edmo has not established\nirreparable harm because she “has not attempted suicide or\nself-castration for years.” That argument overlooks the\nprofound, persistent distress Edmo’s gender dysphoria\ncauses, as well as the credited expert testimony that absent\nGCS, Edmo is at risk of further attempts at self-castration,\nand possibly suicide. The district court did not err in finding\nthat Edmo would be irreparably harmed in the absence of an\ninjunction.\n\n IV. Challenges to the Scope of the Injunction\n\n We turn to the State’s contentions that the district court’s\ninjunction was overbroad.\n\n A. Individual Defendants\n\n The State contends that the injunction should not apply\nto Atencio, Zmuda, Yordy, Siegert, Dr. Young, Dr. Craig,\nDr. Eliason, or Dr. Whinnery because the district court did\nnot find that they, individually, were deliberately indifferent\nto Edmo’s medical needs.\n\n As explained in Section III.A, Edmo has established that\nDr. Eliason was deliberately indifferent to her serious\nmedical needs. The injunction was properly entered against\nhim because he personally participated in the deprivation of\nEdmo’s constitutional rights. See Colwell, 763 F.3d at 1070.\n\n Edmo sued Attencio, Zmuda, and Yordy in their official\ncapacities. An official-capacity suit for injunctive relief is\n\f76 EDMO V. CORIZON\n\nproperly brought against any persons who “would be\nresponsible for implementing any injunctive relief.” Pouncil\nv. Tilton, 704 F.3d 568, 576 (9th Cir. 2012). The State does\nnot contest that Attencio, as Director of IDOC, and Zmuda,\nas Deputy Director of IDOC, would be responsible for\nimplementing any injunctive relief ordered. Edmo properly\nnamed them as defendants to her Eighth Amendment claim\nfor injunctive relief, regardless of their personal\ninvolvement. See Colwell, 763 F.3d at 1070–71 (director of\na state correctional system is a proper defendant in an\nofficial-capacity suit seeking injunctive relief for Eighth\nAmendment violations). Yordy is no longer the Warden of\nISCI, but, by operation of the Federal Rules, his successor,\nAl Ramirez, is “automatically substituted as party” in his\nofficial capacity. Fed. R. Civ. P. 25(d). Ramirez is properly\na defendant to Edmo’s Eighth Amendment claim for\ninjunctive relief, regardless of his personal involvement. See\nColwell, 763 F.3d at 1070–71 (warden is a proper defendant\nin an official-capacity suit seeking injunctive relief for\nEighth Amendment violations). Because Edmo may\nproperly pursue her Eighth Amendment claim for injunctive\nrelief against Attencio, Zmuda, and Ramirez in their official\ncapacities, they are properly included within the scope of the\ndistrict court’s injunction. On remand, the district court shall\namend the injunction to substitute Al Ramirez (or the then-\ncurrent Warden of ISCI) as a party for Yordy.\n\n Edmo also named Yordy as a defendant in his individual\ncapacity. She likewise named Siegert, Dr. Young, Dr. Craig,\nand Dr. Whinnery as defendants in their individual\ncapacities (though she does not argue on appeal that the\ninjunction properly included them). We hold that the\nevidence in the current record is insufficient to conclude that\nthey were deliberately indifferent to Edmo’s serious medical\nneeds. In particular, the record does not show what they\n\f EDMO V. CORIZON 77\n\nknew about Edmo’s condition and what role they played in\nher treatment or lack thereof. Edmo has not established their\nliability, and the district court improperly included them\nwithin the scope of the injunction. We vacate the district\ncourt’s injunction to the extent it applies to Yordy, Siegert,\nDr. Young, Dr. Craig, and Dr. Whinnery in their individual\ncapacities. See California v. Azar, 911 F.3d 558, 585 (9th\nCir. 2018) (vacating in part an overbroad injunction and\nremanding to the district court). On remand, the district\ncourt shall modify the injunction to exclude those defendants\nfrom its scope.\n\n B. Corizon\n\n The State also contends that the injunction should not\napply to Corizon. It urges that Corizon does not have a\npolicy barring GCS and argues that such a policy is a\nprerequisite to liability under Monell v. Department of Social\nServices, 436 U.S. 658 (1978). We have not yet determined\nwhether Monell applies “to private entities acting on behalf\nof state governments,” such as Corizon. Oyenik v. Corizon\nHealth Inc., 696 F. App’x 792, 794 n.1 (9th Cir. 2017). We\nleave that issue for another day. Instead, we vacate the\ninjunction as to Corizon and remand with instructions to the\ndistrict court to modify the injunction to exclude Corizon.\nSee Azar, 911 F.3d at 585. Doing so still provides Edmo the\nrelief she seeks at this stage. 22\n\n\n\n\n 22\n For similar reasons, we need not reach Edmo’s contention and the\ndistrict court’s finding that “Corizon and IDOC have a de facto policy or\npractice of refusing” GCS to prisoners. Edmo, 358 F. Supp. 3d at 1127.\n\f78 EDMO V. CORIZON\n\n C. Relief Ordered\n\n The State next contends that the injunctive relief ordered\nis overbroad because it requires the State to provide Edmo\nall “adequate medical care.” The State misconstrues the\ndistrict court’s order. The order, read in context, requires\ndefendants to provide GCS, as well as “adequate medical\ncare” that is “reasonably necessary” to accomplish that\nend—not every conceivable form of adequate medical care.\nEdmo, 358 F. Supp. 3d at 1129; see also id. at 1109\n(“Plaintiff Adree Edmo alleges that prison authorities\nviolated her Eighth Amendment rights by refusing to provide\nher with gender confirmation surgery. For the reasons\nexplained below, the Court agrees and will order defendants\nto provide her with this procedure, a surgery which is\nconsidered medically necessary under generally accepted\nstandards of care.”); id. at 1110 (“[F]or the reasons explained\nin detail below, IDOC and Corizon will be ordered to\nprovide Ms. Edmo with gender confirmation surgery.”).\n\n The State similarly contends that the injunctive relief\nordered is overbroad because it requires the State to provide\nEdmo surgery even though the defendants are not surgeons\nand no surgeon has evaluated Edmo. We reject this obtuse\nreading of the district court’s order. The district court\nordered the State to “take all actions reasonably necessary to\nprovide Ms. Edmo gender confirmation surgery.” Edmo,\n358 F. Supp. 3d at 1129. That means that the State must take\nsteps within its power to provide GCS to Edmo, such as\nfinding a surgeon and scheduling a surgical evaluation.\nIndeed, we modified our stay of the district court’s order to\npermit a surgical consultation, which went forward in April\n2019. Oral Arg. at 12:00–12:10. The State cannot\nreasonably understand the district court’s December 13,\n2018 order to require that the defendants themselves provide\n\f EDMO V. CORIZON 79\n\nsurgery. To the extent there are issues arising from a surgical\nevaluation, the State can raise those issues with the district\ncourt. 23\n\n V. Challenges to the Procedure Used by the District\n Court\n\n Finally, the State contends that the district court\nimproperly converted an evidentiary hearing on a\npreliminary injunction into a final trial on the merits of\nEdmo’s Eighth Amendment claim for GCS without giving\nthem adequate notice and in violation of their Seventh\nAmendment right to a jury trial. We address and reject each\ncontention.\n\n A. Notice\n\n We first address the State’s contention that the district\ncourt erroneously converted the evidentiary hearing into a\nfinal trial on the merits without giving the State “clear and\nunambiguous notice.” Under Federal Rule of Civil\nProcedure 65(a)(2), “[a] district court may consolidate a\npreliminary injunction hearing with a trial on the merits, but\nonly when it provides the parties with clear and\nunambiguous notice [of the intended consolidation] either\nbefore the hearing commences or at a time which will afford\n\n 23\n The State contends for the first time in its reply brief that the\ninjunctive relief ordered was inappropriate because the WPATH\nStandards of Care require two referrals from qualified mental health\nprofessionals who have independently assessed the patient before GCS\nmay be provided. It similarly contends for the first time in its reply in\nsupport of its motion to dismiss that the order is overbroad because it\ndoes not specify the type of GCS ordered. Because the State did not\npresent these arguments in its opening brief, we do not consider them.\nSee Smith v. Marsh, 194 F.3d 1045, 1052 (9th Cir. 1999).\n\f80 EDMO V. CORIZON\n\nthe parties a full opportunity to present their respective\ncases.” Isaacson v. Horne, 716 F.3d 1213, 1220 (9th Cir.\n2013) (second alteration in original) (quotation omitted).\n“What constitutes adequate notice depends upon the facts of\nthe case.” Michenfelder v. Sumner, 860 F.2d 328, 337 (9th\nCir. 1988).\n\n A party challenging consolidation must show not only\ninadequate notice, but also “substantial prejudice in the\nsense that [it] was not allowed to present material evidence.”\nMichenfelder, 860 F.2d at 337; see also 11A Charles Alan\nWright et al., Federal Practice and Procedure § 2950 (3d\ned. Apr. 2019 update). “We have on occasion upheld a\ndistrict court’s failure to give any notice whatsoever before\nfinally determining the merits after only a preliminary\ninjunction hearing, where the complaining party has failed\nto show how additional evidence could have altered the\noutcome.” Michenfelder, 860 F.2d at 337.\n\n At the outset, we note that the State was provided notice,\ntwice, that the district court considered the evidentiary\nhearing a final trial on the merits of Edmo’s request for GCS.\nAt the beginning of the hearing, the district court explained\n“it’s hard for me to envision this hearing being anything but\na hearing on a final injunction at least as to that part of the\nrelief requested [GCS],” and it asked the parties to address\nby the end of the hearing whether it was for a permanent\ninjunction. At the close of the hearing, the district court\nagain questioned whether it could order GCS in a\npreliminary injunction. It explained that it had, in effect,\n“kind of treated this hearing as the final hearing” on Edmo’s\nrequest for GCS, and it again asked the parties to address in\ntheir oral closings or written briefs whether the hearing was\none for a permanent injunction. The State never answered\nthe court’s question or objected to consolidation, despite the\n\f EDMO V. CORIZON 81\n\ndistrict court specifically noting it had treated the hearing as\nfinal. Cf. Reilly v. United States, 863 F.2d 149, 160 (1st Cir.\n1988) (“[W]hen a trial judge announces a proposed course\nof action which litigants believe to be erroneous, the parties\ndetrimentally affected must act expeditiously to call the error\nto the judge’s attention or to cure the defect, not lurk in the\nbushes waiting to ask for another trial when their litigatory\nmilk curdles.”). This is not a case where the district court\ngave no notice whatsoever.\n\n Regardless, the State has not shown any prejudice. With\nfull awareness of the stakes, the district court permitted the\nparties four months of discovery and held a three-day\nevidentiary hearing. The parties called seven witnesses,\nsubmitted declarations in lieu of live testimony for other\nwitnesses, and submitted thousands of pages of exhibits and\nextensive pre- and post-trial briefing. Most importantly,\nboth parties put on extensive evidence concerning the\ntreatment provided to and withheld from Edmo and why it\nwas or was not appropriate—the key issue at the hearing.\n\n When it comes to identifying prejudice, the State is\ntellingly short on specifics. It indicates that it “would have\nobjected” to consolidation, but it failed to do so despite\nrepeated invitations—indeed, directives—to address the\nissue. The State also urges that it would have requested that\nthe named defendants be able to testify live, but it\nstipulated—knowing full well the stakes of the hearing—to\nsubmit certain testimony via declaration “[i]n lieu of and/or\nin addition to live testimony.” Moreover, the State fails to\nidentify what testimony those witnesses would have offered\nor explain how presenting that testimony live, instead of via\ndeclaration, “could have altered the outcome.”\nMichenfelder, 860 F.2d at 337. The district court did not\n\f82 EDMO V. CORIZON\n\ncommit reversible error in consolidating the evidentiary\nhearing with a trial on the merits of Edmo’s request for GCS.\n\n B. Seventh Amendment\n\n We turn to the State’s related contention that the district\ncourt violated the defendants’ Seventh Amendment right to\na jury trial by converting the evidentiary hearing into a trial\non the merits. We review that contention de novo. Palmer\nv. Valdez, 560 F.3d 965, 968 (9th Cir. 2009).\n\n The Seventh Amendment guarantees the right to a trial\nby jury “[i]n Suits at common law, where the value in\ncontroversy shall exceed twenty dollars.” U.S. Const.\namend. VII. In a case such as this, where legal claims are\njoined with equitable claims, a party “has a right to jury\nconsideration of all legal claims, as well as all issues\ncommon to both claims.” Plummer v. W. Int’l Hotels Co.,\n656 F.2d 502, 504 n.6 (9th Cir. 1981) (citing Curtis v.\nLoether, 415 U.S. 189, 196 n.11 (1974)). “Otherwise, the\ncourt might limit the parties’ opportunity to try to a jury\nevery issue underlying the legal claims by affording\npreclusive effect to its own findings of fact on questions that\nare common to both the legal and equitable claims.” Lacy v.\nCook County, 897 F.3d 847, 858 (7th Cir. 2018).\n\n Like other constitutional rights, the right to a jury trial in\ncivil suits can be waived. See United States v. Moore,\n340 U.S. 616, 621 (1951). It is well established that “[a]\nfailure to object to a proceeding in which the court sits as the\nfinder of fact waives a valid jury demand as to any claims\ndecided in that proceeding, at least where it was clear that\nthe court intended to make fact determinations.” Fillmore v.\nPage, 358 F.3d 496, 503 (7th Cir. 2004) (quotation omitted);\nsee also 9 Wright & Miller, Federal Practice and Procedure\n§ 2321 (“The right to jury trial also may be waived as it has\n\f EDMO V. CORIZON 83\n\nin many, many cases, by conduct, such as failing to object to\nor actually participating in a bench trial . . . .”).\n\n For example, in White v. McGinnis, we held that “[a]\nparty’s vigorous participation in a bench trial, without so\nmuch as a mention of a jury, . . . can only be ascribed to\nknowledgeable relinquishment of the prior jury demand.”\n903 F.2d 699, 703 (9th Cir. 1990) (en banc). We explained\nthat where a party chooses “to argue his case fully before the\ndistrict judge[,] it is not unjust to hold him to that\ncommitment.” Id. By contrast, we have held that “[w]hen a\nparty participates in [a] bench trial ordered by the trial court\nwhile continuing to demand a jury trial, his ‘continuing\nobjection’ is ‘sufficient to preserve his right to appeal the\ndenial of his request for a jury.’” Solis v. County of Los\nAngeles, 514 F.3d 946, 957 (9th Cir. 2008) (quoting United\nStates v. Nordbrock, 941 F.2d 947, 950 (9th Cir. 1991)).\n“This is because the party in such a case is not seeking ‘two\nbites at the procedural apple’ . . . . Rather, when a trial court\ndenies a party a jury trial despite the party’s continuing\ndemand, the party has little choice but to accede to the trial\ncourt’s ruling and participate in the bench trial.” Id. (citation\nomitted); see also Lovelace v. Dall, 820 F.2d 223, 228 (7th\nCir. 1987) (“Another policy justifying the jury demand\nwaiver rule is the view that it is unfair to permit a party to\nhave a trial, discover that it has lost, and then raise the jury\nissue because it is unsatisfied with the result of the trial.”).\n\n The State seeks a second bite at the apple. It vigorously\nparticipated in the evidentiary hearing without ever raising\nthe right to a jury trial. The State remained silent in the face\nof statements from the district court that it was considering\ntreating, and then that it had treated, the hearing as a final\ntrial on the merits, which made it clear that the court\n“intended to make fact determinations.” Fillmore, 358 F.3d\n\f84 EDMO V. CORIZON\n\nat 503. It also remained silent despite the district court\nasking twice whether the hearing was one for a permanent\ninjunction—as clear a time as any to raise any concerns\nabout a jury trial.\n\n The State raised the issue of a jury trial for the first time\non appeal, after the district court ruled against it. Even after\nthe district court’s ruling, the State made no objection or\nclaim to a jury trial. This conduct waived the State’s right to\na jury trial with respect to issues common to Edmo’s request\nfor an injunction ordering GCS and her legal claims.\n\n VI. Conclusion\n\n We apply the dictates of the Eighth Amendment today in\nan area of increased social awareness: transgender health\ncare. We are not the first to speak on the subject, nor will\nwe be the last. Our court and others have been considering\nEighth Amendment claims brought by transgender prisoners\nfor decades. During that time, the medical community’s\nunderstanding of what treatments are safe and medically\nnecessary to treat gender dysphoria has changed as more\ninformation becomes available, research is undertaken, and\nexperience is gained. The Eighth-Amendment inquiry takes\naccount of that developing understanding. See Estelle,\n429 U.S. at 102–03.\n\n We hold that where, as here, the record shows that the\nmedically necessary treatment for a prisoner’s gender\ndysphoria is gender confirmation surgery, and responsible\nprison officials deny such treatment with full awareness of\nthe prisoner’s suffering, those officials violate the Eighth\nAmendment’s prohibition on cruel and unusual punishment.\n\f EDMO V. CORIZON 85\n\n ***\n\n We affirm the district court’s entry of an injunction for\nEdmo. However, we vacate the injunction to the extent it\napplies to Corizon, Yordy, Siegert, Dr. Young, Dr. Craig,\nand Dr. Whinnery, in their individual capacities, and remand\nto the district court to modify the injunction accordingly.\nThe district court shall also modify the injunction to\nsubstitute Al Ramirez in his official capacity as Warden of\nISCI for Yordy.\n\n Although we addressed this appeal on an expedited\nbasis, it has been more than a year since doctors concluded\nthat GCS is medically necessary for Edmo. We urge the\nState to move forward. We emphatically do not speak to\nother cases, but the facts of this case call for expeditious\neffectuation of the injunction.\n\n In light of the nature and urgency of the relief at issue,\nwe will disfavor any motion, absent extraordinary\ncircumstances or consent from all parties, to extend the\nperiod to petition for rehearing or rehearing en banc. Our\nstay of the district court’s December 13, 2018 order shall\nautomatically terminate upon issuance of the mandate.\n\n Costs on appeal are awarded to Edmo.\n\n AFFIRMED IN PART, VACATED IN PART, AND\nREMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: Adree Edmo v. Corizon, Inc.\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-23\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n ADREE EDMO, AKA Mason Edmo, No. 19-35017\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 1:17-cv-00151-\\n BLW\\n CORIZON, INC.; SCOTT ELIASON;\\n MURRAY YOUNG; CATHERINE\\n WHINNERY,\\n Defendants-Appellants,\\n\\n and\\n\\n IDAHO DEPARTMENT OF\\n CORRECTIONS; AL RAMIREZ, in his\\n official capacity as warden of Idaho\\n State Correctional Institution; *\\n HENRY ATENCIO; JEFF ZMUDA;\\n HOWARD KEITH YORDY; RICHARD\\n CRAIG; RONA SIEGERT,\\n Defendants.\\n\\n\\n ADREE EDMO, AKA Mason Edmo, No. 19-35019\\n Plaintiff-Appellee,\\n\\n\\n\\n\\n *\\n Al Ramirez is substituted in his official capacity for his\\npredecessor, Howard Keith Yordy, pursuant to Rule 43(c)(2) of the\\nFederal Rules of Appellate Procedure.\\n\\f2 EDMO V. CORIZON\\n\\n\\n v. D.C. No.\\n 1:17-cv-00151-\\n IDAHO DEPARTMENT OF BLW\\n CORRECTIONS; AL RAMIREZ, in his\\n official capacity as warden of Idaho\\n State Correctional Institution; HENRY OPINION\\n ATENCIO; JEFF ZMUDA; HOWARD\\n KEITH YORDY; RICHARD CRAIG;\\n RONA SIEGERT,\\n Defendants-Appellants,\\n\\n and\\n\\n CORIZON, INC.; SCOTT ELIASON;\\n MURRAY YOUNG; CATHERINE\\n WHINNERY,\\n Defendants.\\n\\n Appeal from the United States District Court\\n for the District of Idaho\\n B. Lynn Winmill, Chief District Judge, Presiding\\n\\n Argued and Submitted May 16, 2019\\n San Francisco, California\\n\\n Filed August 23, 2019\\n\\n Before: M. Margaret McKeown and Ronald M. Gould,\\n Circuit Judges, and Robert S. Lasnik, ** District Judge.\\n\\n Per Curiam Opinion\\n\\n **\\n The Honorable Robert S. Lasnik, United States District Judge for\\nthe Western District of Washington, sitting by designation.\\n\\f EDMO V. CORIZON 3\\n\\n SUMMARY ***\\n\\n\\n Eighth Amendment / Prisoner Rights\\n\\n The panel affirmed the district court’s entry of a\\npermanent injunction in favor of Idaho state prisoner Adree\\nEdmo, but vacated the injunction to the extent it applied to\\ndefendants Corizon, Howard Yordy, Rona Siegert, Dr.\\nYoung, Dr. Craig, and Dr. Whinnery, in their individual\\ncapacities, in Edmo’s action seeking medical treatment for\\ngender dysphoria.\\n\\n The district court concluded that Edmo had established\\nher Eighth Amendment claim. The district court further\\nconcluded that gender confirmation surgery (“GCS”) was\\nmedically necessary for Edmo, and ordered the State to\\nprovide the surgery.\\n\\n The panel credited the district court’s factual findings as\\nlogical and well-supported, and held that the responsible\\nprison authorities were deliberately indifferent to Edmo’s\\ngender dysphoria, in violation of the Eighth Amendment.\\nThe panel held that the record, as construed by the district\\ncourt, established that Edmo had a serious medical need, that\\nthe appropriate medical treatment was GCS, and that prison\\nauthorities had not provided that treatment despite full\\nknowledge of Edmo’s ongoing and extreme suffering and\\nmedical needs. The panel rejected the State’s position that\\nthere was a reasoned disagreement between qualified\\nmedical professionals. The panel emphasized that its\\n\\n\\n ***\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\f4 EDMO V. CORIZON\\n\\nanalysis was individual to Edmo, and rested on the record of\\nthis case.\\n\\n Addressing further aspects of the appeal, the panel\\nrejected the State’s contention that the district court did not\\nmake the Prison Litigation Reform Act’s requisite “need-\\nnarrowness-intrusiveness” findings, causing the injunction\\nto automatically expire and mooting the appeal. The panel\\nheld that the district court’s order, considered as a whole,\\nmade all the findings required by 18 U.S.C. § 3626(a)(1)(A),\\nand Ninth Circuit precedent. The panel also held that the\\npermanent injunction that the district court entered had not\\nexpired, and remained in place, albeit stayed. The panel\\naccordingly denied the State’s motion to dismiss.\\n\\n The panel held that the district court did not err in\\ngranting a permanent injunction. Specifically, the panel\\nheld, based on the district court’s factual findings, that Edmo\\nestablished her Eighth Amendment claim and that she will\\nsuffer irreparable harm – in the form of ongoing mental\\nanguish and possible physical harm – if GCS is not provided.\\nThe State did not dispute that Edmo’s gender dysphoria was\\na sufficiently serious medical need to trigger the State’s\\nobligations under the Eighth Amendment. The panel held\\nthat the district court did not err in crediting the testimony of\\nEdmo’s experts that GCS was medically necessary to treat\\nEdmo’s gender dysphoria and that the State’s failure to\\nprovide that treatment was medically unacceptable. The\\npanel further held that the district court did not err in\\ndiscrediting the State’s experts because aspects of their\\nopinions were illogical and unpersuasive. Also, the panel\\nheld that the record demonstrated that Dr. Eliason acted with\\ndeliberate indifference to Edmo’s serious medical needs.\\nThe panel noted that its decision was in tension with the Fifth\\nCircuit’s decision in Gibson v. Collier, 920 F.3d 212 (5th\\n\\f EDMO V. CORIZON 5\\n\\nCir. 2019), and the panel rejected that decision’s categorical\\nholding that denying GCS cannot, as a matter of law, violate\\nthe Eighth Amendment.\\n\\n The panel held that the district court did not err in finding\\nthat Edmo would be irreparably harmed absent an\\ninjunction. The panel rejected the State’s contentions as to\\nwhy the district court erred in this finding.\\n\\n The panel next considered the State’s challenges to the\\nscope of the injunction. The panel held that the injunction\\nwas properly entered against Dr. Eliason because he\\npersonally participated in the deprivation of Edmo’s\\nconstitutional rights. The panel also held that because Edmo\\nmay properly pursue her Eighth Amendment claim for\\ninjunctive relief against Attencio, Zmuda and Ramirez in\\ntheir official capacities, they were properly included within\\nthe scope of the district court’s injunction. On remand, the\\ndistrict court shall amend the injunction to substitute the\\ncurrent warden as a party for Yordy. The panel vacated the\\ndistrict court’s injunction to the extent it applied to Yordy,\\nSiegert, Dr. Young, Dr. Craig, and Dr. Whinnery in their\\nindividual capacities because the evidence in the record was\\ninsufficient to conclude that they were deliberately\\nindifferent to Edmo’s serious medical needs. The panel\\nvacated the injunction as to Corizon, and remanded with\\ninstructions to the district court to modify the injunction to\\nexclude Corizon. Finally, the panel held that the injunctive\\nrelief ordered was not overbroad.\\n\\n The panel considered the State’s challenges to the\\nprocedure used by the district court. The panel rejected the\\nState’s contention that the district court erroneously\\nconverted the evidentiary hearing into a final trial on the\\nmerits without giving proper notice. The panel held that the\\nState did receive notice, and in any event, the State had not\\n\\f6 EDMO V. CORIZON\\n\\nshown any prejudice. The panel also rejected the State’s\\ncontention that the district court violated defendants’\\nSeventh Amendment right to a jury trial by converting the\\nevidentiary hearing into a trial on the merits. The panel held\\nthat the State’s conduct waived its right to a jury trial with\\nrespect to issues common to Edmo’s request for an\\ninjunction ordering GCS and her legal claims.\\n\\n\\n COUNSEL\\n\\nBrady J. Hall (argued), Special Deputy Attorney General;\\nLawrence G. Wasden, Attorney General; Office of the\\nAttorney General, Boise, Idaho; Marisa S. Crecelius, Moore\\nElia Kraft & Hall LLP, Boise, Idaho; for Defendants-\\nAppellants Idaho Department of Corrections, Henry\\nAtencio, Jeff Zmuda, Howard Keith Yordy, Richard Craig,\\nand Rona Siegert.\\n\\nDylan A. Eaton (argued), J. Kevin West, and Bryce Jensen,\\nParsons Behle & Latimer, Boise, Idaho, for Defendants-\\nAppellants Corizon, Inc.; Scott Eliason; Murray Young; and\\nCatherine Whinnery.\\n\\nLori Rifkin (argued), Hadsell Stormer & Renick LLP,\\nEmeryville, California; Dan Stormer and Shaleen Shanbhag,\\nHadsell Stormer & Renick LLP, Pasadena, California; Craig\\nDurham and Deborah Ferguson, Ferguson Durham PLLC,\\nBoise, Idaho; Amy Whelan and Julie Wilensky, National\\nCenter for Lesbian Rights, San Francisco, California; for\\nPlaintiff-Appellee.\\n\\nDavid M. Shapiro, Sheila A. Bedi, and Vanessa del Valle,\\nRoderick & Solange MacArthur Justice Center, Chicago,\\nIllinois; Molly E. Whitman, Akin Gump Strauss Hauer &\\n\\f EDMO V. CORIZON 7\\n\\nFeld LLP, Dallas, Texas; for Amici Curiae Andrea\\nArmstrong, Sharon Dolovich, Betsy Ginsberg, Michael B.\\nMushlin, Alexander A. Reinert, Laura Rovner, and Margo\\nSchlanger.\\n\\nMolly Kafka and Richard Alan Eppink, ACLU of Idaho\\nFoundation, Boise, Idaho; Devon A. Little and Derek\\nBorchardt, Walden Macht & Haran LLP, New York, New\\nYork; Amy Fettig and Jennifer Wedekind, ACLU National\\nPrison Project, Washington, D.C.; Gabriel Arkles and Rose\\nSaxe, ACLU LGBT & HIV Project/ACLU Foundation, New\\nYork, New York; for Amici Curiae Former Corrections\\nOfficials.\\n\\nDevi M. Rao and Jason T. Perkins, Jenner & Block LLP,\\nWashington, D.C., for Amici Curiae Medical and Mental\\nHealth Professional Organizations.\\n\\nSharif E. Jacob, Ryan K. M. Wong, Kristin E. Hucek, and\\nPatrick E. Murray, Keker Van Nest & Peters LLP, San\\nFrancisco, California, for Amicus Curiae Jody L. Herman.\\n\\nAlan E. Schoenfeld, Wilmer Cutler Pickering Hale and Dorr\\nLLP, New York, New York; Michael Posada, Wilmer Cutler\\nPickering Hale and Dorr LLP, Washington, D.C.; Richard\\nSaenz, Lambda Legal Defense & Education Fund Inc., New\\nYork, New York; A. Chinyere Ezie, Center for\\nConstitutional Rights, New York, New York; for Amici\\nCuriae Civil Rights & Non-Profit Organizations.\\n\\f8 EDMO V. CORIZON\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n The Eighth Amendment prohibits “cruel and unusual\\npunishments.” U.S. Const. amend. VIII. “The Amendment\\nembodies broad and idealistic concepts of dignity, civilized\\nstandards, humanity, and decency . . . .” Estelle v. Gamble,\\n429 U.S. 97, 102 (1976) (quotation omitted). Our society\\nrecognizes that prisoners “retain the essence of human\\ndignity inherent in all persons.” Brown v. Plata, 563 U.S.\\n493, 510 (2011).\\n\\n Consistent with the values embodied by the Eighth\\nAmendment, for more than 40 years the Supreme Court has\\nheld that “deliberate indifference to serious medical needs”\\nof prisoners constitutes cruel and unusual punishment.\\nEstelle, 429 U.S. at 106. When prison authorities do not\\nabide by their Eighth Amendment duty, “the courts have a\\nresponsibility to remedy the resulting . . . violation.” Brown,\\n563 U.S. at 511. We do so here.\\n\\n Adree Edmo (formerly Mason Dean Edmo) is a male-to-\\nfemale transgender prisoner in the custody of the Idaho\\nDepartment of Correction (“IDOC”). Edmo’s sex assigned\\nat birth (male) differs from her gender identity (female). The\\nincongruity causes Edmo to experience persistent distress so\\nsevere it limits her ability to function. She has twice\\nattempted self-castration to remove her male genitalia,\\nwhich cause her profound anguish.\\n\\n Both sides and their medical experts agree: Edmo suffers\\nfrom gender dysphoria, a serious medical condition. They\\nalso agree that the appropriate benchmark regarding\\ntreatment for gender dysphoria is the World Professional\\nAssociation of Transgender Health Standards of Care for the\\n\\f EDMO V. CORIZON 9\\n\\nHealth of Transsexual, Transgender, and Gender\\nNonconforming People (“WPATH Standards of Care”).\\nAnd the State 1 does not seriously dispute that in certain\\ncircumstances, gender confirmation surgery (“GCS”) can be\\na medically necessary treatment for gender dysphoria. The\\nparties’ dispute centers around whether GCS is medically\\nnecessary for Edmo—a question we analyze with deference\\nto the district court’s factual findings.\\n\\n Following four months of intensive discovery and a\\nthree-day evidentiary hearing, the district court concluded\\nthat GCS is medically necessary for Edmo and ordered the\\nState to provide the surgery. Its ruling hinged on findings\\nindividual to Edmo’s medical condition. The ruling also\\nrested on the finding that Edmo’s medical experts testified\\npersuasively that GCS was medically necessary, whereas\\ntestimony from the State’s medical experts deserved little\\nweight. In contrast to Edmo’s experts, the State’s witnesses\\nlacked relevant experience, could not explain their\\ndeviations from generally accepted guidelines, and testified\\nillogically and inconsistently in important ways.\\n\\n The district court’s detailed factual findings were amply\\nsupported by its careful review of the extensive evidence and\\ntestimony. Indeed, they are essentially unchallenged. The\\nappeal boils down to a disagreement about the implications\\nof the factual findings.\\n\\n Crediting, as we must, the district court’s logical, well-\\nsupported factual findings, we hold that the responsible\\n 1\\n In addition to IDOC, Edmo sued Corizon, Inc. (a private for-profit\\ncorporation that provides health care to inmates in IDOC custody) and\\nvarious employees of IDOC and Corizon. The defendants briefed the\\ncase jointly, and for ease of reference we refer to them collectively as\\n“the State.”\\n\\f10 EDMO V. CORIZON\\n\\nprison authorities have been deliberately indifferent to\\nEdmo’s gender dysphoria, in violation of the Eighth\\nAmendment. The record before us, as construed by the\\ndistrict court, establishes that Edmo has a serious medical\\nneed, that the appropriate medical treatment is GCS, and that\\nprison authorities have not provided that treatment despite\\nfull knowledge of Edmo’s ongoing and extreme suffering\\nand medical needs. In so holding, we reject the State’s\\nportrait of a reasoned disagreement between qualified\\nmedical professionals. We also emphasize that the analysis\\nhere is individual to Edmo and rests on the record in this\\ncase. We do not endeavor to project whether individuals in\\nother cases will meet the threshold to establish an Eighth\\nAmendment violation. The district court’s order entering\\ninjunctive relief for Edmo is affirmed, with minor\\nmodifications noted below.\\n\\n Our opinion proceeds as follows. In Part I, we provide\\nbackground on gender dysphoria, the standard of care, and\\nthe evidence considered and factual findings made by the\\ndistrict court. Part II explains why this appeal complies with\\nthe Prison Litigation Reform Act (“PLRA”) and is not moot.\\nIn Part III, we turn to the gravamen of the appeal: Edmo’s\\nEighth Amendment claim and showing of irreparable injury.\\nPart IV addresses the State’s challenges to the injunction’s\\nscope and narrows the injunction as to certain defendants.\\nPart V rejects the State’s objections to the procedure\\nemployed by the district court. We conclude in Part VI.\\n\\f EDMO V. CORIZON 11\\n\\n I. Background 2\\n\\n A. Gender Dysphoria and its Treatment\\n\\n Transgender individuals have a “[g]ender identity”—a\\n“deeply felt, inherent sense” of their gender—that does not\\nalign with their sex assigned at birth. 3 Am. Psychol. Ass’n,\\nGuidelines for Psychological Practice with Transgender\\nand Gender Nonconforming People, 70 Am. Psychologist\\n832, 834 (2015). Recent estimates suggest that\\napproximately 1.4 million transgender adults live in the\\nUnited States, or 0.6 percent of the adult population.\\nAndrew R. Flores et al., The Williams Inst., How Many\\nAdults Identify as Transgender in the United States?, at\\n2 (2016), http://williamsinstitute.law.ucla.edu/wp-content/\\nuploads/How-Many-Adults-Identify-as-Transgender-in-the-\\nUnited-States.pdf.\\n\\n Gender dysphoria 4 is “[d]istress that is caused by a\\ndiscrepancy between a person’s gender identity and that\\nperson’s sex assigned at birth (and the associated gender role\\nand/or primary and secondary sex characteristics).” World\\nProf’l Ass’n for Transgender Health, Standards of Care for\\nthe Health of Transsexual, Transgender, and Gender-\\n\\n 2\\n The following sections are derived from the district court’s factual\\nfindings and the record on appeal.\\n 3\\n At birth, infants are classified as male or female based on visual\\nobservation of their external genitalia. This is a person’s “sex assigned\\nat birth,” but it may not be the person’s gender identity.\\n 4\\n Until recently, the medical community commonly referred to\\ngender dysphoria as “gender identity disorder.” See Kosilek v. Spencer,\\n774 F.3d 63, 68 n.1 (1st Cir. 2014).\\n\\f12 EDMO V. CORIZON\\n\\nNonconforming People 2 (7th ed. 2011) (hereinafter\\n“WPATH SOC”). The Fifth Edition of the American\\nPsychiatric Association’s Diagnostic and Statistical Manual\\nof Mental Disorders (“DSM-5”) sets forth two conditions\\nthat must be met for a person to be diagnosed with gender\\ndysphoria. 5\\n\\n First, there must be “[a] marked incongruence between\\none’s experienced/expressed gender and assigned gender, of\\nat least 6 months’ duration, as manifested by at least two of\\nthe following”:\\n\\n (1) “a marked incongruence between one’s\\n experienced/expressed gender and primary\\n and/or secondary sex characteristics”;\\n\\n (2) “a strong desire to be rid of one’s primary\\n and/or secondary sex characteristics because\\n of a marked incongruence with one’s\\n experienced/expressed gender”;\\n\\n (3) “a strong desire for the primary and/or\\n secondary sex characteristics of the other\\n gender”;\\n\\n (4) “a strong desire to be of the other gender”;\\n\\n (5) “a strong desire to be treated as the other\\n gender”; or\\n\\n\\n\\n\\n 5\\n Each expert in the case used these criteria to determine whether\\nEdmo has gender dysphoria.\\n\\f EDMO V. CORIZON 13\\n\\n (6) “a strong conviction that one has the\\n typical feelings and reactions of the other\\n gender.”\\n\\nAm. Psychiatric Ass’n, Diagnostic and Statistical Manual of\\nMental Disorders 452 (5th ed. 2013) (hereinafter “DSM-5”).\\nSecond, the person’s condition must be associated with\\n“clinically significant distress”—i.e., distress that impairs or\\nseverely limits the person’s ability to function in a\\nmeaningful way and has reached a threshold that requires\\nmedical or surgical intervention, or both. Id. at 453, 458.\\nNot every transgender person has gender dysphoria, and not\\nevery gender dysphoric person has the same medical needs.\\n\\n Gender dysphoria is a serious but treatable medical\\ncondition. Left untreated, however, it can lead to debilitating\\ndistress, depression, impairment of function, substance use,\\nself-surgery to alter one’s genitals or secondary sex\\ncharacteristics, self-injurious behaviors, and even suicide.\\n\\n The district court found that the World Professional\\nAssociation of Transgender Health Standards of Care for the\\nHealth of Transsexual, Transgender, and Gender\\nNonconforming People (“WPATH Standards of Care”) 6\\n“are the internationally recognized guidelines for the\\ntreatment of individuals with gender dysphoria.” Edmo v.\\nIdaho Dep’t of Corr., 358 F. Supp. 3d 1103, 1111 (D. Idaho\\n2018). Most courts agree. See, e.g., De’lonta v. Johnson,\\n708 F.3d 520, 522–23 (4th Cir. 2013); Keohane v. Jones,\\n328 F. Supp. 3d 1288, 1294 (N.D. Fla. 2018), appeal filed,\\n\\n 6\\n The WPATH Standards of Care were formerly referred to as the\\n“Harry Benjamin Standards of Care” and were promulgated by WPATH\\nunder its former name, the “Harry Benjamin International Gender\\nDysphoria Association.” Kosilek, 774 F.3d at 70 & n.3.\\n\\f14 EDMO V. CORIZON\\n\\nNo. 18-14096 (11th Cir. 2018); Norsworthy v. Beard, 87 F.\\nSupp. 3d 1164, 1170 (N.D. Cal.), appeal dismissed &\\nremanded, 802 F.3d 1090 (9th Cir. 2015); Soneeya v.\\nSpencer, 851 F. Supp. 2d 228, 231–32 (D. Mass. 2012). But\\nsee Gibson v. Collier, 920 F.3d 212, 221 (5th Cir. 2019)\\n(“[T]he WPATH Standards of Care reflect not consensus,\\nbut merely one side in a sharply contested medical debate\\nover [GCS].”); cf. Kosilek, 774 F.3d at 76–79 (recounting\\ntestimony questioning the WPATH Standards of Care). And\\nmany of the major medical and mental health groups in the\\nUnited States—including the American Medical\\nAssociation, the American Medical Student Association, the\\nAmerican Psychiatric Association, the American\\nPsychological Association, the American Family Practice\\nAssociation, the Endocrine Society, the National\\nAssociation of Social Workers, the American Academy of\\nPlastic Surgeons, the American College of Surgeons, Health\\nProfessionals Advancing LGBTQ Equality, the HIV\\nMedicine Association, the Lesbian, Bisexual, Gay and\\nTransgender Physician Assistant Caucus, and Mental Health\\nAmerica—recognize the WPATH Standards of Care as\\nrepresenting the consensus of the medical and mental health\\ncommunities regarding the appropriate treatment for\\ntransgender and gender dysphoric individuals.\\n\\n Each expert in this case relied on the WPATH Standards\\nof Care in rendering an opinion. As the State acknowledged\\nto the district court, the WPATH Standards of Care “provide\\nthe best guidance,” and “are the best standards out there.”\\n“There are no other competing, evidence-based standards\\nthat are accepted by any nationally or internationally\\nrecognized medical professional groups.” Edmo, 358 F.\\nSupp. 3d at 1125.\\n\\f EDMO V. CORIZON 15\\n\\n “[B]ased on the best available science and expert\\nprofessional consensus,” the WPATH Standards of Care\\nprovide “flexible clinical guidelines” “to meet the diverse\\nhealth care needs of transsexual, transgender, and gender\\nnonconforming people.” WPATH SOC at 1–2. Treatment\\nunder the WPATH Standards of Care must be\\nindividualized: “[w]hat helps one person alleviate gender\\ndysphoria might be very different from what helps another\\nperson.” Id. at 5. “Clinical departures from the [WPATH\\nStandards of Care] may come about because of a patient’s\\nunique anatomic, social, or psychological situation; an\\nexperienced health professional’s evolving method of\\nhandling a common situation; a research protocol; lack of\\nresources in various parts of the world; or the need for\\nspecific harm reduction strategies.” Id. at 2.\\n\\n The WPATH Standards of Care identify the following\\nevidence-based treatment options for individuals with\\ngender dysphoria:\\n\\n (1) “changes in gender expression and role\\n (which may involve living part time or full\\n time in another gender role, consistent with\\n one’s gender identity)”;\\n\\n (2) “psychotherapy (individual, couple,\\n family, or group) for purposes such as\\n exploring gender identity, role, and\\n expression[,] addressing the negative impact\\n of gender dysphoria and stigma on mental\\n health[,] alleviating internalized\\n transphobia[,] enhancing social and peer\\n support[,] improving body image[,] or\\n promoting resilience”;\\n\\f16 EDMO V. CORIZON\\n\\n (3) “hormone therapy to feminize or\\n masculinize the body”; and\\n\\n (4) “surgery to change primary and/or\\n secondary sex characteristics (e.g.,\\n breasts/chest, external and/or internal\\n genitalia, facial features, body contouring).”\\n\\nId. at 10. The WPATH Standards of Care state that many\\nindividuals “find comfort with their gender identity, role,\\nand expression without surgery.” Id. at 54. For others,\\nhowever, “surgery is essential and medically necessary to\\nalleviate their gender dysphoria.” Id. That group cannot\\nachieve “relief from gender dysphoria . . . without\\nmodification of their primary and/or secondary sex\\ncharacteristics to establish greater congruence with their\\ngender identity.” Id. at 55; see also Jae Sevelius & Valerie\\nJenness, Challenges and Opportunities for Gender-\\nAffirming Healthcare for Transgender Women in Prison,\\n13 Int’l J. Prisoner Health 32, 36 (2017) (“Negative\\noutcomes such as genital self-harm, including autocastration\\nand/or autopenectomy, can arise when gender-affirming\\nsurgeries are delayed or denied.”); George R. Brown &\\nEverett McDuffie, Health Care Policies Addressing\\nTransgender Inmates in Prison Systems in the United States,\\n15 J. Corr. Health Care 280, 287–88 (2009) (describing the\\nauthors’ “firsthand knowledge of completed autocastration\\nand/or autopenectomy in six facilities in four states”).\\n\\n The weight of opinion in the medical and mental health\\ncommunities agrees that GCS is safe, effective, and\\nmedically necessary in appropriate circumstances. See, e.g.,\\nU.S. Dep’t of Health & Human Servs., No. A-13-87,\\nDecision No. 2576, (Dep’t Appeals Bd. May 30, 2014);\\nRandi Ettner, et al., Principles of Transgender Medicine and\\n\\f EDMO V. CORIZON 17\\n\\nSurgery 109–11 (2d ed. 2016); Jordan D. Frey, et al., A\\nHistorical Review of Gender-Affirming Medicine: Focus on\\nGenital Reconstruction Surgery, 14 J. Sexual Med. 991, 991\\n(2017); Cynthia S. Osborne & Anne A. Lawrence, Male\\nPrison Inmates With Gender Dysphoria: When Is Sex\\nReassignment Surgery Appropriate?, 45 Archives of Sexual\\nBehav. 1649, 1651–53 (2016); see also De’lonta, 708 F.3d\\nat 523 (“Pursuant to the Standards of Care, after at least one\\nyear of hormone therapy and living in the patient’s identified\\ngender role, sex reassignment surgery may be necessary for\\nsome individuals for whom serious symptoms persist. In\\nthese cases, the surgery is not considered experimental or\\ncosmetic; it is an accepted, effective, medically indicated\\ntreatment for [gender dysphoria].”).\\n\\n The WPATH criteria for genital reconstruction surgery\\nin male-to-female patients include the following:\\n\\n (1) “persistent, well documented gender\\n dysphoria”;\\n\\n (2) “capacity to make a fully informed\\n decision and to consent for treatment”;\\n\\n (3) “age of majority in a given country”;\\n\\n (4) “if significant medical or mental health\\n concerns are present, they must be well\\n controlled”;\\n\\n (5) “12 continuous months of hormone\\n therapy as appropriate to the patient’s gender\\n goals”; and\\n\\f18 EDMO V. CORIZON\\n\\n (6) “12 continuous months of living in a\\n gender role that is congruent with their\\n gender identity.”\\n\\nWPATH SOC at 60. The parties’ dispute focuses on\\nwhether Edmo satisfied the fourth and sixth criteria.\\n\\n With respect to the fourth criterion, the WPATH\\nStandards of Care provide that coexisting medical or mental\\nhealth concerns unrelated to the person’s gender dysphoria\\ndo not necessarily preclude surgery. Id. at 25. But those\\nconcerns need to be managed prior to, or concurrent with,\\ntreatment of a person’s gender dysphoria. Id. Coexisting\\nmedical or mental health issues resulting from a person’s\\ngender dysphoria are not an impediment under the fourth\\ncriterion. It may be difficult to determine, however, whether\\nmental or medical health concerns result from the gender\\ndysphoria or are unrelated.\\n\\n The WPATH Standards of Care explain that the sixth\\ncriterion—living for 12 months in an identity-congruent\\nrole—is intended to ensure that the person experiences the\\nfull range of “different life experiences and events that may\\noccur throughout the year.” Id. at 61. During that time, the\\npatient should present consistently in her desired gender\\nrole. Id.\\n\\n Scientific studies show that the regret rate for individuals\\nwho undergo GCS is low, in the range of one to two percent.\\nSee, e.g., Osborne & Lawrence, Male Prison Inmates With\\nGender Dysphoria, 45 Archives of Sexual Behav. at 1660;\\nWilliam Byne, et al., Report of the American Psychiatric\\nAssociation Task Force on Treatment of Gender Identity\\nDisorder, 41 Archives of Sexual Behav. 759, 780–81\\n(2012). The district court found, and the State does not\\ndispute on appeal, that Edmo does not have any of the risk\\n\\f EDMO V. CORIZON 19\\n\\nfactors that would make her likely to regret GCS. See Edmo,\\n358 F. Supp. 3d at 1121.\\n\\n The WPATH Standards of Care apply equally to all\\nindividuals “irrespective of their housing situation” and\\nexplicitly state that health care for transgender individuals\\n“living in an institutional environment should mirror that\\nwhich would be available to them if they were living in a\\nnon-institutional setting within the same community.”\\nWPATH SOC at 67. The next update to the WPATH\\nStandards of Care will likewise apply equally to incarcerated\\npersons. The National Commission on Correctional Health\\nCare (“NCCHC”), a leading professional organization in\\nhealth care delivery in the correctional context, endorses the\\nWPATH Standards of Care as the accepted standards for the\\ntreatment of transgender prisoners.\\n\\n In summary, the broad medical consensus in the area of\\ntransgender health care requires providers to individually\\ndiagnose, assess, and treat individuals’ gender dysphoria,\\nincluding for those individuals in institutionalized\\nenvironments. Treatment can and should include GCS when\\nmedically necessary. Failure to follow an appropriate\\ntreatment plan can expose transgender individuals to a\\nserious risk of psychological and physical harm. The State\\ndoes not dispute these points; it contends that GCS is not\\nmedically necessary for Edmo.\\n\\n B. Edmo’s Treatment\\n\\n Edmo is a transgender woman in IDOC custody. Her sex\\nassigned at birth was male, but she identifies as female. In\\nher words, “my brain typically operates female, even though\\nmy body hasn’t corresponded with my brain.”\\n\\f20 EDMO V. CORIZON\\n\\n Edmo has been incarcerated since pleading guilty in\\n2012 to sexual abuse of a 15-year-old male at a house party.\\nEdmo was 21 years old at the time of the criminal offense.\\nEdmo is currently incarcerated at the Idaho State\\nCorrectional Institution (“ISCI”). At the time of the\\nevidentiary hearing, she was 30 years old and due to be\\nreleased from prison in 2021.\\n\\n Edmo has viewed herself as female since age 5 or 6. She\\nstruggled with her gender identity as a child and teenager,\\npresenting herself intermittently as female, but around age\\n20 or 21 she began living fulltime as a woman.\\n\\n Although she identified as female from an early age,\\nEdmo first learned the term “gender dysphoria” and the\\ncontours of that diagnosis around the time of her\\nincarceration. Shortly thereafter, Corizon psychiatrist\\nDr. Scott Eliason diagnosed her with “gender identity\\ndisorder,” now referred to as gender dysphoria. Corizon\\npsychologist Dr. Claudia Lake confirmed that diagnosis.\\n\\n While incarcerated, Edmo has changed her legal name to\\nAdree Edmo and the sex on her birth certificate to “female”\\nto affirm her gender identity. Throughout her incarceration,\\nEdmo has consistently presented as female, despite receiving\\nmany disciplinary offense reports for doing so. For example,\\nwhen able to do so, Edmo has worn her hair in feminine\\nhairstyles and worn makeup, for which she has received\\nmultiple disciplinary offense reports. 7 Medical providers\\nhave documented Edmo’s feminine presentation since 2012.\\n\\n\\n 7\\n Before the evidentiary hearing, Edmo tried to receive access to\\nfemale commissary items, such as women’s underwear. Most of her\\nrequests were denied. On the eve of the evidentiary hearing, IDOC\\n\\f EDMO V. CORIZON 21\\n\\n Neither the parties nor their experts dispute that Edmo\\nsuffers from gender dysphoria. That dysphoria causes Edmo\\nto feel “depressed,” “disgusting,” “tormented,” and\\n“hopeless.”\\n\\n To alleviate Edmo’s gender dysphoria, prison officials\\nhave, since 2012, provided hormone therapy. Edmo has\\nfollowed and complied with her hormone therapy regimen,\\nwhich helps alleviate her gender dysphoria to some extent.\\nThe hormones “clear[] [her] mind” and have resulted in\\nbreast growth, body fat redistribution, and changes in her\\nskin. Today, Edmo is hormonally confirmed, which means\\nthat she has the hormones and secondary sex characteristics\\n(characteristics, such as women’s breasts, that appear during\\npuberty but are not part of the reproductive system) of an\\nadult female. Edmo has gained the maximum physical\\nchanges associated with hormone treatment.\\n\\n Hormone therapy has not completely alleviated Edmo’s\\ngender dysphoria. Edmo continues to experience significant\\ndistress related to gender incongruence. Much of that\\ndistress is caused by her male genitalia. Edmo testified that\\nshe feels “depressed, embarrassed, [and] disgusted” by her\\nmale genitalia and that this is an “everyday reoccurring\\nthought.” Her medical records confirm her disgust, noting\\nrepeated efforts by Edmo to purchase underwear to keep, in\\nEdmo’s words, her “disgusting penis” out of sight.\\n\\n In addition to her gender dysphoria, Edmo suffers from\\nmajor depressive disorder with anxiety and drug and alcohol\\naddiction, although her addiction has been in remission\\n\\n\\namended its policy concerning the treatment of gender dysphoric\\nprisoners to increase transgender women’s access to female commissary\\nitems.\\n\\f22 EDMO V. CORIZON\\n\\nwhile incarcerated. Edmo has taken her prescribed\\nmedications for depression and anxiety. Prison officials\\nhave also provided Edmo mental health treatment to help her\\nwork through her serious underlying mental health issues\\nand a pre-incarceration history of trauma, abuse, and suicide\\nattempts. Edmo sees her psychiatrist when scheduled. But\\nEdmo does not see her treating clinician, Krina Stewart,\\nbecause Edmo does not believe Stewart is qualified to treat\\nher gender dysphoria. Edmo has attended group therapy\\nsessions inconsistently.\\n\\n In September 2015, Edmo attempted to castrate herself\\nfor the first time using a disposable razor blade. 8 Before\\ndoing so, she left a note to alert officials that she was not\\n“trying to commit suicide,” and was instead “only trying to\\nhelp [her]self.” Edmo did not complete the castration,\\nthough she continued to report thoughts of self-castration in\\nthe following months.\\n\\n On April 20, 2016, Dr. Eliason evaluated Edmo for GCS.\\nAt the time, IDOC’s policy concerning the treatment of\\ngender dysphoric prisoners provided that GCS “will not be\\nconsidered for individuals within [IDOC], unless determined\\nmedically necessary by” the treating physician. 9 Corizon’s\\npolicy does not mention GCS.\\n\\n In his evaluation, Dr. Eliason noted that Edmo reported\\nshe was “doing alright.” He also noted that Edmo had been\\non hormone replacement therapy for the last year and a half,\\nbut that she felt she needed more. He reported that Edmo\\n 8\\n She had previously reported thoughts of self-castration to\\nclinicians.\\n 9\\n IDOC revised its policy shortly before the evidentiary hearing, but\\nits revised policy contains functionally identical language.\\n\\f EDMO V. CORIZON 23\\n\\nhad stated that hormone replacement therapy helped\\nalleviate her gender dysphoria, but she remained frustrated\\nwith her male anatomy.\\n\\n Dr. Eliason indicated that Edmo appeared feminine in\\ndemeanor and interaction style. He also indicated that Edmo\\nhad previously attempted to “mutilate her genitalia” because\\nof the severity of her distress. Dr. Eliason later testified that,\\nat the time of his evaluation, he felt that Edmo’s gender\\ndysphoria “had risen to another level,” as evidenced by her\\nself-castration attempt.\\n\\n But Dr. Eliason also flagged that he had spoken to prison\\nstaff about Edmo’s behavior and they explained it was\\n“notable for animated affect and no observed distress.” He\\nsimilarly noted that he had personally observed Edmo and\\ndid not see significant dysphoria; instead, she “looked\\npleasant and had a good mood.”\\n\\n As to GCS, Dr. Eliason explained in his notes that while\\nmedical necessity for GCS is “not very well defined and is\\nconstantly shifting,” in his view, GCS would be medically\\nnecessary in at least three situations: (1) “congenital\\nmalformations or ambiguous genitalia,” (2) “severe and\\ndevastating dysphoria that is primarily due to genitals,” or\\n(3) “some type of medical problem in which endogenous\\nsexual hormones were causing severe physiological\\ndamage.” Dr. Eliason concluded that Edmo “does not meet\\nany of those . . . criteria” and, for that reason, GCS is not\\nmedically necessary for her.\\n\\n Dr. Eliason instead concluded that hormone therapy and\\nsupportive counseling suffice to treat Edmo’s gender\\ndysphoria for the time being, despite recognizing that Edmo\\nhad attempted self-castration on that regimen. Dr. Eliason\\n\\f24 EDMO V. CORIZON\\n\\nindicated that he would continue to monitor and assess\\nEdmo.\\n\\n Dr. Eliason staffed Edmo’s evaluation with Dr. Jeremy\\nStoddart, Dr. Murray Young, and Jeremy Clark, who all\\nagreed with his assessment. They did not observe Edmo;\\nrather, they agreed with Dr. Eliason’s recommended\\ntreatment as he presented it to them. The record is sparse on\\nthe qualifications of Dr. Stoddart and Dr. Young, but Clark\\nhas never personally treated anyone with gender dysphoria\\nand was not qualified under IDOC policy to assess whether\\nGCS would be appropriate for Edmo.\\n\\n Dr. Eliason also discussed his evaluation with IDOC’s\\nManagement and Treatment Committee (“MTC”), a multi-\\ndisciplinary team composed of medical providers, mental\\nhealth clinicians, IDOC’s Chief Psychologist, and prison\\nleadership. The MTC meets periodically to evaluate and\\naddress the unique medical, mental health, and housing\\nneeds of prisoners with gender dysphoria. The committee\\n“does not make any individual treatment decisions\\nregarding” treatment for inmates with gender dysphoria.\\n“Those determinations are made by the individual clinicians\\nor the medical staff employed by Corizon.” The MTC\\nagreed with Dr. Eliason’s assessment.\\n\\n Although not mentioned in his April 20, 2016 notes,\\nDr. Eliason testified at the evidentiary hearing that he\\nconsidered the WPATH Standards of Care when\\ndetermining Edmo’s treatment. Citing those standards,\\nDr. Eliason testified that he did not believe GCS was\\nappropriate for two reasons: (1) because mental health\\nissues separate from Edmo’s gender dysphoria were not\\n“fully in adequate control” and (2) because Edmo had not\\nlived in her identified gender role for 12 months outside of\\nprison. He explained that Edmo needed to experience\\n\\f EDMO V. CORIZON 25\\n\\n“living as a woman” around “her real social network – her\\nfamily and friends on the outside” so that she could\\n“determine whether or not she felt like that was her real\\nidentity.”\\n\\n Edmo was never evaluated for GCS again, but the MTC\\nconsidered her gender dysphoria and treatment plan during\\nlater meetings. The MTC continues to believe that GCS is\\nnot medically necessary or appropriate for Edmo.\\n\\n In December 2016, Edmo tried to castrate herself for the\\nsecond time. A medical note from the incident reports that\\nEdmo said she no longer wanted her testicles. Edmo\\nreported to medical providers that she was “feeling\\nangry/frustrated that [she] was not receiving the help desired\\nrelated to [her] gender dysphoria. Inmate Edmo’s actions\\nwere reported as a method to stop/cease testosterone\\nproduction in Edmo’s body. Edmo denied suicidal ideation\\n. . . .”\\n\\n Edmo’s second attempt was more successful than the\\nfirst. She was able to open her testicle sac with a razor blade\\nand remove one testicle. She abandoned her attempt,\\nhowever, when there was too much blood to continue. She\\nthen sought medical assistance and was transported to a\\nhospital, where her testicle was repaired. Edmo was\\nreceiving hormone therapy both times she attempted self-\\ncastration.\\n\\n Edmo testified that she was disappointed in herself for\\ncoming so close but failing to complete her self-castration\\nattempts. She also testified that she continues to actively\\nthink about self-castration. To avoid acting on those\\nthoughts and impulses, Edmo “self-medicat[es]” by cutting\\nher arms with a razor. She says that the physical pain helps\\n\\f26 EDMO V. CORIZON\\n\\nto ease the “emotional torment” and mental anguish her\\ngender dysphoria causes her.\\n\\n Edmo further testified that she expects GCS to help\\nalleviate some of her gender dysphoria. In particular, she\\ntestified that she expects GCS to help her avoid having “as\\nmuch depression about myself and my physical body. I\\ndon’t think I will be so anxious that people are always\\nknowing I’m different . . . .” Edmo recognizes, however,\\nthat GCS “is not a fix-all”: “[i]t’s not a magic operation. . . .\\nI’m still going to have to face the same stressors that we all\\nface in everyday life . . . .”\\n\\n C. Initiation of this Action\\n\\n Edmo filed a pro se complaint on April 6, 2017. She also\\nmoved for a temporary restraining order, a preliminary\\ninjunction, and the appointment of counsel.\\n\\n Edmo’s motion for appointment of counsel was granted\\nin part, and counsel for Edmo appeared in June and August\\n2017. Counsel withdrew Edmo’s pro se motion for\\npreliminary injunction shortly thereafter.\\n\\n On September 1, 2017, Edmo filed an amended\\ncomplaint asserting claims under 42 U.S.C. § 1983, the\\nEighth Amendment, the Fourteenth Amendment, the\\nAmericans with Disabilities Act, the Affordable Care Act,\\nand for common law negligence. She named as defendants\\nIDOC, Henry Atencio (Director of IDOC), Jeff Zmuda\\n(Deputy Director of IDOC), Howard Keith Yordy (former\\nWarden of ISCI), Dr. Richard Craig (Chief Psychologist at\\nISCI), Rona Siegert (Health Services Director at ISCI),\\nCorizon, Dr. Eliason, Dr. Young, and Dr. Catherine\\nWhinnery (Corizon employee).\\n\\f EDMO V. CORIZON 27\\n\\n Through counsel, Edmo filed a renewed motion for a\\npreliminary injunction on June 1, 2018. Among other relief,\\nEdmo sought an order requiring the State to provide her with\\na referral to a qualified surgeon and access to GCS.\\n\\n The State moved to extend the time to respond to Edmo’s\\nmotion. After a status conference, the district court set an\\nevidentiary hearing for October 10, 11, and 12, 2018. The\\ncourt permitted the parties to undertake four months of\\nextensive fact and expert discovery in preparation for the\\nhearing.\\n\\n D. The Evidentiary Hearing\\n\\n At the evidentiary hearing, each side had eight hours to\\npresent its case. The district court heard live testimony from\\nseven witnesses over three days. It also considered\\nthousands of pages of exhibits, including Edmo’s medical\\nrecords. With the parties’ agreement, the court also\\npermitted the State to submit declarations in lieu of live\\ntestimony and permitted Edmo to impeach the declarations\\nwith deposition testimony.\\n\\n At the outset of the hearing, the district court noted that\\n“[w]e’re here on a hearing for a temporary injunction,” but\\nit explained that “it’s hard for me to envision this hearing\\nbeing anything but a hearing on a final injunction[,] at least\\nas to” the injunctive relief ordering GCS. The court stated\\nthat it was unsure whether that made a difference, and it\\nasked the parties to address at some point whether the\\nhearing was for a preliminary injunction or a permanent\\ninjunction. Notably, the State did not do so.\\n\\n The district court heard testimony from three percipient\\nwitnesses: Edmo, Dr. Eliason (the Corizon physician), and\\nJeremy Clark (an IDOC clinician who did not meet IDOC’s\\n\\f28 EDMO V. CORIZON\\n\\ncriteria to assess Edmo for GCS). Their relevant testimony\\nis largely recounted above.\\n\\n It also heard testimony from four expert witnesses, two\\neach for Edmo and the State. Dr. Randi Ettner, Ph.D. in\\npsychology, testified first for Edmo. Dr. Ettner is one of the\\nauthors of the current (seventh) version of the WPATH\\nStandards of Care. She has been a WPATH member since\\n1993 and chairs its Institutionalized Persons Committee. Dr.\\nEttner has authored or edited many peer-reviewed\\npublications on the treatment of gender dysphoria and\\ntransgender health care more broadly, including the leading\\ntextbook used in medical schools on the subject. She also\\ntrains medical and mental health providers on treating people\\nwith gender dysphoria. Dr. Ettner has been retained as an\\nexpert witness on gender dysphoria and its treatment in\\nmany court cases, and she has been appointed as an\\nindependent expert by one federal court to evaluate an\\nincarcerated person for GCS.\\n\\n Dr. Ettner has evaluated, diagnosed, and treated between\\n2,500 and 3,000 individuals with gender dysphoria. She has\\nreferred about 300 people for GCS. She has also refused to\\nrecommend surgery for some patients who have requested it.\\nShe believes that not everyone who has gender dysphoria\\nneeds GCS. Dr. Ettner also has “[e]xtensive experience”\\ntreating and providing post-operative care for patients who\\nhave undergone GCS.\\n\\n Dr. Ettner has assessed approximately 30 incarcerated\\nindividuals with gender dysphoria for GCS and other\\nmedical care, but she has not treated incarcerated patients.\\nShe has not worked in a prison and she is not a Certified\\nCorrectional Healthcare Professional.\\n\\f EDMO V. CORIZON 29\\n\\n Based on her evaluation of Edmo and a review of\\nEdmo’s medical records, Dr. Ettner diagnosed Edmo with\\ngender dysphoria, depressive disorder, anxiety, and suicidal\\nideation. In Dr. Ettner’s opinion, GCS is medically\\nnecessary for Edmo and should be immediately performed.\\nShe explained that most patients with gender dysphoria do\\nnot require GCS, but Edmo requires it because hormone\\ntherapy has been inadequate for her and Edmo has attempted\\nto remove her own testicles. Dr. Ettner further explained that\\nGCS would give Edmo congruent genitalia, eliminating the\\nsevere distress Edmo experiences due to her male anatomy.\\n\\n Dr. Ettner further opined that Edmo meets the WPATH\\ncriteria for GCS. She explained that Edmo has “persistent\\nand well-documented long-standing gender dysphoria”;\\nEdmo “has no thought disorders and no impaired reality\\ntesting”; Edmo is the age of majority in this country;\\nalthough Edmo has depression and anxiety, those conditions\\ndo not “impair her ability to undergo surgery” because they\\nare “as controlled as [they] can be”; Edmo has had six years\\nof hormone therapy; and Edmo has lived for more than one\\nyear “as a woman to the best of her ability in a male prison.”\\n\\n More specifically, as to the fourth criterion, Dr. Ettner\\nopined that Edmo does not have mental health concerns that\\nwould preclude GCS. She explained that Edmo’s depression\\nand anxiety are as “controlled as can be” because Edmo “is\\ntaking the maximum amount of medication that controls\\ndepression.” Dr. Ettner noted that Edmo has complied with\\ntaking her prescribed medications and that psychotherapy is\\nnot “a precondition for surgery” under the WPATH\\nStandards of Care. She also flagged that Edmo has the\\ncapacity to comply with her postsurgical treatment, as\\nevidenced by her compliance with her hormone therapy to\\ndate.\\n\\f30 EDMO V. CORIZON\\n\\n As to the clinical significance of Edmo’s self-castration\\nattempts and cutting behaviors, Dr. Ettner explained that\\nneither behavior indicates that Edmo has inadequately\\ncontrolled mental health concerns. Rather, those behaviors\\nindicate “the need for treatment for gender dysphoria.”\\nDr. Ettner explained that\\n\\n when an individual who is not psychotic or\\n delusional attempts what we call surgical\\n self-treatment – because we don’t regard\\n removal of the testicles or attempted removal\\n of the testicles as either mutilation or self-\\n harm – we regard it as an intentional attempt\\n to remove the target organ that produces\\n testosterone, which, in fact, is the cure for\\n gender dysphoria.\\n\\nIn Dr. Ettner’s opinion, Edmo’s depression and anxiety “will\\nbe attenuated post surgery.”\\n\\n Dr. Ettner opined that Edmo satisfies the sixth criterion\\nbecause she has lived “as a woman to the best of her ability\\nin a male prison.” Dr. Ettner based her opinion on Edmo’s\\n“appearance . . . , her disciplinary records, which indicated\\nthat she had attempted to wear her hair in a feminine\\nhairstyle and to wear makeup even though that was against\\nthe rules and she was – received some sort of disciplinary\\naction for that, and her – the way that she was receiving\\nfemale undergarments and had developed the stigma of\\nfemininity, the secondary sex characteristics, breast\\ndevelopment, et cetera.”\\n\\n Dr. Ettner opined that if Edmo does not receive GCS,\\n“[t]he risks would be, as typical in inadequately treated or\\nuntreated gender dysphoria, either surgical self-treatment,\\nemotional decompensation, or suicide.” Dr. Ettner\\n\\f EDMO V. CORIZON 31\\n\\nexplained that Edmo “is at particular risk of suicide given\\nthat she has a high degree of suicide ideation.” If, on the\\nother hand, Edmo receives surgery, Dr. Ettner opined that\\n\\n [i]t would eliminate the gender dysphoria. It\\n would provide a level of wellbeing that she\\n hasn’t had previously. It would eliminate\\n 80 percent of the testosterone in her body,\\n necessitating a lower dose of hormones going\\n forward, which would be particularly helpful\\n given that she has elevated liver enzymes.\\n And it would, I believe, eliminate much of the\\n depression and the attendant symptoms that\\n she is experiencing.\\n\\n Dr. Ryan Gorton, M.D., also testified for Edmo.\\nDr. Gorton is an emergency medicine physician. He also\\nworks pro bono at a clinic serving uninsured patients or those\\nwith Medicare or Medicaid. Many of those patients have\\nmental health conditions or have been in prison. He has\\npublished peer-reviewed articles on the treatment of gender\\ndysphoria, and he has been qualified as an expert witness in\\ncases involving transgender health care. Dr. Gorton also\\nprovides training on transgender health care issues to many\\ngroups, is a member of WPATH, and serves on WPATH’s\\nTransgender Medicine and Research Committee and its\\nInstitutionalized Persons Committee.\\n\\n Dr. Gorton has been the primary care physician for about\\n400 patients with gender dysphoria. At the time of the\\nevidentiary hearing, Dr. Gorton was treating approximately\\n100 patients with gender dysphoria. Dr. Gorton has assessed\\npatients for gender dysphoria, initiated and monitored\\nhormone treatment, referred patients for mental health\\ntreatment, and determined the appropriateness of GCS. At\\n\\f32 EDMO V. CORIZON\\n\\nthe time of the evidentiary hearing, Dr. Gorton was\\nproviding follow-up care for about 30 patients who had\\nvaginoplasty. Dr. Gorton has no experience treating\\ntransgender inmates and is not a Certified Correctional\\nHealthcare Professional.\\n\\n Based on his review of Edmo’s medical records and his\\nin-person evaluation of Edmo, Dr. Gorton opined that GCS\\nis medically necessary for Edmo and that she meets the\\nWPATH criteria for GCS. He explained that Edmo has\\n“persistent well-documented gender dysphoria,” as shown in\\nher prison medical records; she has the capacity “to make a\\nfully informed decision and to consent for treatment”\\nbecause “she didn’t seem at all impaired in her decision-\\nmaking capacity”; she is the age of majority; she has\\ndepression and anxiety, “but they are not to a level that\\nwould preclude her getting [GCS]”; she had 12 consecutive\\nmonths of hormone therapy; and she has been living in her\\n“target gender role . . . despite an environment that’s very\\nhostile to that and some negative consequences that she has\\nexperienced because of that.”\\n\\n Dr. Gorton further opined that if Edmo “is not provided\\nsurgery, there is a very substantial chance she will try to\\nattempt self-surgery again. And that’s especially worrisome\\ngiven her attempts have been progressive. . . . So I think she\\nmight be successful” on her next attempt. He predicted that\\nthere is little chance that Edmo’s gender dysphoria will\\nimprove without surgery. Conversely, Dr. Gorton\\nanticipated that Edmo is unlikely to regret surgery because\\n“her gender dysphoria is very genital-focused” and regret\\nrates among GCS patients are very low.\\n\\n Dr. Gorton also opined that Edmo’s self-castration\\nattempts demonstrate “that she has severe genital-focused\\ngender dysphoria and that she is not getting the medically\\n\\f EDMO V. CORIZON 33\\n\\nnecessary treatment to alleviate that.” He elaborated that\\nEdmo’s depression and anxiety are not driving Edmo’s self-\\ncastration attempts: “there [are] a lot of people with\\ndepression and anxiety who don’t remove their testicles.”\\n\\n Finally, Dr. Gorton criticized Dr. Eliason’s evaluation of\\nEdmo. He explained that he disagreed with Dr. Eliason’s\\nconclusion that Edmo does not need GCS and he also\\ndisagreed with the three “criteria” Dr. Eliason gave for when\\nGCS would be necessary. Dr. Gorton criticized Dr.\\nEliason’s first criterion—that GCS could be needed where\\nthere is “congenital malformation or ambiguous genitalia”—\\nbecause that situation “isn’t even germane to transgender\\npeople”; rather, it relates to “people with intersex\\nconditions.” As to the second criterion—that GCS could be\\nneeded when a patient is suffering from “severe and\\ndevastating gender dysphoria that is primarily due to\\ngenitals”—Dr. Gorton pointed out that the WPATH\\nStandards of Care for surgery require only “clear and\\nsignificant dysphoria.” And even applying Dr. Eliason’s\\nhigher bar, Dr. Gorton explained that Edmo would still\\nqualify for GCS because she has twice attempted self-\\ncastration, demonstrating “severe genital-focused\\ndysphoria.” Finally, Dr. Gorton characterized Dr. Eliason’s\\nthird criterion—that GCS could be needed in situations when\\n“endogenous sexual hormones were causing severe\\nphysiological damage”—as “bizarre.” Dr. Gorton could not\\nconjure “a clinical circumstance where that would be the\\ncase that your hormones that your body produces are\\nattacking you . . . . I just don’t understand what [Dr. Eliason]\\nis talking about there.”\\n\\n Dr. Keelin Garvey, M.D., testified for the State.\\nDr. Garvey is a psychiatrist and Certified Correctional\\nHealthcare Professional. As the former Chief Psychiatrist of\\n\\f34 EDMO V. CORIZON\\n\\nthe Massachusetts Department of Corrections, Dr. Garvey\\nchaired the Gender Dysphoria Treatment Committee. She\\ndirectly treated a “couple of patients” with gender dysphoria\\nearlier in her career as Deputy Medical Director, but she has\\nnot done so in recent years. Prior to evaluating Edmo, Dr.\\nGarvey had never evaluated a patient in person to determine\\nwhether that person needed GCS. Dr. Garvey has never\\nrecommended a patient for GCS, and she has not done\\nfollow-up care with a person who has received GCS.\\n\\n Based on her evaluation of Edmo and a review of\\nEdmo’s medical records, Dr. Garvey diagnosed Edmo with\\ngender dysphoria, major depressive disorder, alcohol use\\ndisorder, stimulant use disorder, and opioid use disorder.\\nShe explained that the latter three are in remission.\\n\\n Relying on the WPATH Standards of Care, Dr. Garvey\\nopined that GCS is not medically necessary for Edmo. 10 Dr.\\nGarvey first explained that Edmo does not meet the first\\nWPATH Standards of Care criterion—“persistent, well\\ndocumented gender dysphoria”—because of a lack of\\nevidence in pre-incarceration medical records that Edmo\\npresented as female before her time in prison. Dr. Garvey\\nacknowledged, however, that Edmo has been presenting as\\nfemale since 2012 and that she has been diagnosed with\\ngender dysphoria since that time.\\n\\n Dr. Garvey then explained that Edmo does not meet the\\nfourth criterion—“medical/mental health concerns must be\\nwell controlled”—because Edmo “is actively self-injuring.”\\nDr. Garvey elaborated that “self-injury in any form is never\\n\\n 10\\n Dr. Garvey testified that she relies on the WPATH Standards of\\nCare and the NCCHC guidelines adopting those standards when treating\\ninmates with gender dysphoria.\\n\\f EDMO V. CORIZON 35\\n\\nconsidered a healthy or productive coping mechanism” and\\nthat she would like to see Edmo “develop further coping\\nskills that she would be able to use following surgery so that\\nshe is not engaging in self-injury after surgery.” Dr.\\nGarvey’s concern is that GCS is a “stressful undertaking”\\nand Edmo lacks “effective coping strategies” to deal with the\\nstress.\\n\\n Finally, Dr. Garvey testified that Edmo does not meet the\\nsixth criterion—“12 continuous months of living in a gender\\nrole that is congruent with gender identity”—because Edmo\\nhas not presented as female outside of prison and “there [are]\\nchallenges to using her time in a men’s prison as this real-\\nlife experience because it doesn’t offer her the opportunity\\nto actually experience all those things she is going to go\\nthrough on the outside.”\\n\\n Dr. Joel Andrade, Ph.D. in social work, also testified for\\nthe State. He is a licensed clinical social worker and is a\\nCertified Correctional Healthcare Professional with an\\nemphasis in mental health. Dr. Andrade has over a decade\\nof experience providing and supervising the provision of\\ncorrectional mental health care, including directing and\\noverseeing the treatment of inmates diagnosed with gender\\ndysphoria in the custody of the Massachusetts Department\\nof Corrections in his roles as clinical director, chair of the\\nGender Dysphoria Supervision Group, and member of the\\nGender Dysphoria Treatment Committee.\\n\\n As a member of the Gender Dysphoria Treatment\\nCommittee, Dr. Andrade recommended GCS for two\\ninmates. But the recommendations were contingent on the\\ninmates living in a women’s prison for approximately\\n12 months before the surgery. The Massachusetts\\nDepartment of Corrections, like IDOC, houses prisoners\\n\\f36 EDMO V. CORIZON\\n\\naccording to their genitals, so the inmates had not been\\nmoved (nor had their surgery occurred).\\n\\n Dr. Andrade has never directly treated patients with\\ngender dysphoria, nor has he been a treating clinician for a\\npatient who has had GCS. His “experience with gender\\ndysphoria comes almost exclusively from [his] participation\\non the Massachusetts Department of Corrections[’] Gender\\nDysphoria Treatment Committee and Supervision Group.”\\nDr. Andrade did not qualify, under the IDOC gender\\ndysphoria policy in effect at the time of his assessment of\\nEdmo, to assess a person for GCS because he is neither a\\npsychologist nor a physician.\\n\\n Based on his evaluation of Edmo and a review of her\\nmedical records, Dr. Andrade diagnosed Edmo with “major\\ndepressive disorder, recurrent, in partial remission,”\\n“generalized anxiety disorder,” “alcohol use disorder,\\nsevere,” and gender dysphoria. Dr. Andrade also diagnosed\\nEdmo with borderline personality disorder. The district\\ncourt did not credit this diagnosis, however, because no other\\nperson (including the State’s other expert, Dr. Garvey) has\\never diagnosed Edmo with borderline personality disorder\\nand Dr. Andrade was unable to identify his criteria for this\\ndiagnosis. Edmo, 358 F. Supp. 3d at 1120. The record\\namply supports the district court’s finding in this respect.\\n\\n Dr. Andrade opined that Edmo does not meet the\\nWPATH criteria for GCS. He explained that, based on his\\nreview of Edmo’s pre-incarceration records, Edmo did not\\npresent as female or discuss her gender dysphoria before\\nincarceration. Dr. Andrade testified that he would like to see\\nEdmo live as female outside of a correctional setting before\\nreceiving GCS, or, at the least, live in a women’s prison first.\\nIDOC, however, houses prisoners according to their\\ngenitals. Dr. Andrade also explained that Edmo needs to\\n\\f EDMO V. CORIZON 37\\n\\nwork through some of her trauma, particularly sexual abuse\\nthat she suffered, and other mental health concerns before\\nreceiving surgery. Dr. Andrade opined that Edmo’s mental\\nhealth issues will not be cured by GCS.\\n\\n At the close of the hearing, the district court reiterated\\nthat it was unsure “how we can hear [Edmo’s request for\\nGCS] on a preliminary injunction. . . . [I]f I order it, then it’s\\ndone.” The court further suggested that the request for GCS\\ncould “only be resolved in a final hearing” and noted that it\\nhad, in effect, “treated this hearing as [a] final hearing on the\\nissue.”\\n\\n The court, as it had done at the outset of the hearing,\\nasked the parties to address whether the hearing was for a\\npreliminary or permanent injunction. In response, Edmo\\ncontended that the court could order GCS in a preliminary\\ninjunction. The State did not address the court’s question. It\\ninstead contended that the standard for a mandatory\\ninjunction—which can be preliminary or permanent—\\nshould apply.\\n\\n E. The District Court’s Decision\\n\\n The district court rendered its decision on December 13,\\n2018. After recounting the evidence and making extensive\\nfactual findings, the district court began its analysis by\\nnoting that it was unsure whether the standard for a\\npreliminary injunction or the standard for a permanent\\ninjunction applied. The court noted that “the nature of the\\nrelief requested in this case, coupled with the extensive\\nevidence presented by the parties over a 3-day evidentiary\\nhearing, [may have] effectively converted these proceedings\\ninto a final trial on the merits of the plaintiff’s request for\\npermanent injunctive relief.” Edmo, 358 F. Supp. 3d at 1122\\nn.1. It also indicated that “both parties appear to have treated\\n\\f38 EDMO V. CORIZON\\n\\nthe evidentiary hearing” as a final trial on the merits. Id. The\\ndistrict court explained that the difference was immaterial,\\nhowever, because Edmo was entitled to relief under either\\nstandard. Id.\\n\\n On the merits, the district court concluded that Edmo had\\nestablished her Eighth Amendment claim. The district court\\nfirst held that Edmo suffers from gender dysphoria, which is\\nundisputedly “a serious medical condition.” Id. at 1124.\\n\\n It then concluded that GCS is medically necessary to\\ntreat Edmo’s gender dysphoria. See id. at 1124–26. In a\\ncarefully considered, 45-page opinion, the district court\\nspecifically found “credible the testimony of Plaintiff’s\\nexperts Drs. Ettner and Gorton, who have extensive personal\\nexperience treating individuals with gender dysphoria both\\nbefore and after receiving gender confirmation surgery,” and\\nwho opined that GCS was medically necessary. Id. at 1125.\\nThe court rejected the contrary opinions of the State’s\\nexperts because “neither Dr. Garvey nor Dr. Andrade has\\nany direct experience with patients receiving gender\\nconfirmation surgery or assessing patients for the medical\\nnecessity of gender confirmation surgery,” and neither of the\\nState’s experts had meaningful “experience treating patients\\nwith gender dysphoria other than assessing them for the\\nexistence of the condition.” Id. The district court also noted\\nthat the State’s “experts appear to misrepresent the WPATH\\nStandards of Care by concluding that Ms. Edmo, despite\\npresenting as female since her incarceration in 2012, cannot\\nsatisfy the WPATH criteria because she has not presented as\\nfemale outside of the prison setting.” Id. As the district\\ncourt noted, “there is no requirement in the WPATH\\nStandards of Care that a patient live for twelve months in his\\nor her gender role outside of prison before becoming eligible\\nfor” GCS. Id. (quotation omitted).\\n\\f EDMO V. CORIZON 39\\n\\n Finally, the district court explained that the State was\\ndeliberately indifferent to Edmo’s gender dysphoria because\\nit “fail[ed] to provide her with available treatment that is\\ngenerally accepted in the field as safe and effective, despite\\nher actual harm and ongoing risk of future harm including\\nself-castration attempts, cutting, and suicidal ideation.” Id.\\nat 1126–27. The district court also stated that the evidence\\n“suggest[ed] that Ms. Edmo has not been provided gender\\nconfirmation surgery because Corizon and IDOC have a de\\nfacto policy or practice of refusing this treatment for gender\\ndysphoria to prisoners,” which amounts to deliberate\\nindifference. Id. at 1127.\\n\\n After analyzing the merits, the district court concluded\\nthat Edmo satisfied the other prerequisites to injunctive\\nrelief. Id. at 1127–28. The district court found that, given\\nEdmo’s continuing emotional distress and self-castration\\nattempts, “Edmo is at serious risk of life-threatening self-\\nharm” if she does not receive GCS. Id. at 1128. The State,\\non the other hand, had not shown that it would be harmed if\\nordered to provide GCS, so the equities favored Edmo. Id.\\n\\n Having concluded that Edmo was entitled to an\\ninjunction, the court ordered the State “to provide Plaintiff\\nwith adequate medical care, including gender confirmation\\nsurgery.” Id. at 1129. It ordered the State to “take all actions\\nreasonably necessary to provide Ms. Edmo gender\\nconfirmation surgery as promptly as possible and no later\\nthan six months from the date of this order.” Id.\\n\\n F. Appellate Proceedings\\n\\n The State filed timely notices of appeal on January 9,\\n2019. It also asked the district court to stay its order pending\\nappeal. The district court denied the State’s motion on\\nMarch 4.\\n\\f40 EDMO V. CORIZON\\n\\n The State then filed in this court a motion to stay pending\\nappeal. A motions panel granted that motion. Edmo\\nsubsequently moved to amend the stay to allow her to\\nundergo a previously scheduled pre-surgery consultation.\\nThe motions panel granted that motion and amended the\\nstay.\\n\\n On April 3, the State filed an “urgent motion” to dismiss\\nthis appeal as moot. We indicated on April 5 that our court\\nwould consider that motion with the merits, not on an urgent\\nbasis.\\n\\n After hearing oral argument on May 16, we ordered a\\nlimited remand to the district court to clarify three points.\\nRelevant here, we asked the district court to clarify whether\\nit granted Edmo a permanent injunction in its December 13,\\n2018 order. The district court clarified that it “granted\\npermanent injunctive relief.” Edmo v. Idaho Dep’t of Corr.,\\nNo. 1:17-CV-00151-BLW, 2019 WL 2319527, at *2 (D.\\nIdaho May 31, 2019). We also asked the district court to\\nclarify whether it had concluded that Edmo had succeeded\\non the merits of her Eighth Amendment claim. The district\\ncourt responded that it had. Id.\\n\\n Having received the district court’s response to our\\nlimited remand order, we proceed to the issues on appeal.\\nThe State challenges the district court’s grant of injunctive\\nrelief to Edmo on multiple grounds. It contends that this\\nappeal is moot because the injunction did not comply with\\nthe PLRA and has, for that reason, automatically expired. It\\ncontends that the decision not to provide GCS to Edmo\\nreflects a difference of prudent medical opinion and cannot\\nsupport an Eighth Amendment claim. It contends that Edmo\\nwill not be irreparably harmed absent an injunction. It\\ncontends that the injunction is overbroad. Finally, it\\ncontends that, to the extent the district court converted the\\n\\f EDMO V. CORIZON 41\\n\\nevidentiary hearing into a final trial on the merits of Edmo’s\\nrequest for GCS, it was provided inadequate notice and the\\ncourt violated its right to a jury trial.\\n\\n II. Mootness\\n\\n “We first address, as we must, the question of mootness\\n. . . .” Shell Offshore Inc. v. Greenpeace, Inc., 815 F.3d 623,\\n628 (9th Cir. 2016). An appeal is moot “[w]hen events\\nchange such that the appellate court can no longer grant ‘any\\neffectual relief whatever to the prevailing party.’” Id.\\n(quoting City of Erie v. Pap’s A.M., 529 U.S. 277, 287\\n(2000)). In those circumstances, we “lack[] jurisdiction and\\nmust dismiss the appeal.” Id.\\n\\n The State contends that the injunction does not comply\\nwith provisions of the PLRA and, for that reason, has\\nautomatically expired under the terms of the statute.\\nRelevant here, the PLRA provides that a\\n\\n court shall not grant or approve any\\n prospective relief unless the court finds that\\n such relief is narrowly drawn, extends no\\n further than necessary to correct the violation\\n of the Federal right, and is the least intrusive\\n means necessary to correct the violation of\\n the Federal right. The court shall give\\n substantial weight to any adverse impact on\\n public safety or the operation of a criminal\\n justice system caused by the relief.\\n\\n18 U.S.C. § 3626(a)(1)(A). Courts often refer to this\\nprovision as the “need-narrowness-intrusiveness” inquiry.\\nGraves v. Arpaio, 623 F.3d 1043, 1048 n.1 (9th Cir. 2010)\\n(per curiam) (quoting Pierce v. County of Orange, 526 F.3d\\n1190, 1205 (9th Cir. 2008)). The PLRA further provides that\\n\\f42 EDMO V. CORIZON\\n\\nany “[p]reliminary injunctive relief shall automatically\\nexpire on the date that is 90 days after its entry, unless the\\ncourt makes the findings required under subsection (a)(1)\\n[quoted above] for the entry of prospective relief and makes\\nthe order final before the expiration of the 90-day period.”\\n18 U.S.C. § 3626(a)(2).\\n\\n The State contends that the district court did not make\\nthe PLRA’s requisite need-narrowness-intrusiveness\\nfindings or make its order final within 90 days, causing the\\ninjunction to expire under 18 U.S.C. § 3626(a)(2).\\nGenerally, the expiration of an injunction challenged on\\nappeal moots the appeal. See Kitlutsisti v. ARCO Alaska,\\nInc., 782 F.2d 800, 801 (9th Cir. 1986); see also United\\nStates v. Sec’y, Fla. Dep’t of Corr., 778 F.3d 1223, 1228–29\\n(11th Cir. 2015). The State asserts separate, albeit\\noverlapping, contentions in their motion to dismiss this\\nappeal and in their briefing. We reject those arguments.\\n\\n A. Need-Narrowness-Intrusiveness Findings\\n\\n The State first contends that the district court did not\\nmake the PLRA’s need-narrowness-intrusiveness findings,\\ncausing the injunction to automatically expire and mooting\\nthis appeal. 11 As we have explained in prior decisions, the\\nPLRA “has not substantially changed the threshold findings\\nand standards required to justify an injunction.” Gomez v.\\nVernon, 255 F.3d 1118, 1129 (9th Cir. 2001). When\\n“determining the appropriateness of the relief ordered,”\\nappellate “courts must do what they have always done”:\\n\\n 11\\n We question whether the State’s need-narrowness-intrusiveness\\nchallenge, properly understood, implicates mootness. But because the\\nresult is the same, we accept the State’s framing for purposes of our\\nanalysis.\\n\\f EDMO V. CORIZON 43\\n\\n“consider the order as a whole.” Armstrong v.\\nSchwarzenegger, 622 F.3d 1058, 1070 (9th Cir. 2010).\\nDistrict courts must make need-narrowness-intrusiveness\\n“findings sufficient to allow a ‘clear understanding’ of the\\nruling,” but they need not “make such findings on a\\nparagraph by paragraph, or even sentence by sentence,\\nbasis.” Id. (quotation omitted). “What is important, and\\nwhat the PLRA requires, is a finding that the set of reforms\\nbeing ordered—the ‘relief’—corrects the violations of\\nprisoners’ rights with the minimal impact possible on\\ndefendants’ discretion over their policies and procedures.”\\nId.\\n\\n Here, the district court made the necessary need-\\nnarrowness-intrusiveness findings. At the start of its\\nDecember 13, 2018 order, the district court explained that\\nany injunction must meet the PLRA’s need-narrowness-\\nintrusiveness requirement. See Edmo, 358 F. Supp. 3d at\\n1122. The district court then explained how the relief being\\nordered, GCS, “corrects the violations of” Edmo’s rights.\\nSee Armstrong, 622 F.3d at 1071. Specifically, the district\\ncourt explained that GCS is medically necessary to alleviate\\nEdmo’s gender dysphoria and that the State’s denial of GCS\\namounts to deliberate indifference in violation of the Eighth\\nAmendment. See Edmo, 358 F. Supp. 3d at 1116–21, 1123–\\n27, 1129. The district court limited the relief ordered to have\\n“the minimal impact possible on [the State’s] discretion over\\ntheir policies and procedures.” See Armstrong, 622 F.3d at\\n1071. Specifically, the district court limited the relief to\\n“actions reasonably necessary” to provide GCS, cautioned\\nthat its conclusion is based on “the unique facts and\\ncircumstances presented” by Edmo, and noted that its\\n“decision is not intended, and should not be construed, as a\\ngeneral finding that all inmates suffering from gender\\ndysphoria are entitled to [GCS].” Edmo, 358 F. Supp. 3d at\\n\\f44 EDMO V. CORIZON\\n\\n1110, 1129. Finally, the district court rejected the notion that\\ninjunctive relief would have “any adverse impact on public\\nsafety or the operation of a criminal justice system.”\\n18 U.S.C. § 3626(a)(1)(A). It explained that the State had\\n“made no showing that an order requiring them to provide”\\nGCS to Edmo “causes them injury.” Edmo, 358 F. Supp. 3d\\nat 1128. The district court’s order, considered as a whole,\\nmade all the findings required by 18 U.S.C. § 3626(a)(1)(A)\\nand our precedent. See Armstrong, 622 F.3d at 1070.\\n\\n B. Finality\\n\\n The State next argues that the injunction has\\nautomatically expired under the PLRA because the district\\ncourt did not make its order “final” within 90 days of\\nentering injunctive relief. See 18 U.S.C. § 3626(a)(2); see\\nalso Sec’y, Fla. Dep’t of Corr., 778 F.3d at 1228–29\\n(holding that an appeal of a preliminary injunction was moot\\nbecause the district court “did not issue an order finalizing\\nits [preliminary-injunction] order,” and “[a]s a result, the\\npreliminary injunction expired by operation of law” 90 days\\nlater). The PLRA provision cited by the State applies to\\npreliminary injunctive relief, not permanent injunctive\\nrelief. See 18 U.S.C. § 3626(a)(2). The permanent\\ninjunction that the district court entered has not expired. See\\nEdmo, 358 F. Supp. 3d at 1122 n.1 (concluding that Edmo is\\n“entitled to relief” under the permanent injunction standard);\\nsee also Edmo, 2019 WL 2319527, at *2 (clarifying on\\nlimited remand that the district court granted Edmo a\\npermanent injunction). It remains in place, albeit stayed.\\n\\f EDMO V. CORIZON 45\\n\\n There is a live controversy on appeal. 12 We accordingly\\nDENY the State’s motion to dismiss and proceed to the\\nmerits of the appeal.\\n\\n III. Challenges to the District Court’s Grant of\\n Injunctive Relief\\n\\n An injunction is an “extraordinary remedy never\\nawarded as of right.” Winter v. Nat. Res. Def. Council, Inc.,\\n555 U.S. 7, 24 (2008). “To be entitled to a permanent\\ninjunction, a plaintiff must demonstrate: (1) actual success\\non the merits; (2) that it has suffered an irreparable injury;\\n(3) that remedies available at law are inadequate; (4) that the\\nbalance of hardships justify a remedy in equity; and (5) that\\nthe public interest would not be disserved by a permanent\\ninjunction.” 13 Indep. Training & Apprenticeship Program\\n\\n\\n\\n 12\\n Even construed as a preliminary injunction, the district court’s\\nDecember 13, 2018 order is not moot. On May 31, 2019, the district\\ncourt, incorporating its previous findings, renewed the injunction. See\\nEdmo, 2019 WL 2319527, at *2. Because the district court renewed the\\ninjunction, we can consider its merits. See Mayweathers v. Newland,\\n258 F.3d 930, 935–36 (9th Cir. 2001) (holding that district courts may\\nrenew preliminary injunctions under the PLRA while an appeal is\\npending, and considering the merits of the renewed injunction). And we\\nhave jurisdiction under 28 U.S.C. § 1292(a)(1) regardless of whether the\\ndistrict court’s order is considered a preliminary or permanent\\ninjunction. See Hendricks v. Bank of Am., N.A., 408 F.3d 1127, 1131\\n(9th Cir. 2005) (preliminary injunction); TransWorld Airlines, Inc. v.\\nAm. Coupon Exch., Inc., 913 F.2d 676, 680–81 (9th Cir. 1990)\\n(permanent injunction where the “district court retained jurisdiction to\\ndetermine damages” and to adjudicate a separate claim).\\n 13\\n We agree with the State that the injunction is mandatory, as\\nopposed to prohibitory, because it requires the State to act. Based on that\\ndistinction, the State argues that Edmo must satisfy a higher burden of\\n\\f46 EDMO V. CORIZON\\n\\nv. Cal. Dep’t of Indus. Relations, 730 F.3d 1024, 1032 (9th\\nCir. 2013) (citing eBay Inc. v. MercExch., L.L.C., 547 U.S.\\n388, 391 (2006)).\\n\\n We review for abuse of discretion the district court’s\\ndecision to grant a permanent injunction. Ariz. Dream Act\\nCoal. v. Brewer, 855 F.3d 957, 965 (9th Cir. 2017). We\\n\\n\\nproof to be entitled to injunctive relief, and that the district court failed\\nto hold Edmo to that burden. On that point, we disagree.\\n\\n The State errs by relying on cases that concern mandatory\\npreliminary injunctions. Because mandatory preliminary injunctions go\\n“well beyond simply maintaining the status quo [p]endente lite,” they are\\n“particularly disfavored” and “are not issued in doubtful cases.” Marlyn\\nNutraceuticals, Inc. v. Mucos Pharma GmbH & Co., 571 F.3d 873, 879\\n(9th Cir. 2009) (alteration in original) (quoting Anderson v. United\\nStates, 612 F.2d 1112, 1114–15 (9th Cir. 1980)). The calculus is\\ndifferent in the context of permanent injunctions. A plaintiff must show\\nactual success on the merits, see Amoco Prod. Co. v. Village of Gambell,\\n480 U.S. 531, 546 n.12 (1987), so there is no concern that a mandatory\\npermanent injunction will upset the status quo only for a later trial on the\\nmerits to show that the plaintiff was not entitled to equitable relief. As a\\nresult, a plaintiff need not show that “extreme or very serious damage\\nwill result,” as is required for mandatory preliminary injunctions.\\n\\n As we have explained, the district court granted Edmo injunctive\\nrelief under both the preliminary and permanent injunction standards.\\nSee Edmo, 358 F. Supp. 3d at 1122 n.1; see also Edmo, 2019 WL\\n2319527, at *2. Because the standard for granting permanent injunctive\\nrelief is higher (in that it requires actual success on the merits) and the\\nState contends in its opening brief that we should review the injunction\\nas a permanent injunction, we consider whether the district court erred\\nin granting Edmo permanent injunctive relief. But we would also affirm\\nunder the mandatory preliminary injunction standard, because the district\\ncourt correctly applied the proper standard for mandatory preliminary\\ninjunctive relief, and not the lower standard for prohibitory preliminary\\ninjunctions. See Edmo, 358 F. Supp. 3d at 1122, 1128.\\n\\f EDMO V. CORIZON 47\\n\\nreview “any determination underlying the grant of an\\ninjunction by the standard that applies to that\\ndetermination.” Ting v. AT&T, 319 F.3d 1126, 1134–35 (9th\\nCir. 2003). Accordingly, the district court’s factual findings\\non Edmo’s Eighth Amendment claim are reviewed for clear\\nerror. See Graves, 623 F.3d at 1048. Clear error exists if the\\nfinding is “illogical, implausible, or without support in\\ninferences that may be drawn from the facts in the record.”\\nLa Quinta Worldwide LLC v. Q.R.T.M., S.A. de C.V.,\\n762 F.3d 867, 879 (9th Cir. 2014) (quoting Herb Reed\\nEnters., LLC v. Florida Entm’t Mgmt., Inc., 736 F.3d 1239,\\n1247 (9th Cir. 2013)). We review de novo the district court’s\\n“conclusion that the facts . . . demonstrate an Eighth\\nAmendment violation.” Hallett v. Morgan, 296 F.3d 732,\\n744 (9th Cir. 2002).\\n\\n The State contends that the district court erred in\\ngranting an injunction because (1) Edmo’s Eighth\\nAmendment claim fails and (2) Edmo has not shown that she\\nwill suffer irreparable injury in the absence of an\\ninjunction. 14 We disagree. We hold, based on the district\\ncourt’s factual findings, that Edmo established her Eighth\\nAmendment claim and that she will suffer irreparable\\nharm—in the form of ongoing mental anguish and possible\\nphysical harm—if GCS is not provided.\\n\\n A. The Merits of Edmo’s Eighth Amendment Claim\\n\\n “[D]eliberate indifference to serious medical needs of\\nprisoners” violates the Eighth Amendment. Estelle,\\n429 U.S. at 104. Because “society takes from prisoners the\\nmeans to provide for their own needs,” Brown, 563 U.S.\\n\\n 14\\n Because the State does not contest the other injunction factors, we\\ndo not address them.\\n\\f48 EDMO V. CORIZON\\n\\nat 510, the government has an “obligation to provide medical\\ncare for those whom it is punishing by incarceration,”\\nEstelle, 429 U.S. at 103.\\n\\n To establish a claim of inadequate medical care, a\\nprisoner must first “show a ‘serious medical need’ by\\ndemonstrating that ‘failure to treat a prisoner’s condition\\ncould result in further significant injury or the ‘unnecessary\\nand wanton infliction of pain.’” Jett v. Penner, 439 F.3d\\n1091, 1096 (9th Cir. 2006) (quoting McGuckin v. Smith,\\n974 F.2d 1050, 1059 (9th Cir. 1991), overruled on other\\ngrounds by WMX Techs., Inc. v. Miller, 104 F.3d 1133 (9th\\nCir. 1997) (en banc)). Serious medical needs can relate to\\n“physical, dental and mental health.” Hoptowit v. Ray,\\n682 F.2d 1237, 1253 (9th Cir. 1982), abrogated on other\\ngrounds by Sandin v. Conner, 515 U.S. 472 (1995).\\n\\n The State does not dispute that Edmo’s gender dysphoria\\nis a sufficiently serious medical need to trigger the State’s\\nobligations under the Eighth Amendment. Nor could it.\\nGender dysphoria is a “serious . . . medical condition” that\\ncauses “clinically significant distress”—distress that impairs\\nor severely limits an individual’s ability to function in a\\nmeaningful way. DSM-5 at 453, 458. As Edmo testified,\\nher gender dysphoria causes her to feel “depressed,”\\n“disgusting,” “tormented,” and “hopeless,” and it has caused\\npast efforts and active thoughts of self-castration. As this\\nand many other courts have recognized, Edmo’s gender\\ndysphoria is a sufficiently serious medical need to implicate\\nthe Eighth Amendment. See Rosati v. Igbinoso, 791 F.3d\\n1037, 1039–40 (9th Cir. 2015); Kosilek, 774 F.3d at 86;\\nDe’lonta, 708 F.3d at 525; Battista v. Clarke, 645 F.3d 449,\\n452 (1st Cir. 2011); Allard v. Gomez, 9 F. App’x 793, 794\\n(9th Cir. 2001); White v. Farrier, 849 F.2d 322, 325 (8th Cir.\\n1988); Meriwether v. Faulkner, 821 F.2d 408, 412 (7th Cir.\\n\\f EDMO V. CORIZON 49\\n\\n1987) (and cases cited therein); Norsworthy, 87 F. Supp. 3d\\nat 1187; Konitzer v. Frank, 711 F. Supp. 2d 874, 905 (E.D.\\nWis. 2010).\\n\\n If, as here, a prisoner establishes a sufficiently serious\\nmedical need, that prisoner must then “show the [official’s]\\nresponse to the need was deliberately indifferent.” Jett,\\n439 F.3d at 1096. An inadvertent or negligent failure to\\nprovide adequate medical care is insufficient to establish a\\nclaim under the Eighth Amendment. Estelle, 429 U.S.\\nat 105–06; see also Farmer v. Brennan, 511 U.S. 825, 835\\n(1994) (“ordinary lack of due care” is insufficient to\\nestablish an Eighth Amendment claim). In other words,\\n“[m]edical malpractice does not become a constitutional\\nviolation merely because the victim is a prisoner.” Estelle,\\n429 U.S. at 106. To “show deliberate indifference, the\\nplaintiff must show that the course of treatment the [official]\\nchose was medically unacceptable under the circumstances\\nand that the [official] chose this course in conscious\\ndisregard of an excessive risk to the plaintiff’s health.”\\nHamby v. Hammond, 821 F.3d 1085, 1092 (9th Cir. 2016)\\n(quoting Snow v. McDaniel, 681 F.3d 978, 988 (9th Cir.\\n2012), overruled in part on other grounds by Peralta v.\\nDillard, 744 F.3d 1076 (9th Cir. 2014) (en banc)).\\n\\n 1. The Medical Necessity of GCS for Edmo\\n\\n The crux of the State’s appeal is that it provided adequate\\nand medically acceptable care to Edmo.\\n\\n Accepted standards of care and practice within the\\nmedical community are highly relevant in determining what\\ncare is medically acceptable and unacceptable. See Allard v.\\nBaldwin, 779 F.3d 768, 772 (8th Cir. 2015); Henderson v.\\nGhosh, 755 F.3d 559, 566 (7th Cir. 2014) (per curiam).\\nTypically, “[a] difference of opinion between a physician\\n\\f50 EDMO V. CORIZON\\n\\nand the prisoner—or between medical professionals—\\nconcerning what medical care is appropriate does not\\namount to deliberate indifference.” Snow, 681 F.3d at 987;\\nsee also Gibson, 920 F.3d at 220. But that is true only if the\\ndueling opinions are medically acceptable under the\\ncircumstances. See Toguchi v. Chung, 391 F.3d 1051, 1058\\n(9th Cir. 2004) (a mere “difference of medical opinion . . .\\n[is] insufficient, as a matter of law, to establish deliberate\\nindifference,” but not if the “chosen course of treatment ‘was\\nmedically unacceptable under the circumstances’”\\n(alterations in original) (quoting Jackson v. McIntosh,\\n90 F.3d 330, 332 (9th Cir. 1996))).\\n\\n “In deciding whether there has been deliberate\\nindifference to an inmate’s serious medical needs, we need\\nnot defer to the judgment of prison doctors or\\nadministrators.” Hunt v. Dental Dep’t, 865 F.2d 198, 200\\n(9th Cir. 1989). Nor does it suffice for “correctional\\nadministrators wishing to avoid treatment . . . simply to find\\na single practitioner willing to attest that some well-accepted\\ntreatment is not necessary.” Kosilek, 774 F.3d at 90 n.12. In\\nthe final analysis under the Eighth Amendment, we must\\ndetermine, considering the record, the judgments of prison\\nmedical officials, and the views of prudent professionals in\\nthe field, whether the treatment decision of responsible\\nprison authorities was medically acceptable.\\n\\n Reviewing the record and the district court’s extensive\\nfactual findings, we conclude that Edmo has established that\\nthe “course of treatment” chosen to alleviate her gender\\ndysphoria “was medically unacceptable under the\\ncircumstances.” Hamby, 821 F.3d at 1092 (quoting Snow,\\n681 F.3d at 988). This conclusion derives from the district\\ncourt’s factual findings, which are not “illogical,\\nimplausible, or without support in inferences that may be\\n\\f EDMO V. CORIZON 51\\n\\ndrawn from the facts in the record.” La Quinta Worldwide\\nLLC, 762 F.3d at 879 (quotation omitted).\\n\\n In particular, and as we will explain, this is not a case of\\ndueling experts, as the State paints it. The district court\\npermissibly credited the opinions of Edmo’s experts that\\nGCS is medically necessary to treat Edmo’s gender\\ndysphoria and that the State’s failure to provide that\\ntreatment is medically unacceptable. Edmo’s experts are\\nwell-qualified to render such opinions, and they logically\\nand persuasively explained the necessity of GCS and applied\\nthe WPATH Standards of Care—the undisputed starting\\npoint in determining the appropriate treatment for gender\\ndysphoric individuals. On the other side of the coin, the\\ndistrict court permissibly discredited the contrary opinions\\nof the State’s treating physician and medical experts. Those\\nindividuals lacked expertise and incredibly applied (or did\\nnot apply, in the case of the State’s treating physician) the\\nWPATH Standards of Care. In other words, the district court\\ndid not clearly err in making its credibility determinations,\\nso it is not our role to reevaluate them. The credited\\ntestimony establishes that GCS is medically necessary.\\n\\n a. Expert Testimony\\n\\n Turning first to the expert testimony offered, the district\\ncourt credited the testimony of Edmo’s experts that GCS is\\nmedically necessary to treat Edmo’s gender dysphoria and\\nthat the State’s failure to provide that treatment is medically\\nunacceptable. See Edmo, 358 F. Supp. 3d at 1120–21, 1125.\\nDr. Ettner and Dr. Gorton opined that GCS is medically\\nnecessary because Edmo’s current treatment has been\\ninadequate, as evidenced by her self-castration attempts.\\nThey also opined that if Edmo does not receive GCS, there\\nis little chance that her gender dysphoria will improve and\\nshe is at risk of committing self-surgery again, suicide, and\\n\\f52 EDMO V. CORIZON\\n\\nfurther emotional decompensation. On the other hand,\\nproviding GCS to Edmo would, in the opinions of Dr. Ettner\\nand Dr. Gorton, align Edmo’s genitalia with her gender\\nidentity, thereby eliminating the severe distress Edmo\\nexperiences from her male genitalia.\\n\\n In sharp contrast, the district court gave “virtually no\\nweight” to the opinions of the State’s experts. Edmo, 358 F.\\nSupp. 3d at 1126. Dr. Garvey and Dr. Andrade, who\\npurported to rely on the WPATH Standards of Care, opined\\nthat GCS is not medically necessary for Edmo.\\n\\n The district court did not err in crediting the testimony of\\nEdmo’s experts and discounting the testimony of the State’s\\nexperts. Dr. Ettner and Dr. Gorton are well-qualified to\\nopine on the medical necessity of GCS. Both have\\nsubstantial experience treating individuals with gender\\ndysphoria. Dr. Ettner has evaluated, diagnosed, and treated\\nbetween 2,500 and 3,000 individuals with gender dysphoria,\\nwhile Dr. Gorton has been the primary care physician for\\napproximately 400 patients with gender dysphoria. Both\\nhave substantial experience evaluating whether GCS is\\nmedically necessary for patients. Dr. Ettner has evaluated\\nhundreds of people for GCS, referring approximately 300\\nwhile refusing others, and Dr. Gorton routinely determines\\nthe appropriateness of GCS for patients. They also have\\nexperience providing follow-up care for patients who have\\nundergone GCS. And both have published peer-reviewed\\narticles concerning the treatment of gender dysphoria.\\n\\n The State’s experts, by contrast, have substantial\\nexperience providing health care in institutional settings, but\\nlack meaningful experience directly treating people with\\ngender dysphoria. Dr. Garvey directly treated a “couple of\\npatients” with gender dysphoria early in her career, while\\nDr. Andrade has never provided direct treatment for patients\\n\\f EDMO V. CORIZON 53\\n\\nwith gender dysphoria. Moreover, prior to evaluating Edmo,\\nneither had ever evaluated someone in person to determine\\nthe medical necessity of GCS. Relatedly, Dr. Garvey and\\nDr. Andrade have never provided follow-up care for a person\\nwho has received GCS. Indeed, Dr. Andrade did not even\\nqualify under IDOC policy to assess a person for GCS. And\\nneither Dr. Garvey nor Dr. Andrade has published a peer-\\nreviewed article concerning the treatment of gender\\ndysphoria.\\n\\n Neither Dr. Ettner nor Dr. Gorton have treated prisoners\\nwith gender dysphoria, nor are they Certified Correctional\\nHealthcare Professionals. But both serve on WPATH’s\\nInstitutionalized Persons Committee, which “looks at the\\ncare and the assessment of individuals who are incarcerated\\nand develops standards for treatment” of such individuals.\\nThey are thus familiar with medical treatment in prison\\nsettings. Moreover, Dr. Ettner has assessed approximately\\n30 incarcerated persons with gender dysphoria for GCS and\\nother medical care.\\n\\n More to the point, the more relevant experience for\\ndetermining the medical necessity of GCS is having treated\\nindividuals with gender dysphoria, having evaluated\\nindividuals for GCS, and having treated them post-\\noperatively. Such experience lends itself to fundamental\\nknowledge of whether GCS is necessary and the potential\\nrisks of providing or foregoing the surgery. Edmo’s experts\\nhave the requisite experience; the State’s experts do not. For\\nthat reason alone, the district court did not clearly err in\\ncrediting the opinions of Edmo’s experts over those of the\\nState. 15 See Caro v. Woodford, 280 F.3d 1247, 1253 (9th\\n\\n 15\\n The State contends that neither Dr. Ettner nor Dr. Gorton was\\nqualified to offer expert opinions as to the appropriate medical care for\\n\\f54 EDMO V. CORIZON\\n\\nCir. 2002) (explaining that we “must afford the District\\nCourt considerable deference in its determination that the\\nwitnesses were qualified to draw [their] conclusions”).\\n\\n Independent of the experts’ qualifications, the district\\ncourt did not err in crediting the opinions of Edmo’s experts\\nover those of the State because aspects of Dr. Garvey’s and\\nDr. Andrade’s opinions ran contrary to the established\\nstandards of care in the area of transgender health care—the\\nWPATH Standards of Care—which they purported to\\napply. 16 See Edmo, 358 F. Supp. 3d at 1125.\\n\\n\\nEdmo because neither is a psychiatrist. So far as we can discern, the\\nargument is that because a psychiatrist (Dr. Eliason) evaluated Edmo for\\nGCS, only other psychiatrists are qualified to opine as to the medical\\nnecessity of GCS and to contradict his assessment. See Oral Arg.\\nat 10:00–10:30. We reject that contention. Edmo’s experts, as\\nexplained, have significant experience evaluating patients for GCS—\\nprecisely what Dr. Eliason did. On the basis of their medical experience\\ntreating persons with gender dysphoria, they are well-qualified to render\\nan opinion on the medical necessity of GCS and whether failure to\\nprovide the surgery is medically acceptable. See Fed. R. Evid. 702.\\n 16\\n The State contends that the district court erred in requiring strict\\nadherence to the flexible WPATH Standards of Care and in concluding\\nthat any deviation from those standards is medically unacceptable. But\\nthe district court correctly recognized that the WPATH Standards of\\nCare are flexible, see Edmo, 358 F. Supp. 3d at 1111, and it appropriately\\nused them as a starting point to gauge the credibility of each expert’s\\ntestimony, see id. at 1125–26. Tellingly, each expert for Edmo and the\\nState likewise used the WPATH Standards of Care as a starting point.\\nAs the district court recognized: “There are no other competing,\\nevidence-based standards that are accepted by any nationally or\\ninternationally recognized medical professional groups.” Id. at 1125.\\nAnd as the State acknowledged at the evidentiary hearing, the “WPATH\\nstandards of care in the seventh edition do provide the best guidance”\\nand “are the best standards out there.” For these reasons, the WPATH\\nStandards of Care establish a useful starting point for analyzing the\\n\\f EDMO V. CORIZON 55\\n\\n For example, both Dr. Garvey and Dr. Andrade\\nexpressed the view that Edmo does not meet the sixth\\nWPATH criterion, “12 continuous months of living in a\\ngender role that is congruent with gender identity.” WPATH\\nSOC at 60. They pointed out that Edmo has not presented\\nas female outside of prison and urged that she needs real-life\\nexperiences in the community before undergoing GCS.\\n\\n These opinions run head-on into the WPATH Standards\\nof Care. The WPATH standards, which the NCCHC\\nendorses as the accepted standards for the treatment of\\ntransgender inmates, apply\\n\\n in their entirety . . . to all transsexual,\\n transgender, and gender nonconforming\\n people, irrespective of their housing\\n situation. People should not be discriminated\\n against in their access to appropriate health\\n care based on where they live, including\\n institutional environments such as prisons\\n .... Health care for transsexual,\\n transgender, and gender nonconforming\\n people living in an institutional environment\\n should mirror that which would be available\\n to them if they were living in a non-\\n institutional setting within the same\\n community.\\n\\n\\n\\ncredibility and weight to be given to each expert’s opinion and whether\\nthat opinion was consistent with established standards of care. The State\\ndoes not contest the district court’s finding that the WPATH Standards\\nof Care are the “internationally recognized guidelines for the treatment\\nof individuals with gender dysphoria.” Id. at 1111. They are the gold\\nstandard on this issue.\\n\\f56 EDMO V. CORIZON\\n\\n All elements of assessment and treatment as\\n described in the [Standards of Care] can be\\n provided to people living in institutions.\\n Access to these medically necessary\\n treatments should not be denied on the basis\\n of institutionalization or housing\\n arrangements.\\n\\nWPATH SOC at 67. Dr. Garvey and Dr. Andrade’s view—\\nthat GCS cannot be medically indicated for transgender\\ninmates who did not present in a gender-congruent manner\\nbefore incarceration—contradicts these accepted standards.\\nDr. Garvey and Dr. Andrade would deny GCS to a class of\\npeople because of their “institutionalization,” which the\\nWPATH Standards of Care explicitly disavow. They\\nprovide no persuasive explanation for their deviation. 17 And\\nnothing in the WPATH Standards of Care or the law\\nsupports excluding an entire class of gender dysphoric\\nindividuals from eligibility for GCS.\\n\\n Both Dr. Garvey and Dr. Andrade also relied on Edmo’s\\nfailure to attend psychotherapy sessions as an indication that\\nher mental health concerns are not well controlled. But\\npsychotherapy is not a precondition for surgery under the\\nWPATH Standards of Care. WPATH SOC at 28–29.\\n\\n We acknowledge that the WPATH Standards of Care are\\nflexible, and a simple deviation from those standards does\\nnot alone establish an Eighth Amendment claim. But the\\n 17\\n In concluding that Edmo does not meet the sixth WPATH\\ncriterion, Dr. Garvey expressed concern that there is a lack of evidence\\nregarding GCS in prison settings. That rationale acts as self-fulfilling\\nprophecy. If prisons and prison officials deny GCS to prisoners because\\nof a lack of data, the data will never be generated, and the cycle will\\ncontinue.\\n\\f EDMO V. CORIZON 57\\n\\nState’s experts purported to be applying those standards and\\nyet did so in a way that directly contradicted them. These\\nunsupported and unexplained deviations offer a further\\nreason why the district court did not clearly err in\\ndiscounting the testimony of the State’s experts. See Caro,\\n280 F.3d at 1253.\\n\\n Finally, the district court did not err in discrediting the\\nState’s experts because aspects of their opinions were\\nillogical and unpersuasive. For example, Dr. Garvey and\\nDr. Andrade expressed the view that Edmo does not meet the\\nfirst WPATH criterion—“persistent, well documented\\ngender dysphoria,” WPATH SOC at 60—because of a lack\\nof evidence from pre-incarceration records of Edmo\\npresenting as female. But both experts acknowledged that\\nEdmo has been diagnosed with and treated for gender\\ndysphoria since 2012—i.e., for six years as of the evidentiary\\nhearing. Neither Dr. Garvey nor Dr. Andrade questioned\\nEdmo’s diagnosis, and both agree that she currently suffers\\ngender dysphoria. There can be no doubt that Edmo has\\n“persistent, well documented gender dysphoria,” so their\\nopinion is inexplicable.\\n\\n Dr. Garvey’s and Dr. Andrade’s opinions on this point\\nalso ignore that individuals with gender dysphoria do not\\nalways experience symptoms early in life or throughout their\\nlife, or do not identify them as such. As Dr. Ettner testified,\\n“gender dysphoria intensifies with age.” And as with\\ntreatment for any other medical condition, treatment for\\ngender dysphoria must be based on a patient’s current\\nsituation.\\n\\n The opinions of Edmo’s experts are notably devoid of\\nthese flaws. Dr. Ettner and Dr. Gorton cogently and\\npersuasively explained why GCS is medically necessary for\\nEdmo and why Edmo meets the WPATH criteria for GCS.\\n\\f58 EDMO V. CORIZON\\n\\n For example, consistent with the WPATH Standards of\\nCare, Dr. Ettner explained that Edmo has lived for\\n“12 continuous months . . . in a gender role that is congruent\\nwith gender identity” (the sixth WPATH criterion) because\\nshe has lived “as a woman to the best of her ability in a male\\nprison.” In support of her opinion, Dr. Ettner cited Edmo’s\\n“appearance . . . , her disciplinary records, which indicated\\nthat she had attempted to wear her hair in a feminine\\nhairstyle and to wear makeup even though that was against\\nthe rules and she was – received some sort of disciplinary\\naction for that, and her – the way that she was receiving\\nfemale undergarments and had developed the stigma of\\nfemininity, the secondary sex characteristics, breast\\ndevelopment, et cetera.” Dr. Gorton similarly explained that\\nEdmo satisfies the sixth WPATH criterion because she has\\nlived for years in her “target gender role . . . despite an\\nenvironment that’s very hostile to that and some negative\\nconsequences that she has experienced because of that.”\\n\\n Moreover, both Dr. Ettner and Dr. Gorton offered\\nreasoned explanations tying Edmo’s self-castration attempts\\nto her severe gender dysphoria. Dr. Ettner explained that\\ndoctors regard “surgical self-treatment . . . as an intentional\\nattempt to remove the target organ that produces\\ntestosterone, which, in fact, is the cure for gender\\ndysphoria.” As Dr. Gorton elaborated, Edmo’s self-\\ncastration attempts demonstrate deficient treatment for\\n“severe genital-focused gender dysphoria.” He rejected the\\nnotion that Edmo’s depression and anxiety drove her self-\\ncastration attempts: “there [are] a lot of people with\\ndepression and anxiety who don’t remove their testicles.”\\n\\n In light of the experts’ backgrounds and experience, and\\nthe reasonableness, consistency, and persuasiveness of their\\nopinions, the district court did not err in crediting the\\n\\f EDMO V. CORIZON 59\\n\\nopinions of Edmo’s experts and giving little weight to those\\nof the State’s experts. The district court carefully examined\\nthe voluminous record, extensive testimony, and conflicting\\nexpert opinions in this case and set forth clear reasons,\\nsupported by the record, for relying on the testimony of\\nEdmo’s experts. See La Quinta Worldwide, 762 F.3d at 879\\n(a factual finding is clear error if it is “illogical, implausible,\\nor without support in inferences that may be drawn from the\\nfacts in the record”); Caro, 280 F.3d at 1253; Beech Aircraft\\nCorp. v. United States, 51 F.3d 834, 838 (9th Cir. 1995) (per\\ncuriam). The credited expert testimony established that GCS\\nis medically necessary to alleviate Edmo’s gender dysphoria.\\n\\n b. Dr. Eliason’s Assessment\\n\\n Turning from the expert testimony offered, the State\\ncontends that Edmo’s experts, at most, created a dispute of\\nprofessional judgment with Edmo’s treating psychiatrist, Dr.\\nEliason, who it urges reasonably concluded that GCS is\\ninappropriate for Edmo. If that is the case, the argument\\ngoes, then Edmo’s Eighth Amendment claim fails because\\nthe dispute is merely a “difference of opinion . . . between\\nmedical professionals” about “what medical care is\\nappropriate.” Snow, 681 F.3d at 987. The problem for the\\nState is that Dr. Eliason’s decision “was medically\\nunacceptable under the circumstances.” Toguchi, 391 F.3d\\nat 1058 (quoting Jackson, 90 F.3d at 332).\\n\\n In particular, as the district court found, Dr. Eliason did\\nnot follow accepted standards of care in the area of\\ntransgender health care. See Edmo, 358 F. Supp. 3d at 1126.\\nDr. Eliason explained in his notes that, in his view, GCS is\\nmedically necessary in three situations: “congenital\\nmalformation or ambiguous genitalia,” “severe and\\ndevastating dysphoria that is primarily due to genitals,” or\\n“some type of medical problem in which endogenous sexual\\n\\f60 EDMO V. CORIZON\\n\\nhormones were causing severe physiological damage.” The\\nconclusion of his notes—“[t]his inmate does not meet any of\\nthose [three] criteria”—suggests that he views those as the\\nonly three scenarios in which GCS would be medically\\nnecessary, an impression he did not dispel during his\\ntestimony. Those “criteria” (Dr. Eliason’s term), however,\\nbear little resemblance to the widely accepted, evidence-\\nbased criteria set out in the WPATH’s Standards of Care. As\\nDr. Eliason acknowledged, the NCCHC endorses the\\nWPATH Standards of Care as the accepted standards for the\\ntreatment of transgender prisoners. And as the district court\\nfound and the State does not contest, “[t]here are no other\\ncompeting, evidence-based standards that are accepted by\\nany nationally or internationally recognized medical\\nprofessional groups.” Id. at 1125. Dr. Eliason did not follow\\nthese standards in rendering his decision.\\n\\n The State challenges the district court’s finding that\\nDr. Eliason “did not apply the WPATH Criteria,” id. at 1126,\\non two grounds. First, citing Dr. Eliason’s testimony at the\\nevidentiary hearing, it urges that Dr. Eliason concluded that\\nGCS was not medically necessary for Edmo because Edmo’s\\nmental health issues were not well controlled (the fourth\\nWPATH criterion) and she had not consistently presented as\\nfemale outside of prison (the sixth).\\n\\n The district court’s rejection of this post hoc explanation\\nwas not clear error. Neither of the explanations offered by\\nDr. Eliason during the evidentiary hearing appears in\\nDr. Eliason’s notes. Nor did he give these reasons during his\\ndeposition. Their absence is conspicuous, given that\\nDr. Eliason took the time to indicate instances where, in his\\nopinion, GCS is appropriate and to explain that Edmo did\\nnot satisfy his “criteria.”\\n\\f EDMO V. CORIZON 61\\n\\n Second, the State highlights that Dr. Eliason’s notes\\nrecommend further “supportive counseling” for Edmo and\\nindicate that Edmo was up for parole. The State construes\\nthese notes as shorthand for the fourth and sixth WPATH\\ncriteria, respectively. The State’s proposed reading of\\nDr. Eliason’s notes is unreasonable. His notes are clear that\\nGCS is not needed because Edmo did not meet his three\\n“criteria,” and the district court was well within its\\nfactfinding discretion in rejecting the State’s strained\\nreading. We therefore conclude that the district court\\nreasonably found that Dr. Eliason “did not rely upon any\\nfinding that Ms. Edmo did not meet the WPATH criteria in\\nconcluding in his April 2016 assessment that she did not\\nmeet the criteria for gender confirmation surgery.” Id.\\nat 1120.\\n\\n Notably, neither Dr. Eliason nor the State has offered any\\nexplanation or support for Dr. Eliason’s “criteria.”\\nDr. Eliason testified that he could not recall where he came\\nup with them.\\n\\n Nor has Dr. Eliason or the State contended that\\nDr. Eliason’s criteria were a reasonable deviation or\\nmodification of the WPATH Standards of Care. In any\\nevent, we could not accept that argument. Dr. Eliason’s\\ncriteria—apparently invented out of whole cloth—are so far\\nafield from the WPATH standards that we cannot\\ncharacterize his decision as a flexible application of or\\ndeviation from those standards. Indeed, as Dr. Gorton\\nexplained, two of Dr. Eliason’s criteria are inapplicable to\\nthe care of transgender individuals. Dr. Eliason’s criterion\\nof “congenital malformation or ambiguous genitalia” “isn’t\\n. . . germane to transgender people.” His statement that GCS\\ncould be needed when “endogenous sexual hormones were\\ncausing severe physiological damage,” is, in Dr. Gorton’s\\n\\f62 EDMO V. CORIZON\\n\\nwords, “bizarre. I can’t think of a clinical circumstance\\nwhere . . . your hormones that your body produces are\\nattacking you . . . . I just don’t understand what [Dr. Eliason]\\nis talking about there.”\\n\\n Dr. Eliason, in short, did not follow the accepted\\nstandards of care in the area of transgender health care, nor\\ndid he reasonably deviate from or flexibly apply them.\\nDr. Eliason did not apply the established standards, even as\\na starting point, in his evaluation.\\n\\n Putting to the side Dr. Eliason’s failure to follow or\\nreasonably deviate from the accepted standards of care, his\\ndecision was internally contradictory in an important way.\\nHis notes reflect that GCS would be medically necessary if\\na person is suffering “severe and devastating gender\\ndysphoria that is primarily due to genitals.” At his\\ndeposition, Dr. Eliason conceded that self-castration could\\nshow gender dysphoria sufficiently severe to satisfy that\\ncriterion. And at the evidentiary hearing, he acknowledged\\nthat Edmo “does primarily meet that criteri[on].” Thus, even\\nunder Dr. Eliason’s own criteria, Edmo should have been\\nprovided GCS. Neither Dr. Eliason nor the State has\\nreconciled this important contradiction between\\nDr. Eliason’s criteria and his determination.\\n\\n In sum, Dr. Eliason’s evaluation was not an exercise of\\nmedically acceptable professional judgment. Dr. Eliason’s\\ndecision was based on inexplicable criteria far afield from\\nthe recognized standards of care and, even applying\\nDr. Eliason’s criteria, Edmo qualifies for GCS. Given the\\ncredited expert testimony that GCS is necessary to treat\\nEdmo’s gender dysphoria, Dr. Eliason’s contrary\\n\\f EDMO V. CORIZON 63\\n\\ndetermination was “medically unacceptable under the\\ncircumstances.” 18 Snow, 681 F.3d at 988.\\n\\n 2. Deliberate Indifference\\n\\n The State next contends that even if the treatment\\nprovided Edmo was medically unacceptable, no defendant\\nacted “in conscious disregard of an excessive risk to\\n[Edmo’s] health.” Hamby, 821 F.3d at 1092 (quoting Snow,\\n681 F.3d at 988). We disagree.\\n\\n The record demonstrates that Dr. Eliason acted with\\ndeliberate indifference to Edmo’s serious medical needs.\\nDr. Eliason knew, as of the time of his evaluation, that Edmo\\nhad attempted to castrate herself. He also knew that Edmo\\nsuffers from gender dysphoria; he knew she experiences\\n“clinically significant” distress that impairs her ability to\\nfunction. He acknowledged that Edmo’s self-castration\\nattempt was evidence that Edmo’s gender dysphoria, in his\\nwords, “had risen to another level.” Dr. Eliason nonetheless\\ncontinued with Edmo’s ineffective treatment plan.\\n\\n Edmo then tried to castrate herself a second time, in\\nDecember 2016. Dr. Eliason knew of that nearly\\n\\n 18\\n Dr. Eliason was not alone in his decision. Dr. Stoddart,\\nDr. Young, and Jeremy Clark agreed with his assessment, as did the\\nMTC. The State contends that such general agreement demonstrates that\\nDr. Eliason’s decision was reasonable. But general agreement in a\\nmedically unacceptable form of treatment does not somehow make it\\nreasonable. This is especially so in light of the limited review those\\nindividuals performed: Dr. Stoddard, Dr. Young, and Jeremy Clark\\nagreed with Dr. Eliason’s recommended treatment as he presented it to\\nthem and without personally evaluating Edmo, and the MTC “does not\\nmake any individual treatment decisions regarding [gender dysphoric]\\ninmates. Those determinations are made by the individual clinicians or\\nthe medical staff employed by Corizon,” like Dr. Eliason.\\n\\f64 EDMO V. CORIZON\\n\\ncatastrophic event, but he did not reevaluate or recommend\\na change to Edmo’s treatment plan, despite indicating in his\\nApril 2016 evaluation that he would continue to monitor and\\nassess Edmo’s condition. Dr. Eliason continued to see Edmo\\nafter that time, and he considered Edmo’s treatment as a\\nmember of the MTC. At no point did Dr. Eliason change his\\nmind or the treatment plan regarding surgery. Under these\\ncircumstances, we conclude that Dr. Eliason knew of and\\ndisregarded the substantial risk of severe harm to Edmo.\\nFarmer, 511 U.S. at 837.\\n\\n The State urges that neither Dr. Eliason nor any other\\ndefendant acted with deliberate indifference because none\\nacted with “malice, intent to inflict pain, or knowledge that\\n[the] recommended course of treatment was medically\\ninappropriate.” The State misstates the standard. A prisoner\\n“must show that prison officials ‘kn[e]w [ ] of and\\ndisregard[ed]’ the substantial risk of harm,’ but the officials\\nneed not have intended any harm to befall the inmate; ‘it is\\nenough that the official acted or failed to act despite his\\nknowledge of a substantial risk of serious harm.’” Lemire v.\\nCal. Dep’t of Corr. & Rehab., 726 F.3d 1062, 1074 (9th Cir.\\n2013) (alterations in original) (quoting Farmer, 511 U.S.\\nat 837, 842). Neither the Supreme Court nor this court has\\never required a plaintiff to show a “sinister [prison official]\\nwith improper motives,” as the State would require. It is\\nenough that Dr. Eliason knew of and disregarded an\\nexcessive risk to Edmo’s health by rejecting her request for\\nGCS and then never re-evaluating his decision despite\\nongoing harm to Edmo.\\n\\n The State also contends that because the defendants\\nprovided some care to Edmo, no defendant could have been\\ndeliberately indifferent. The provision of some medical\\ntreatment, even extensive treatment over a period of years,\\n\\f EDMO V. CORIZON 65\\n\\ndoes not immunize officials from the Eighth Amendment’s\\nrequirements. See Lopez v. Smith, 203 F.3d 1122, 1132 (9th\\nCir. 2000) (en banc) (explaining that “[a] prisoner need not\\nprove that he was completely denied medical care” to make\\nout an Eighth Amendment claim); see also De’lonta,\\n708 F.3d at 526 (“[J]ust because [officials] have provided\\nDe’lonta with some treatment consistent with the GID\\nStandards of Care, it does not follow that they have\\nnecessarily provided her with constitutionally adequate\\ntreatment.”). As the Fourth Circuit has aptly analogized,\\n\\n imagine that prison officials prescribe a\\n painkiller to an inmate who has suffered a\\n serious injury from a fall, but that the\\n inmate’s symptoms, despite the medication,\\n persist to the point that he now, by all\\n objective measure, requires evaluation for\\n surgery. Would prison officials then be free\\n to deny him consideration for surgery,\\n immunized from constitutional suit by the\\n fact they were giving him a painkiller? We\\n think not.\\n\\nDe’lonta, 708 F.3d at 526. Here, although the treatment\\nprovided Edmo was important, it stopped short of what was\\nmedically necessary.\\n\\n 3. Out-of-Circuit Precedent\\n\\n Our decision cleaves to settled Eighth Amendment\\njurisprudence, which requires a fact-specific analysis of the\\nrecord (as construed by the district court) in each case. See\\nPatel v. Kent Sch. Dist., 648 F.3d 965, 975 (9th Cir. 2011)\\n(“Deliberate-indifference cases are by their nature highly\\nfact-specific . . . .”); see also Rachel v. Troutt, 820 F.3d 390,\\n394 (10th Cir. 2016) (“Each step of this [deliberate\\n\\f66 EDMO V. CORIZON\\n\\nindifference] inquiry is fact-intensive.” (quoting Hartsfield\\nv. Colburn, 491 F.3d 394, 397 (8th Cir. 2007))); Roe v.\\nElyea, 631 F.3d 843, 859 (7th Cir. 2011) (“[I]nmate medical\\ncare decisions must be fact-based with respect to the\\nparticular inmate, the severity and stage of his condition, the\\nlikelihood and imminence of further harm and the efficacy\\nof available treatments.”); Youmans v. Gagnon, 626 F.3d\\n557, 564 (11th Cir. 2010) (“Judicial decisions addressing\\ndeliberate indifference to a serious medical need, like\\ndecisions in the Fourth Amendment search-and-seizure\\nrealm, are very fact specific.”); Chance v. Armstrong,\\n143 F.3d 698, 703 (2d Cir. 1998) (“Whether a course of\\ntreatment was the product of sound medical judgment,\\nnegligence, or deliberate indifference depends on the facts\\nof the case.”).\\n\\n Several years ago, the First Circuit, sitting en banc,\\nemployed that fact-based approach to evaluate a gender\\ndysphoric prisoner’s Eighth Amendment claim seeking\\nGCS. The First Circuit confronted the following record:\\ncredited expert testimony disagreed as to whether GCS was\\nmedically necessary; the prisoner’s active treatment plan,\\nwhich did not include GCS, had “led to a significant\\nstabilization in her mental state”; and a report and testimony\\nfrom correctional officials detailed significant security\\nconcerns that would arise if the prisoner underwent GCS.\\nKosilek, 774 F.3d at 86–96. “After carefully considering the\\ncommunity standard of medical care, the adequacy of the\\nprovided treatment, and the valid security concerns\\narticulated by the DOC,” a 3–2 majority of the en banc court\\nconcluded that the plaintiff had not demonstrated GCS was\\nmedically necessary treatment for her gender dysphoria. Id.\\nat 68.\\n\\f EDMO V. CORIZON 67\\n\\n Our approach mirrors the First Circuit’s, but the\\nimportant factual differences between cases yield different\\noutcomes. Notably, the security concerns in Kosilek, which\\nthe First Circuit afforded “wide-ranging deference,” are\\ncompletely absent here. Id. at 92. The State does not so\\nmuch as allude to them. The medical evidence also differs.\\nIn Kosilek, qualified and credited experts disagreed about\\nwhether GCS was necessary. Id. at 90. As explained above,\\nthe district court’s careful factual findings admit of no such\\ndisagreement here. Rather, they unequivocally establish that\\nGCS is the safe, effective, and medically necessary treatment\\nfor Edmo’s severe gender dysphoria.\\n\\n We recognize, however, that our decision is in tension\\nwith Gibson v. Collier. In that case, the Fifth Circuit held,\\nin a split decision, that “[a] state does not inflict cruel and\\nunusual punishment by declining to provide [GCS] to a\\ntransgender inmate.” 920 F.3d at 215. It did so on a “sparse\\nrecord”—which included only the WPATH Standards of\\nCare and was notably devoid of “witness testimony or\\nevidence from professionals in the field”—compiled by a\\npro se plaintiff. Id. at 220. Despite the sparse record, a 2–1\\nmajority of the Gibson panel concluded that “there is no\\nconsensus in the medical community about the necessity and\\nefficacy of [GCS] as a treatment for gender dysphoria. . . .\\nThis on-going medical debate dooms Gibson’s claim.” Id.\\nat 221.\\n\\n We respectfully disagree with the categorical nature of\\nour sister circuit’s holding. Most fundamentally, Gibson\\nrelies on an incorrect, or at best outdated, premise: that\\n“[t]here is no medical consensus that [GCS] is a necessary\\nor even effective treatment for gender dysphoria.” Id. at 223.\\n\\n As the record here demonstrates and the State does not\\nseriously dispute, the medical consensus is that GCS is\\n\\f68 EDMO V. CORIZON\\n\\neffective and medically necessary in appropriate\\ncircumstances. The WPATH Standards of Care—which are\\nendorsed by the American Medical Association, the\\nAmerican Medical Student Association, the American\\nPsychiatric Association, the American Psychological\\nAssociation, the American Family Practice Association, the\\nEndocrine Society, the National Association of Social\\nWorkers, the American Academy of Plastic Surgeons, the\\nAmerican College of Surgeons, Health Professionals\\nAdvancing LGBTQ Equality, the HIV Medicine\\nAssociation, the Lesbian, Bisexual, Gay and Transgender\\nPhysician Assistant Caucus, and Mental Health America—\\nrecognize this fact. WPATH SOC at 54–55. Each expert in\\nthis case agrees. As do others in the medical community.\\nSee, e.g., U.S. Dep’t of Health & Human Servs., No. A-13-\\n87, Decision No. 2576; Bao Ngoc N. Tran, et al., Gender\\nAffirmation Surgery: A Synopsis Using American College of\\nSurgeons National Surgery Quality Improvement Program\\nand National Inpatient Sample Databases, 80 Annals Plastic\\nSurgery S229, S234 (2018); Frey, A Historical Review of\\nGender-Affirming Medicine, 14 J. Sexual Med. at 991; see\\nalso What We Know Project, Ctr. for the Study of\\nInequality, Cornell Univ., What Does the Scholarly\\nResearch Say About the Effect of Gender Transition on\\nTransgender Well-Being?, https://whatweknow.inequality.c\\nornell.edu/topics/lgbt-equality/what-does-the-scholarly-rese\\narch-say-about-the-well-being-of-transgender-people/ (last\\nvisited July 10, 2019) (reviewing the available literature and\\nfinding “a robust international consensus in the peer-\\nreviewed literature that gender transition, including medical\\ntreatments such as hormone therapy and surgeries, improves\\nthe overall well-being of transgender individuals”). The\\nFifth Circuit is the outlier.\\n\\f EDMO V. CORIZON 69\\n\\n Gibson’s broad holding stemmed from a dismaying\\ndisregard for procedure. As noted, the “sparse” summary\\njudgment record that the pro se plaintiff developed included\\n“only the WPATH Standards of Care.” Gibson, 920 F.3d at\\n221. Perhaps that factual deficiency doomed Gibson’s\\nEighth Amendment claim. See id. at 223–24. But to reach\\nits broader holding that denying GCS cannot, as a matter of\\nlaw, violate the Eighth Amendment—in other words, to\\nreject every conceivable Eighth Amendment claim based on\\nthe denial of GCS—the Fifth Circuit coopted the record from\\nKosilek, a First Circuit decision that predates Gibson by four\\nyears. Id. at 221–23. We doubt the analytical value of such\\nan anomalous procedural approach.\\n\\n Worse yet, the medical opinions from Kosilek do not\\nsupport the Fifth Circuit’s categorical holding. Dr. Chester\\nSchmidt’s and Dr. Stephen Levine’s testimony in Kosilek,\\nwhich the Fifth Circuit relied on, do not support the\\nproposition that GCS is never medically necessary. Dr.\\nSchmidt and Dr. Levine testified that GCS was not necessary\\nin the factual circumstances of that case, that is, based on the\\nunique medical needs of the prisoner at issue. See Kosilek,\\n774 F.3d at 76–79.\\n\\n The only suggestion in Kosilek that GCS is never\\nmedically necessary is in the First Circuit’s recitation of the\\ntestimony of Dr. Cynthia Osborne. See Gibson, 920 F.3d\\nat 221. The First Circuit recounted that Dr. Osborne testified\\nthat she “did not view [GCS] as medically necessary in light\\nof the ‘whole continuum from noninvasive to invasive’\\ntreatment options available to individuals with” gender\\ndysphoria. Kosilek, 774 F.3d at 77. To the extent this vague\\nportrait of Dr. Osborne’s testimony conveys her belief that\\nGCS is never medically necessary, she has apparently\\nchanged her view in the more than ten years since she\\n\\f70 EDMO V. CORIZON\\n\\ntestified in Kosilek. Like both sides and all four medical\\nexperts who testified here, Dr. Osborne now agrees that GCS\\n“can be medically necessary for some, though not all,\\npersons with [gender dysphoria], including some prison\\ninmates.” Osborne & Lawrence, Male Prison Inmates With\\nGender Dysphoria, 45 Archives of Sexual Behav. at 1651.\\nIn her and her co-author’s words, “[GCS] is a safe, effective,\\nand widely accepted treatment for [gender dysphoria];\\ndisputing the medical necessity of [GCS] based on assertions\\nto the contrary is unsupportable.” Id. The predicate medical\\nopinions that Gibson is premised upon, then, do not support\\nthe Fifth Circuit’s view that GCS is never medically\\nnecessary. The consensus is that GCS is effective and\\nmedically necessary in appropriate circumstances. 19\\n\\n Gibson is unpersuasive for several additional reasons. It\\ndirectly conflicts with decisions of this circuit, the Fourth\\n\\n 19\\n We do not suggest that every member of the medical and mental\\nhealth communities agrees that GCS may be medically necessary. There\\nare outliers. But when the medical consensus is that a treatment is\\neffective and medically necessary under the circumstances, prison\\nofficials render unacceptable care by following the views of outliers\\nwithout offering a credible medical basis for deviating from the accepted\\nview. See Kosilek, 774 F.3d at 90 n.12 (explaining that it is not enough\\nfor “correctional administrators wishing to avoid treatment . . . simply to\\nfind a single practitioner willing to attest that some well-accepted\\ntreatment is not necessary”); Hamilton v. Endell, 981 F.2d 1062, 1067\\n(9th Cir. 1992) (“By choosing to rely upon a medical opinion which a\\nreasonable person would likely determine to be inferior, the prison\\nofficials took actions which may have amounted to the denial of medical\\ntreatment, and the unnecessary and wanton infliction of pain.” (quotation\\nomitted)), overruled in part on other grounds as recognized in Snow,\\n681 F.3d at 986; cf. also Bragdon v. Abbott, 524 U.S. 624, 650 (1998)\\n(“A health care professional who disagrees with the prevailing medical\\nconsensus may refute it by citing a credible scientific basis for deviating\\nfrom the accepted norm.”).\\n\\f EDMO V. CORIZON 71\\n\\nCircuit, and the Seventh Circuit, all of which have held that\\ndenying surgical treatment for gender dysphoria can pose a\\ncognizable Eighth Amendment claim. Rosati, 791 F.3d at\\n1040 (alleged blanket ban on GCS and denial of GCS to\\nplaintiff with severe symptoms, including repeated self-\\ncastration attempts, states an Eighth Amendment claim);\\nFields v. Smith, 653 F.3d 550, 552–53, 558–59 (7th Cir.\\n2011) (law banning hormone treatment and GCS, even if\\nmedically necessary, violates the Eighth Amendment);\\nDe’lonta, 708 F.3d at 525 (alleged denial of an evaluation\\nfor GCS states an Eighth Amendment claim). 20 Relatedly,\\nGibson eschews Eighth Amendment precedent requiring a\\ncase-by-case determination of the medical necessity of a\\nparticular treatment. See, e.g., Colwell v. Bannister,\\n763 F.3d 1060, 1068 (9th Cir. 2014) (holding that the\\n“blanket, categorical denial of medically indicated surgery\\nsolely on the basis of an administrative policy . . . is the\\nparadigm of deliberate indifference” (quotation omitted));\\nRoe, 631 F.3d at 859.\\n\\n In this latter respect, Gibson also contradicts and\\nmisconstrues the precedent it purports to follow: Kosilek.\\nAccording to the Gibson majority, “the majority in Kosilek\\neffectively allowed a blanket ban on sex reassignment\\nsurgery.” 920 F.3d at 216. Not so. The First Circuit did\\nprecisely what we do here: assess whether the record before\\nit demonstrated deliberate indifference to the plaintiff’s\\n\\n 20\\n The Fifth Circuit unpersuasively attempted to reconcile its\\ndecision with Rosati and De’lonta, pointing out that those decisions\\n“allowed Eighth Amendment claims for [GCS] to survive motions to\\ndismiss, without addressing the merits.” Gibson, 920 F.3d at 223 n.8.\\nBut if Gibson is correct that failing to provide GCS cannot amount to\\ndeliberate indifference, then a plaintiff cannot state an Eighth\\nAmendment claim based on the denial of GCS. Rosati and De’lonta\\nwould necessarily have been decided differently under Gibson’s holding.\\n\\f72 EDMO V. CORIZON\\n\\ngender dysphoria. On the record before it, the First Circuit\\ndetermined that either of two courses of treatment (one\\nincluded GCS and one did not) were medically acceptable.\\nKosilek, 774 F.3d at 90. In light of those medically\\nacceptable alternatives, the First Circuit explained that it was\\nnot its place to “second guess medical judgments or to\\nrequire that the DOC adopt the more compassionate of two\\nadequate options.” Id. (quotation omitted). It expressly\\ncautioned that the opinion should not be read to “create a de\\nfacto ban against [GCS] as a medical treatment for any\\nincarcerated individual,” as “any such policy would conflict\\nwith the requirement that medical care be individualized\\nbased on a particular prisoner’s serious medical needs.” Id.\\nat 91 (citing Roe, 631 F.3d at 862–63). The Fifth Circuit\\ndisregarded these words of warning. 21\\n\\n ***\\n\\n In summary, Edmo has established that she suffers from\\na “serious medical need,” Jett, 439 F.3d at 1096, and that the\\ntreatment provided was “medically unacceptable under the\\ncircumstances” and chosen “in conscious disregard of an\\nexcessive risk” to her health, Hamby, 821 F.3d at 1092. She\\nestablished her Eighth Amendment claim of deliberate\\nindifference as to Defendant-Appellant Dr. Eliason.\\n\\n\\n\\n 21\\n Gibson’s final, originalist rationale—that it cannot be cruel and\\nunusual to deny a surgery that has only once been provided to an inmate,\\n920 F.3d at 226–28—warrants little discussion. Gibson’s originalist\\nunderstanding of the Eighth Amendment does not control; Estelle does,\\nand under Estelle a plaintiff establishes an Eighth Amendment claim by\\ndemonstrating that prison officials were deliberately indifferent to a\\nserious medical need. 429 U.S. at 106. This standard protects the\\nevolving standards of decency enshrined in the Eighth Amendment.\\n\\f EDMO V. CORIZON 73\\n\\n B. Irreparable Harm\\n\\n The State next contends that the district court erred in\\nfinding that Edmo would be irreparably harmed absent an\\ninjunction.\\n\\n In reaching its conclusion, the district court found that\\nEdmo experiences ongoing “clinically significant distress,”\\nmeaning “the distress impairs or severely limits [her] ability\\nto function in a meaningful way.” Edmo, 358 F. Supp. 3d\\nat 1110–11. This finding is supported by Edmo’s testimony\\nthat her gender dysphoria causes her to feel “depressed,”\\n“disgusting,” “tormented,” and “hopeless”; that she actively\\nexperiences thoughts of self-castration; and that she “self-\\nmedicat[es]” by cutting her arms with a razor to avoid acting\\non those thoughts and impulses. The district court also\\nfound that in the absence of surgery, Edmo “will suffer\\nserious psychological harm and will be at high risk of self-\\ncastration and suicide.” Id. at 1128. This finding is\\nsupported by the credited expert testimony of Dr. Ettner and\\nDr. Gorton, who detailed the escalating risks of self-surgery,\\nsuicide, and emotional decompensation should Edmo be\\ndenied surgery.\\n\\n It is no leap to conclude that Edmo’s severe, ongoing\\npsychological distress and the high risk of self-castration and\\nsuicide she faces absent surgery constitute irreparable harm.\\nSee Stanley v. Univ. of S. Cal., 13 F.3d 1313, 1324 n.5 (9th\\nCir. 1994); Thomas v. County of Los Angeles, 978 F.2d 504,\\n511 (9th Cir. 1992); Chalk v. U.S. Dist. Ct. Cent. Dist. of\\nCal., 840 F.2d 701, 709 (9th Cir. 1988). Moreover, the\\ndeprivation of Edmo’s constitutional right to adequate\\nmedical care is sufficient to establish irreparable harm. See\\nNelson v. NASA, 530 F.3d 865, 882 (9th Cir. 2008) (“Unlike\\nmonetary injuries, constitutional violations cannot be\\nadequately remedied through damages and therefore\\n\\f74 EDMO V. CORIZON\\n\\ngenerally constitute irreparable harm.”), rev’d and\\nremanded on other grounds, 562 U.S. 134 (2011).\\n\\n The State offers three contentions as to why the district\\ncourt erred in finding that Edmo would be irreparably\\ninjured in the absence of an injunction. None is persuasive.\\n\\n First, the State argues that the “long delay” of “nearly a\\nyear” between Edmo filing her Amended Complaint and her\\npreliminary injunction motion “implies a lack of urgency\\nand irreparable harm.” We disagree. The procedural history\\ndemonstrates that Edmo did not sit on her rights. Proceeding\\npro se, Edmo moved for preliminary injunctive relief when\\nshe filed her original complaint. The court then appointed\\ncounsel for Edmo, and shortly after appearing, appointed\\ncounsel withdrew Edmo’s motion and filed an amended\\ncomplaint. To assess the urgency of surgery, Edmo’s\\ncounsel promptly sought access to Edmo’s medical records,\\nwhich the State did not produce until more than six months\\nlater. Edmo moved for injunctive relief shortly thereafter.\\nDuring that time, Edmo and her counsel diligently\\ninvestigated and compiled the necessary record to move for\\ninjunctive relief. That it took them months to do their\\ndiligence does not suggest that Edmo will not be harmed\\nabsent an injunction.\\n\\n Second, the State contends that Edmo has not established\\nirreparable injury because both she and her expert,\\nDr. Gorton, agree that GCS is not an emergency surgery and\\nthat the State should have six months to provide such\\nsurgery. The State’s argument would preclude courts from\\nordering non-emergent medical care, even if the Eighth\\nAmendment demands it. That is untenable. The State also\\nignores the rationale for the six-month time period. As Dr.\\nGorton explained, all patients who receive GCS “are seen,\\nthey are evaluated, there is a process you have to go\\n\\f EDMO V. CORIZON 75\\n\\nthrough.” In his experience, that process typically concludes\\nwithin six months. That Edmo requested relief on a\\nreasonable timeline, based on the medical evidence, does not\\nundermine the strong evidence of irreparable injury.\\n\\n Third, the State contends that Edmo has not established\\nirreparable harm because she “has not attempted suicide or\\nself-castration for years.” That argument overlooks the\\nprofound, persistent distress Edmo’s gender dysphoria\\ncauses, as well as the credited expert testimony that absent\\nGCS, Edmo is at risk of further attempts at self-castration,\\nand possibly suicide. The district court did not err in finding\\nthat Edmo would be irreparably harmed in the absence of an\\ninjunction.\\n\\n IV. Challenges to the Scope of the Injunction\\n\\n We turn to the State’s contentions that the district court’s\\ninjunction was overbroad.\\n\\n A. Individual Defendants\\n\\n The State contends that the injunction should not apply\\nto Atencio, Zmuda, Yordy, Siegert, Dr. Young, Dr. Craig,\\nDr. Eliason, or Dr. Whinnery because the district court did\\nnot find that they, individually, were deliberately indifferent\\nto Edmo’s medical needs.\\n\\n As explained in Section III.A, Edmo has established that\\nDr. Eliason was deliberately indifferent to her serious\\nmedical needs. The injunction was properly entered against\\nhim because he personally participated in the deprivation of\\nEdmo’s constitutional rights. See Colwell, 763 F.3d at 1070.\\n\\n Edmo sued Attencio, Zmuda, and Yordy in their official\\ncapacities. An official-capacity suit for injunctive relief is\\n\\f76 EDMO V. CORIZON\\n\\nproperly brought against any persons who “would be\\nresponsible for implementing any injunctive relief.” Pouncil\\nv. Tilton, 704 F.3d 568, 576 (9th Cir. 2012). The State does\\nnot contest that Attencio, as Director of IDOC, and Zmuda,\\nas Deputy Director of IDOC, would be responsible for\\nimplementing any injunctive relief ordered. Edmo properly\\nnamed them as defendants to her Eighth Amendment claim\\nfor injunctive relief, regardless of their personal\\ninvolvement. See Colwell, 763 F.3d at 1070–71 (director of\\na state correctional system is a proper defendant in an\\nofficial-capacity suit seeking injunctive relief for Eighth\\nAmendment violations). Yordy is no longer the Warden of\\nISCI, but, by operation of the Federal Rules, his successor,\\nAl Ramirez, is “automatically substituted as party” in his\\nofficial capacity. Fed. R. Civ. P. 25(d). Ramirez is properly\\na defendant to Edmo’s Eighth Amendment claim for\\ninjunctive relief, regardless of his personal involvement. See\\nColwell, 763 F.3d at 1070–71 (warden is a proper defendant\\nin an official-capacity suit seeking injunctive relief for\\nEighth Amendment violations). Because Edmo may\\nproperly pursue her Eighth Amendment claim for injunctive\\nrelief against Attencio, Zmuda, and Ramirez in their official\\ncapacities, they are properly included within the scope of the\\ndistrict court’s injunction. On remand, the district court shall\\namend the injunction to substitute Al Ramirez (or the then-\\ncurrent Warden of ISCI) as a party for Yordy.\\n\\n Edmo also named Yordy as a defendant in his individual\\ncapacity. She likewise named Siegert, Dr. Young, Dr. Craig,\\nand Dr. Whinnery as defendants in their individual\\ncapacities (though she does not argue on appeal that the\\ninjunction properly included them). We hold that the\\nevidence in the current record is insufficient to conclude that\\nthey were deliberately indifferent to Edmo’s serious medical\\nneeds. In particular, the record does not show what they\\n\\f EDMO V. CORIZON 77\\n\\nknew about Edmo’s condition and what role they played in\\nher treatment or lack thereof. Edmo has not established their\\nliability, and the district court improperly included them\\nwithin the scope of the injunction. We vacate the district\\ncourt’s injunction to the extent it applies to Yordy, Siegert,\\nDr. Young, Dr. Craig, and Dr. Whinnery in their individual\\ncapacities. See California v. Azar, 911 F.3d 558, 585 (9th\\nCir. 2018) (vacating in part an overbroad injunction and\\nremanding to the district court). On remand, the district\\ncourt shall modify the injunction to exclude those defendants\\nfrom its scope.\\n\\n B. Corizon\\n\\n The State also contends that the injunction should not\\napply to Corizon. It urges that Corizon does not have a\\npolicy barring GCS and argues that such a policy is a\\nprerequisite to liability under Monell v. Department of Social\\nServices, 436 U.S. 658 (1978). We have not yet determined\\nwhether Monell applies “to private entities acting on behalf\\nof state governments,” such as Corizon. Oyenik v. Corizon\\nHealth Inc., 696 F. App’x 792, 794 n.1 (9th Cir. 2017). We\\nleave that issue for another day. Instead, we vacate the\\ninjunction as to Corizon and remand with instructions to the\\ndistrict court to modify the injunction to exclude Corizon.\\nSee Azar, 911 F.3d at 585. Doing so still provides Edmo the\\nrelief she seeks at this stage. 22\\n\\n\\n\\n\\n 22\\n For similar reasons, we need not reach Edmo’s contention and the\\ndistrict court’s finding that “Corizon and IDOC have a de facto policy or\\npractice of refusing” GCS to prisoners. Edmo, 358 F. Supp. 3d at 1127.\\n\\f78 EDMO V. CORIZON\\n\\n C. Relief Ordered\\n\\n The State next contends that the injunctive relief ordered\\nis overbroad because it requires the State to provide Edmo\\nall “adequate medical care.” The State misconstrues the\\ndistrict court’s order. The order, read in context, requires\\ndefendants to provide GCS, as well as “adequate medical\\ncare” that is “reasonably necessary” to accomplish that\\nend—not every conceivable form of adequate medical care.\\nEdmo, 358 F. Supp. 3d at 1129; see also id. at 1109\\n(“Plaintiff Adree Edmo alleges that prison authorities\\nviolated her Eighth Amendment rights by refusing to provide\\nher with gender confirmation surgery. For the reasons\\nexplained below, the Court agrees and will order defendants\\nto provide her with this procedure, a surgery which is\\nconsidered medically necessary under generally accepted\\nstandards of care.”); id. at 1110 (“[F]or the reasons explained\\nin detail below, IDOC and Corizon will be ordered to\\nprovide Ms. Edmo with gender confirmation surgery.”).\\n\\n The State similarly contends that the injunctive relief\\nordered is overbroad because it requires the State to provide\\nEdmo surgery even though the defendants are not surgeons\\nand no surgeon has evaluated Edmo. We reject this obtuse\\nreading of the district court’s order. The district court\\nordered the State to “take all actions reasonably necessary to\\nprovide Ms. Edmo gender confirmation surgery.” Edmo,\\n358 F. Supp. 3d at 1129. That means that the State must take\\nsteps within its power to provide GCS to Edmo, such as\\nfinding a surgeon and scheduling a surgical evaluation.\\nIndeed, we modified our stay of the district court’s order to\\npermit a surgical consultation, which went forward in April\\n2019. Oral Arg. at 12:00–12:10. The State cannot\\nreasonably understand the district court’s December 13,\\n2018 order to require that the defendants themselves provide\\n\\f EDMO V. CORIZON 79\\n\\nsurgery. To the extent there are issues arising from a surgical\\nevaluation, the State can raise those issues with the district\\ncourt. 23\\n\\n V. Challenges to the Procedure Used by the District\\n Court\\n\\n Finally, the State contends that the district court\\nimproperly converted an evidentiary hearing on a\\npreliminary injunction into a final trial on the merits of\\nEdmo’s Eighth Amendment claim for GCS without giving\\nthem adequate notice and in violation of their Seventh\\nAmendment right to a jury trial. We address and reject each\\ncontention.\\n\\n A. Notice\\n\\n We first address the State’s contention that the district\\ncourt erroneously converted the evidentiary hearing into a\\nfinal trial on the merits without giving the State “clear and\\nunambiguous notice.” Under Federal Rule of Civil\\nProcedure 65(a)(2), “[a] district court may consolidate a\\npreliminary injunction hearing with a trial on the merits, but\\nonly when it provides the parties with clear and\\nunambiguous notice [of the intended consolidation] either\\nbefore the hearing commences or at a time which will afford\\n\\n 23\\n The State contends for the first time in its reply brief that the\\ninjunctive relief ordered was inappropriate because the WPATH\\nStandards of Care require two referrals from qualified mental health\\nprofessionals who have independently assessed the patient before GCS\\nmay be provided. It similarly contends for the first time in its reply in\\nsupport of its motion to dismiss that the order is overbroad because it\\ndoes not specify the type of GCS ordered. Because the State did not\\npresent these arguments in its opening brief, we do not consider them.\\nSee Smith v. Marsh, 194 F.3d 1045, 1052 (9th Cir. 1999).\\n\\f80 EDMO V. CORIZON\\n\\nthe parties a full opportunity to present their respective\\ncases.” Isaacson v. Horne, 716 F.3d 1213, 1220 (9th Cir.\\n2013) (second alteration in original) (quotation omitted).\\n“What constitutes adequate notice depends upon the facts of\\nthe case.” Michenfelder v. Sumner, 860 F.2d 328, 337 (9th\\nCir. 1988).\\n\\n A party challenging consolidation must show not only\\ninadequate notice, but also “substantial prejudice in the\\nsense that [it] was not allowed to present material evidence.”\\nMichenfelder, 860 F.2d at 337; see also 11A Charles Alan\\nWright et al., Federal Practice and Procedure § 2950 (3d\\ned. Apr. 2019 update). “We have on occasion upheld a\\ndistrict court’s failure to give any notice whatsoever before\\nfinally determining the merits after only a preliminary\\ninjunction hearing, where the complaining party has failed\\nto show how additional evidence could have altered the\\noutcome.” Michenfelder, 860 F.2d at 337.\\n\\n At the outset, we note that the State was provided notice,\\ntwice, that the district court considered the evidentiary\\nhearing a final trial on the merits of Edmo’s request for GCS.\\nAt the beginning of the hearing, the district court explained\\n“it’s hard for me to envision this hearing being anything but\\na hearing on a final injunction at least as to that part of the\\nrelief requested [GCS],” and it asked the parties to address\\nby the end of the hearing whether it was for a permanent\\ninjunction. At the close of the hearing, the district court\\nagain questioned whether it could order GCS in a\\npreliminary injunction. It explained that it had, in effect,\\n“kind of treated this hearing as the final hearing” on Edmo’s\\nrequest for GCS, and it again asked the parties to address in\\ntheir oral closings or written briefs whether the hearing was\\none for a permanent injunction. The State never answered\\nthe court’s question or objected to consolidation, despite the\\n\\f EDMO V. CORIZON 81\\n\\ndistrict court specifically noting it had treated the hearing as\\nfinal. Cf. Reilly v. United States, 863 F.2d 149, 160 (1st Cir.\\n1988) (“[W]hen a trial judge announces a proposed course\\nof action which litigants believe to be erroneous, the parties\\ndetrimentally affected must act expeditiously to call the error\\nto the judge’s attention or to cure the defect, not lurk in the\\nbushes waiting to ask for another trial when their litigatory\\nmilk curdles.”). This is not a case where the district court\\ngave no notice whatsoever.\\n\\n Regardless, the State has not shown any prejudice. With\\nfull awareness of the stakes, the district court permitted the\\nparties four months of discovery and held a three-day\\nevidentiary hearing. The parties called seven witnesses,\\nsubmitted declarations in lieu of live testimony for other\\nwitnesses, and submitted thousands of pages of exhibits and\\nextensive pre- and post-trial briefing. Most importantly,\\nboth parties put on extensive evidence concerning the\\ntreatment provided to and withheld from Edmo and why it\\nwas or was not appropriate—the key issue at the hearing.\\n\\n When it comes to identifying prejudice, the State is\\ntellingly short on specifics. It indicates that it “would have\\nobjected” to consolidation, but it failed to do so despite\\nrepeated invitations—indeed, directives—to address the\\nissue. The State also urges that it would have requested that\\nthe named defendants be able to testify live, but it\\nstipulated—knowing full well the stakes of the hearing—to\\nsubmit certain testimony via declaration “[i]n lieu of and/or\\nin addition to live testimony.” Moreover, the State fails to\\nidentify what testimony those witnesses would have offered\\nor explain how presenting that testimony live, instead of via\\ndeclaration, “could have altered the outcome.”\\nMichenfelder, 860 F.2d at 337. The district court did not\\n\\f82 EDMO V. CORIZON\\n\\ncommit reversible error in consolidating the evidentiary\\nhearing with a trial on the merits of Edmo’s request for GCS.\\n\\n B. Seventh Amendment\\n\\n We turn to the State’s related contention that the district\\ncourt violated the defendants’ Seventh Amendment right to\\na jury trial by converting the evidentiary hearing into a trial\\non the merits. We review that contention de novo. Palmer\\nv. Valdez, 560 F.3d 965, 968 (9th Cir. 2009).\\n\\n The Seventh Amendment guarantees the right to a trial\\nby jury “[i]n Suits at common law, where the value in\\ncontroversy shall exceed twenty dollars.” U.S. Const.\\namend. VII. In a case such as this, where legal claims are\\njoined with equitable claims, a party “has a right to jury\\nconsideration of all legal claims, as well as all issues\\ncommon to both claims.” Plummer v. W. Int’l Hotels Co.,\\n656 F.2d 502, 504 n.6 (9th Cir. 1981) (citing Curtis v.\\nLoether, 415 U.S. 189, 196 n.11 (1974)). “Otherwise, the\\ncourt might limit the parties’ opportunity to try to a jury\\nevery issue underlying the legal claims by affording\\npreclusive effect to its own findings of fact on questions that\\nare common to both the legal and equitable claims.” Lacy v.\\nCook County, 897 F.3d 847, 858 (7th Cir. 2018).\\n\\n Like other constitutional rights, the right to a jury trial in\\ncivil suits can be waived. See United States v. Moore,\\n340 U.S. 616, 621 (1951). It is well established that “[a]\\nfailure to object to a proceeding in which the court sits as the\\nfinder of fact waives a valid jury demand as to any claims\\ndecided in that proceeding, at least where it was clear that\\nthe court intended to make fact determinations.” Fillmore v.\\nPage, 358 F.3d 496, 503 (7th Cir. 2004) (quotation omitted);\\nsee also 9 Wright & Miller, Federal Practice and Procedure\\n§ 2321 (“The right to jury trial also may be waived as it has\\n\\f EDMO V. CORIZON 83\\n\\nin many, many cases, by conduct, such as failing to object to\\nor actually participating in a bench trial . . . .”).\\n\\n For example, in White v. McGinnis, we held that “[a]\\nparty’s vigorous participation in a bench trial, without so\\nmuch as a mention of a jury, . . . can only be ascribed to\\nknowledgeable relinquishment of the prior jury demand.”\\n903 F.2d 699, 703 (9th Cir. 1990) (en banc). We explained\\nthat where a party chooses “to argue his case fully before the\\ndistrict judge[,] it is not unjust to hold him to that\\ncommitment.” Id. By contrast, we have held that “[w]hen a\\nparty participates in [a] bench trial ordered by the trial court\\nwhile continuing to demand a jury trial, his ‘continuing\\nobjection’ is ‘sufficient to preserve his right to appeal the\\ndenial of his request for a jury.’” Solis v. County of Los\\nAngeles, 514 F.3d 946, 957 (9th Cir. 2008) (quoting United\\nStates v. Nordbrock, 941 F.2d 947, 950 (9th Cir. 1991)).\\n“This is because the party in such a case is not seeking ‘two\\nbites at the procedural apple’ . . . . Rather, when a trial court\\ndenies a party a jury trial despite the party’s continuing\\ndemand, the party has little choice but to accede to the trial\\ncourt’s ruling and participate in the bench trial.” Id. (citation\\nomitted); see also Lovelace v. Dall, 820 F.2d 223, 228 (7th\\nCir. 1987) (“Another policy justifying the jury demand\\nwaiver rule is the view that it is unfair to permit a party to\\nhave a trial, discover that it has lost, and then raise the jury\\nissue because it is unsatisfied with the result of the trial.”).\\n\\n The State seeks a second bite at the apple. It vigorously\\nparticipated in the evidentiary hearing without ever raising\\nthe right to a jury trial. The State remained silent in the face\\nof statements from the district court that it was considering\\ntreating, and then that it had treated, the hearing as a final\\ntrial on the merits, which made it clear that the court\\n“intended to make fact determinations.” Fillmore, 358 F.3d\\n\\f84 EDMO V. CORIZON\\n\\nat 503. It also remained silent despite the district court\\nasking twice whether the hearing was one for a permanent\\ninjunction—as clear a time as any to raise any concerns\\nabout a jury trial.\\n\\n The State raised the issue of a jury trial for the first time\\non appeal, after the district court ruled against it. Even after\\nthe district court’s ruling, the State made no objection or\\nclaim to a jury trial. This conduct waived the State’s right to\\na jury trial with respect to issues common to Edmo’s request\\nfor an injunction ordering GCS and her legal claims.\\n\\n VI. Conclusion\\n\\n We apply the dictates of the Eighth Amendment today in\\nan area of increased social awareness: transgender health\\ncare. We are not the first to speak on the subject, nor will\\nwe be the last. Our court and others have been considering\\nEighth Amendment claims brought by transgender prisoners\\nfor decades. During that time, the medical community’s\\nunderstanding of what treatments are safe and medically\\nnecessary to treat gender dysphoria has changed as more\\ninformation becomes available, research is undertaken, and\\nexperience is gained. The Eighth-Amendment inquiry takes\\naccount of that developing understanding. See Estelle,\\n429 U.S. at 102–03.\\n\\n We hold that where, as here, the record shows that the\\nmedically necessary treatment for a prisoner’s gender\\ndysphoria is gender confirmation surgery, and responsible\\nprison officials deny such treatment with full awareness of\\nthe prisoner’s suffering, those officials violate the Eighth\\nAmendment’s prohibition on cruel and unusual punishment.\\n\\f EDMO V. CORIZON 85\\n\\n ***\\n\\n We affirm the district court’s entry of an injunction for\\nEdmo. However, we vacate the injunction to the extent it\\napplies to Corizon, Yordy, Siegert, Dr. Young, Dr. Craig,\\nand Dr. Whinnery, in their individual capacities, and remand\\nto the district court to modify the injunction accordingly.\\nThe district court shall also modify the injunction to\\nsubstitute Al Ramirez in his official capacity as Warden of\\nISCI for Yordy.\\n\\n Although we addressed this appeal on an expedited\\nbasis, it has been more than a year since doctors concluded\\nthat GCS is medically necessary for Edmo. We urge the\\nState to move forward. We emphatically do not speak to\\nother cases, but the facts of this case call for expeditious\\neffectuation of the injunction.\\n\\n In light of the nature and urgency of the relief at issue,\\nwe will disfavor any motion, absent extraordinary\\ncircumstances or consent from all parties, to extend the\\nperiod to petition for rehearing or rehearing en banc. Our\\nstay of the district court’s December 13, 2018 order shall\\nautomatically terminate upon issuance of the mandate.\\n\\n Costs on appeal are awarded to Edmo.\\n\\n AFFIRMED IN PART, VACATED IN PART, AND\\nREMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4098,"cells":{"id":{"kind":"number","value":4656014,"string":"4,656,014"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Town of Colorado City"},"decision_date":{"kind":"string","value":"2019-08-26"},"docket_number":{"kind":"string","value":"17-16472"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: MICHAEL DALY HAWKINS and MILAN D. SMITH, JR., Circuit Judges, and BARBARA M. G. LYNN, * District Judge.\", \"parties\": \"\", \"opinions\": [{\"author\": \"M. SMITH, Circuit Judge:\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n UNITED STATES OF AMERICA, No. 17-16472\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:12-cv-08123-\\n HRH\\n TOWN OF COLORADO CITY,\\n ARIZONA; TWIN CITY WATER\\n AUTHORITY, INC., OPINION\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n H. Russel Holland, District Judge, Presiding\\n\\n Argued and Submitted April 18, 2019\\n San Francisco, California\\n\\n Filed August 26, 2019\\n\\n Before: MICHAEL DALY HAWKINS and MILAN D.\\n SMITH, JR., Circuit Judges, and BARBARA M. G.\\n LYNN, * District Judge.\\n\\n Opinion by Judge Milan D. Smith, Jr.\\n\\n\\n *\\n The Honorable Barbara M. G. Lynn, Chief United States District\\nJudge for the Northern District of Texas, sitting by designation.\\n\\f2 UNITED STATES V. TOWN OF COLORADO CITY\\n\\n SUMMARY**\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s judgment in favor\\nof the United States in its action against the Town of\\nColorado City, Arizona brought under the Violent Crime\\nControl and Law Enforcement Act of 1994, 34 U.S.C.\\n§ 12601, which prohibits any governmental authority from\\nengaging in a pattern or practice of conduct by law\\nenforcement officers or government agents that deprives\\npersons of rights, privileges, or immunities secured or\\nprotected by the Constitution or laws of the United States.\\n\\n The United States brought a civil action against the\\nmunicipal defendants and their utility providers alleging a\\npattern or practice of discrimination against residents who\\nwere not members of Fundamentalist Church of Jesus Christ\\nof Latter-Day Saints. The essential allegation of the United\\nStates was that defendants functioned as an arm of the\\nChurch and conspired with Church leaders to use municipal\\nresources to advance Church interests.\\n\\n The panel held that, in holding that defendants violated\\n§ 12601, the district court correctly interpreted the statute to\\nallow for respondeat superior liability. The panel rejected\\nthe assertion that § 12601 requires the United States to\\ndemonstrate that the Towns instituted an official municipal\\npolicy of violating residents’ constitutional rights and\\ntherefore declined to extend the holding in Monell v.\\nDepartment of Social Services, 436 U.S. 658 (1978) to\\nclaims pursuant to § 12601. The panel held that, had\\nCongress wished to eliminate respondeat superior liability\\nunder § 12601, it could have easily done so with explicit\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 3\\n\\nstatutory language. Its decision not to do so suggested that\\nit intended for § 12601, like most civil rights statutes, to\\nallow for respondeat superior liability.\\n\\n The panel held that it was not necessary to address\\nColorado City’s arguments about the district court’s Fourth\\nAmendment-related factual findings because, even if those\\narguments were correct, the error was harmless. The panel\\nfurther held that the district court did not err in admitting the\\nstatements of Church leaders under the co-conspirator\\nexception to the rule against hearsay. The panel held that\\ntaken together, the evidence was sufficient to establish that\\ndefendants conspired with Church members to advance the\\nChurch’s illicit objectives. While certain other statements\\nadmitted by the district court did not fall under the co-\\nconspirator exception, the district court did not err in\\nadmitting them because they were otherwise admissible.\\nThe panel concluded that because of the overwhelming\\nevidence that Colorado City deprived non-Church residents\\nof their constitutional rights, it was more probable than not\\nthat the court would have reached the same verdict on the\\nUnited States’ § 12601 claim even if the challenged\\nstatements had been excluded.\\n\\n\\n COUNSEL\\n\\nJeffrey C. Matura (argued) and Melissa England, Barrett &\\nMatura P.C., Scottsdale, Arizona; R. Blake Hamilton,\\nDurham Jones & Pinegar P.C., Salt Lake City, Utah; for\\nDefendants-Appellants.\\n\\f4 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nChristine A. Monta (argued) and Thomas E. Chandler,\\nAttorneys; John M. Gore, Acting Assistant Attorney\\nGeneral; Appellate Section, Civil Rights Division, United\\nStates Department of Justice, Washington, D.C.; for\\nPlaintiff-Appellee.\\n\\n\\n OPINION\\n\\nM. SMITH, Circuit Judge:\\n\\n When the United States suspected the Town of Colorado\\nCity, Arizona (Colorado City) and Hildale City, Utah\\n(collectively the Towns) of engaging in a pattern or practice\\nof violating the constitutional rights of residents who were\\nnot members of the Fundamentalist Church of Jesus Christ\\nof Latter-Day Saints (FLDS or the Church), it sued the\\nTowns pursuant to the Violent Crime Control and Law\\nEnforcement Act of 1994, 34 U.S.C. § 12601 (formerly\\ncodified at 42 U.S.C. § 14141). 1 After a 44-day trial, a jury\\nreturned an advisory verdict finding the Towns liable. The\\ndistrict court handed down a judgment holding that the\\nTowns violated § 12601, and granted injunctive relief\\nagainst the Towns.\\n\\n Colorado City2 appeals the district court’s decision on\\nthree grounds, all of which fail. The district court correctly\\ninterpreted § 12601 when it concluded that the statute does\\n\\n 1\\n The United States also sued the Towns pursuant to the Fair\\nHousing Act, 42 U.S.C. § 3601 et seq., but this appeal does not concern\\nthat claim.\\n 2\\n Although Hildale City also appealed the district court’s decision,\\nit has since withdrawn from this proceeding.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 5\\n\\nnot require an official municipal policy of violating\\nconstitutional rights in order for the United States to prevail.\\nColorado City’s arguments about the district court’s factual\\nfindings, even if they are correct, do not entitle it to relief\\nbecause the district court’s judgment is supported on other\\ngrounds. The district court did not err in admitting several\\nstatements that Colorado City contends were hearsay. We\\naffirm the district court.\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\n Straddling the Utah and Arizona border, the Short Creek\\nCommunity is a religious settlement composed of the\\nTowns. Most residents are FLDS members and follow the\\nteachings of Warren Jeffs, whom they sustain as a prophet\\nand leader of the Church. Since becoming the head of the\\nChurch in 2002, Jeffs has promulgated a strict set of rules for\\nFLDS members, such as prohibitions on: vacations, toys,\\nattendance at public schools, and displays of affection\\nbetween husbands and wives.\\n\\n The United States brought a civil action against the\\nTowns and their municipal utility providers, Twin City\\nWater Authority and Twin City Power, alleging a pattern or\\npractice of discrimination against non-FLDS residents. The\\nessential allegation of the United States was that the Towns\\nfunctioned as an arm of the Church and conspired with\\nFLDS leaders to use the Towns’ municipal resources to\\nadvance Church interests. The complaint stated a claim\\nagainst the Towns pursuant to § 12601 for violating the\\nEstablishment Clause of the First Amendment, the Fourth\\nAmendment’s prohibition on unreasonable searches and\\nseizures, and the Equal Protection Clause of the Fourteenth\\nAmendment. Because § 12601 does not provide a right to a\\njury trial, the parties agreed that a jury would render only an\\nadvisory verdict.\\n\\f6 UNITED STATES V. TOWN OF COLORADO CITY\\n\\n At trial, the United States argued that FLDS leaders\\nselected the Towns’ leaders and members of the Colorado\\nCity Marshal’s Office (Marshals), which served as the police\\ndepartment for the Towns. The United States offered\\ntestimony that the FLDS “ran the [Towns’] government” and\\nthat the Towns’ government “was a part of the [C]hurch.” It\\nalso offered evidence demonstrating that FLDS leaders\\ninstructed local government officials on how to perform their\\njobs in a way that advanced the Church’s interests.\\nMarshals, for example, ignored violations of the law—such\\nas underage marriage, unlicensed drug distribution, and food\\nstamp fraud—by FLDS members.\\n\\n The Marshal’s Office worked closely with FLDS\\nleaders. Marshals helped FLDS leaders evade service of\\nprocess by the FBI, and ran computer checks of license\\nplates of unfamiliar cars, when asked to do so by FLDS\\nleaders. Cooperation between the Church and the Towns\\neven extended to sharing tangible resources. For example,\\nthe Marshal’s Office provided equipment such as tasers and\\nnight-vision binoculars to Church Security, the FLDS’s\\nprivate security force.\\n\\n The Marshal’s Office also helped Jeffs after he became\\na fugitive. Less than three years after Jeffs became head of\\nthe Church, the United States secured a warrant for his arrest\\non charges of sexual misconduct with children. The FBI\\nsought the help of the Marshal’s Office to locate Jeffs, but\\nthe Marshals did not cooperate; instead, they hindered the\\nFBI’s investigation and helped Jeffs hide for over a year.\\nThe Marshals also provided Jeffs with financial assistance\\nand information on the activities of federal law enforcement\\nto help him evade capture. The Marshals even helped\\ndestroy evidence of the crimes for which Jeffs was accused\\nby burglarizing a former FLDS member’s business.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 7\\n\\n The United States also presented evidence that members\\nof the Marshal’s Office discriminated against non-FLDS\\nresidents. It contended that the Marshals failed to provide\\neffective police protection to residents who were not FLDS\\nmembers. One non-FLDS resident testified, for example,\\nthat a Marshal drove to his home, walked out of his car, and\\n“just came over and grabbed my arm and [] bent it up around\\nmy back.” Although the resident explained that he had a\\nlegal right to occupy the property and presented an\\noccupancy agreement, he was charged with trespassing.\\n\\n The jury returned an advisory verdict finding the Towns\\nliable under § 12601. After an evidentiary hearing, the\\ndistrict court issued a judgment holding the Towns liable\\nunder § 12601 for engaging in a pattern or practice of\\nviolating the First, Fourth, and Fourteenth Amendment\\nrights of their residents. The court determined that the\\nMarshal’s Office “fostered excessive government\\nentanglement with religion” in an effort to “endors[e],\\nfavor[], or promot[e] the FLDS Church at the expense of\\nnon-FLDS residents.” The court also concluded that\\nmembers of the Marshal’s Office “selectively enforce[ed]\\nthe law based upon religion” and arrested several residents\\nwho were not FLDS members without probable cause. The\\ndistrict court ordered injunctive relief requiring the Towns\\nto, among other things, work with a court-appointed monitor\\nto institute national guidelines for constitutional policing.\\n\\n Although the Towns appealed the district court’s finding\\nof liability under § 12601, Hildale City has since withdrawn\\nfrom this appeal. Accordingly, we address only Colorado\\nCity’s arguments.\\n\\n STANDARD OF REVIEW AND JURISDICTION\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\\n\\f8 UNITED STATES V. TOWN OF COLORADO CITY\\n\\n Whether the district court correctly interpreted 34 U.S.C.\\n§ 12601 is a legal question that we review de novo.\\nGoTo.com, Inc. v. Walt Disney Co., 202 F.3d 1199, 1204\\n(9th Cir. 2000). We review the court’s factual findings for\\nclear error. United States v. Christensen, 828 F.3d 763, 815\\n(9th Cir. 2015). We review de novo the court’s\\ninterpretation of the hearsay rule, but review the court’s\\nadmission of evidence for abuse of discretion. United States\\nv. Morales, 720 F.3d 1194, 1199 (9th Cir. 2013).\\n\\n ANALYSIS\\n\\nI. 34 U.S.C. § 12601\\n\\n The principal dispute in this case concerns the proper\\ninterpretation of 34 U.S.C. § 12601. That statute prohibits\\n\\n any governmental authority, or any agent\\n thereof, or any person acting on behalf of a\\n governmental authority, [from engaging] in a\\n pattern or practice of conduct by law\\n enforcement officers or by officials or\\n employees of any governmental agency . . .\\n that deprives persons of rights, privileges, or\\n immunities secured or protected by the\\n Constitution or laws of the United States.\\n\\n Colorado City argues that the district court erred by\\nconstruing the statute as imposing liability on governments\\nfor patterns of constitutional violations committed by their\\nofficers and agents. It asserts that § 12601 requires the\\nUnited States to demonstrate that the Towns “instituted an\\nofficial municipal policy” of violating residents’\\nconstitutional rights. The United States, on the other hand,\\ncontends that the statute “imposes liability on municipalities\\nfor patterns of constitutional violations [that] their law\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 9\\n\\nenforcement officers commit, without requiring an\\nadditional showing that the municipality’s policy or custom\\ncaused those violations.” This issue—whether § 12601\\nimposes respondeat superior liability3—is one of first\\nimpression in our circuit. Cf. United States v. County. of\\nMaricopa, 889 F.3d 648, 653 (9th Cir. 2018) (finding it\\nunnecessary to decide “whether the language of § 12601\\nimposes liability on the basis of general agency principles”).\\n\\n Colorado City relies on the premise that, by including\\n“pattern or practice” in § 12601, Congress used “language\\nwith a well-defined meaning [] developed under [Monell v.\\nDepartment of Social Services, 436 U.S. 658 (1978)] for\\nmunicipal liability.” That contention, however, confuses the\\nrelationship between general liability rules in civil rights\\nstatutes and the Supreme Court’s decision in Monell.\\n\\n “[T]he general rule regarding actions under civil rights\\nstatutes is that respondeat superior applies.” Bonner v.\\nLewis, 857 F.2d 559, 566 (9th Cir. 1988). In Monell, the\\nCourt carved out an exception to this general rule by holding\\nthat a municipality may not be held liable pursuant to\\n42 U.S.C. § 1983 for the actions of its subordinates. Instead,\\nto establish municipal liability, a plaintiff must show that a\\nlocal government’s “policy or custom” led to the plaintiff’s\\ninjury. Monell, 436 U.S. at 694. In reaching its holding, the\\nCourt relied on “the language of § 1983, read against the\\nbackground of the [statute’s] legislative history.” Id. at 691.\\nBecause § 1983 imposes liability only where a state actor,\\n“under color of some official policy, ‘causes’ an employee\\n\\n 3\\n Respondeat superior is “[t]he doctrine holding an employer or\\nprincipal liable for the employee’s or agent’s wrongful acts committed\\nwithin the scope of employment or agency.” Black’s Law Dictionary\\n(11th ed. 2019).\\n\\f10 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nto violate another’s constitutional rights,” the Court\\nreasoned that Congress did not intend to impose vicarious\\nliability on municipalities “solely on the basis of the\\nexistence of an employer-employee relationship with a\\ntortfeasor.” Id. at 692. Moreover, in the Civil Rights Act of\\n1871—the predecessor statute to § 1983—Congress “did not\\nintend municipalities to be held liable unless action pursuant\\nto official municipal policy of some nature caused a\\nconstitutional tort.” Id. at 691.\\n\\n Monell’s holding remains the exception to the general\\nrule. 4 We have declined to bar respondeat superior liability\\nin other contexts. In Bonner, for example, we held that\\nrespondeat superior liability applies to claims pursuant to\\n§ 504 of the Rehabilitation Act of 1973 because “[t]he\\napplication of respondeat superior . . . [is] entirely\\nconsistent with the policy of that statute, which is to\\neliminate discrimination against the handicapped.” 857 F.2d\\nat 566–67 (quoting Patton v. Cumpson, 498 F. Supp. 933,\\n943 (S.D.N.Y. 1980)). And, in Duvall v. County of Kitsap,\\nwe held that respondeat superior liability applies to claims\\nbrought pursuant to Title II of the Americans with\\nDisabilities Act, 42 U.S.C. § 12132. 260 F.3d 1124, 1141\\n(9th Cir. 2001).\\n\\n We likewise decline to extend Monell’s holding to\\nclaims pursuant to § 12601. Several features of the statutory\\ntext lead us to that conclusion. See Esquivel-Quintana v.\\n\\n\\n 4\\n The Supreme Court has held that respondeat superior liability is\\nalso unavailable against local governments pursuant to 42 U.S.C. § 1981.\\nJett v. Dallas Indep. Sch. Dist., 491 U.S. 701, 733 (1989). That decision\\nwas based on the fact that “the express cause of action for damages\\ncreated by § 1983 constitutes the exclusive federal remedy for violation\\nof the rights guaranteed in § 1981.” Id.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 11\\n\\nSessions, 137 S. Ct. 1562, 1568 (2017) (“We begin, as\\nalways, with the text.”).\\n\\n First, § 12601, unlike § 1983, does not include the words\\n“under color of any law, statute, ordinance, regulation,\\ncustom or usage.” That difference is important because, by\\nincluding “custom” in § 1983, Congress expressly\\ncontemplated imposing liability on actors who violated\\nconstitutional rights under an official policy. The absence of\\nthat language from § 12601, therefore, suggests that\\nCongress did not intend to limit liability to those acting\\nunder an official law or policy. Instead, the plain text of\\n§ 12601 shows that any government agent who engages in a\\npattern or practice of conduct that deprives persons of their\\nconstitutional rights violates § 12601.\\n\\n Second, § 12601 does not limit liability to those who\\n“cause [citizens or persons] to be subjected” to a deprivation\\nof their constitutional rights. The Monell Court interpreted\\nthat language, which appears in § 1983, as imposing liability\\n“on a government that, under color of some official policy,\\n‘causes’ an employee to violate another’s constitutional\\nrights.” Monell, 436 U.S. at 692. The lack of that causal\\nphrase in § 12601 suggests that Congress did not intend to\\nlimit local governments’ liability to situations when “the\\naction that is alleged to be unconstitutional implements or\\nexecutes a policy statement, ordinance, regulation, or\\ndecision officially adopted and promulgated by that body’s\\nofficers.” Id. at 690. Taken together, these statutory clues\\npersuade us that Congress intended to allow for respondeat\\nsuperior liability against local governments pursuant to\\n§ 12601.\\n\\n In arguing that the statutory text supports its position,\\nColorado City relies on the fact that the phrase “pattern or\\npractice” appears in both § 1983 and § 12601. That phrase,\\n\\f12 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nit claims, “refers to the same language necessary to show a\\n‘custom’ under Monell.”\\n\\n We acknowledge that Congress used “pattern or\\npractice” in both statutes, and are mindful that “[a] basic\\nprinciple of interpretation is that courts ought to interpret\\nsimilar language in the same way, unless context indicates\\nthat they should do otherwise.” Shirk v. United States ex rel.\\nDep’t of Interior, 773 F.3d 999, 1004 (9th Cir. 2014). That\\nprinciple, however, does not necessarily support Colorado\\nCity’s argument, for Congress has also used “pattern or\\npractice” literally, rather than as a term of art, in several\\nstatutes. See, e.g., 42 U.S.C. § 2000e-6(a) (authorizing the\\nAttorney General to pursue injunctive relief in cases alleging\\na pattern or practice of employment discrimination);\\n42 U.S.C. § 3614(a) (authorizing the Attorney General to\\nbring civil action in cases involving a pattern or practice of\\nFair Housing Act violations); 42 U.S.C. § 10101(e)\\n(authorizing courts to find a pattern or practice of voting\\nrights deprivations). Under those statutes, the United States\\nmust demonstrate only that the conduct alleged “was not an\\nisolated or accidental or peculiar event.” United States v.\\nIronworkers Local 86, 443 F.2d 544, 552 (9th Cir. 1971). It\\nneed not show the existence of an official policy or custom.\\n\\n For this reason, Congress’s use of “pattern or practice”\\nin § 12601 does not support the weight that Colorado City\\nwishes to place upon it. Congress could have used the phrase\\nto refer to an official policy or custom, as in § 1983, but it\\nalso could have used the phrase to refer to a regular event, as\\nin the statutes cited above.\\n\\n Our interpretation of the statute aligns with our\\nrecognition that although “[§] 12601 shares important\\nsimilarities with § 1983[,] . . . . the language of § 12601 goes\\neven further than § 1983.” County of Maricopa, 889 F.3d\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 13\\n\\nat 653. Had Congress wished to eliminate respondeat\\nsuperior liability under § 12601, it could have easily done so\\nwith explicit statutory language. See McNary v. Haitian\\nRefugee Ctr., Inc., 498 U.S. 479, 494 (1991). Its decision\\nnot to do so suggests that it intended for § 12601, like most\\ncivil rights statutes, to allow for respondeat superior\\nliability.\\n\\n Unable to muster support for its position in the statutory\\ntext, Colorado City urges us to examine § 12601’s legislative\\nhistory. But the Supreme Court has admonished that\\n“legislative history is not the law.” Epic Sys. Corp. v. Lewis,\\n138 S. Ct. 1612, 1631 (2018). That principle is particularly\\nsalient in a case where the legislative history “is virtually\\nnon-existent.” Marshall Miller, Police Brutality, 17 Yale L.\\n& Pol’y Rev. 149, 167 (1998); cf. Azar v. Allina Health\\nServs., 139 S. Ct. 1804, 1814 (2019) (declining to accord\\nweight to the Medicare Act’s legislative history when “the\\nlegislative history presented . . . is ambiguous at best”).\\n\\n Even if this case warranted consideration of it, the\\nlegislative history that Colorado City relies on does not\\nsupport its argument. Colorado City focuses on the\\nlegislative history of § 12601’s predecessor bill, the Police\\nAccountability Act of 1991 (PAA). H.R. 2972, 102d Cong.\\n(1991). That history suggests that Congress enacted the\\nPAA to “close [a] gap in the law”—the Justice Department’s\\ninability “to address systematic patterns or practices of\\npolice misconduct.” H.R. Rep. No. 102-242, pt. 1, at 137,\\n138 (1991). The House Judiciary Committee did not define\\n“pattern or practice,” but did mention “[t]wo cases [that]\\nillustrate both the need for this authority and how it will\\nwork.” Id. One case involved police officers from Mason\\nCounty, Washington who beat “citizens in four incidents”\\ndue to “the lack of training.” Id. at 138–39. The other case,\\n\\f14 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nwhich arose from Goldsboro, North Carolina, involved “a\\nyoung black man who was strangled to death by city police\\nofficers.” Id. at 139. Congress did not suggest that either\\nincident arose from an official policy or custom.\\nNonetheless, Colorado City contends that this history—\\nalthough history of a different and superseded law—\\ndemonstrates that Congress did not intend to impose\\nrespondeat superior liability under § 12601.\\n\\n We disagree. At best, the PAA’s legislative history\\nsupports the argument that Congress passed § 12601’s\\npredecessor bill to allow the United States to prosecute\\nmunicipalities when local police departments violate\\nconstitutional rights—whether or not those violations arose\\nfrom an official policy or custom. Our construction of the\\nstatute, which allows local governments to be held liable\\nwhen their agents engage in a pattern or practice of conduct\\nthat deprives persons of their constitutional rights, accords\\nwith that statutory purpose.\\n\\n When interpreting legislation, our role “is to apply the\\nstatute as it is written—even if we think some other approach\\nmight ‘accor[d] with good policy.’” Burrage v. United\\nStates, 571 U.S. 204, 218 (2014) (alteration in original)\\n(quoting Comm’r v. Lundy, 516 U.S. 235, 252 (1996)).\\nSection 12601 provides a civil cause of action to the United\\nStates Attorney General when a local government’s agents\\n“engage in a pattern or practice of conduct . . . that deprives\\npersons of rights, privileges, or immunities secured or\\nprotected by the Constitution or laws of the United States.”\\n34 U.S.C. § 12601. Because the statutory language does not\\ndemonstrate that Congress intended to exclude local\\ngovernments from respondeat superior liability, we hold\\nthat § 12601 imposes liability based on general agency\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 15\\n\\nprinciples. Accordingly, the district court did not err in its\\nconstruction of § 12601. 5\\n\\nII. Factual Findings\\n\\n Colorado City next argues that the district court made\\nthree mistakes in its factual findings related to the Towns’\\nalleged violations of the Fourth Amendment. First, it argues\\nthat the court erroneously included legal conclusions in its\\nfindings of fact, in violation of Federal Rule of Civil\\nProcedure 52(a). Second, it contends that the district court\\nimproperly adopted the government’s proposed findings of\\nfact and failed to independently evaluate the evidence.\\nThird, it argues that the court improperly made summary\\nfindings in its written judgment.\\n\\n We need not address these alleged errors because they\\ndo not affect the district court’s holding that Colorado City\\nviolated § 12601. As the district court stated in its judgment,\\na finding “of a pattern or practice of violating any one of the\\nthree constitutional amendments in question”—the First,\\nFourth, and Fourteenth Amendments—“would entitle [the\\nUnited States] to relief.” The district court held that the\\nTowns violated all three constitutional amendments.\\n\\n 5\\n We reject Colorado City’s argument that our interpretation of\\n34 U.S.C. § 12601 violates § 5 of the Fourteenth Amendment. It argues\\nthat the district court violated the Supreme Court’s decision in City of\\nCanton v. Harris because it “interpreted [§ 12601] to impose vicarious\\nliability upon the Towns without requiring the United States to show that\\nthe Towns were responsible for the alleged misconduct.” 489 U.S. 378\\n(1989). Not so. The Court’s decision in City of Canton, which stated\\nthat permitting respondeat superior liability against local governments\\nunder § 1983 “would implicate serious questions of federalism,” was\\nlimited to that statute. Id. at 392. The Court did not hold that it was\\nunconstitutional to permit respondeat superior liability against local\\ngovernments in any context.\\n\\f16 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nBecause Colorado City’s arguments relate only to the court’s\\nfindings of fact as to the Fourth Amendment violation, the\\ncourt’s judgment stands even if Colorado City is correct.\\nAccordingly, any such purported error by the district court\\nwas harmless.\\n\\n Colorado City urges us to nevertheless address the merits\\nof its argument because “the district court’s erroneous\\nfactual findings exposed the Towns to unfounded liability\\nthrough lawsuits that Isaac Wyler, Patrick Pipkin, and\\nAndrew Chatwin filed for unlawful arrest.” Those lawsuits,\\nhowever, are irrelevant to our harmless error analysis\\nbecause the relevant question is “whether the [] verdict\\nactually rendered in this trial” was attributable to the district\\ncourt’s error. Sullivan v. Louisiana, 508 U.S. 275, 279\\n(1993). Thus, the fact that other parties may have sued the\\nTowns does not affect our conclusion.\\n\\nIII. Admission of FLDS Leaders’ Statements\\n\\n Finally, we turn to Colorado City’s argument that the\\ndistrict court erred when it admitted the statements of\\nvarious FLDS leaders under the co-conspirator exception to\\nthe rule against hearsay.\\n\\n Before trial, the United States moved to admit several\\nFLDS leaders’ statements under the co-conspirator\\nexception. Fed. R. Evid. 801(d)(2)(E); see Bourjaily v.\\nUnited States, 483 U.S. 171, 173 (1987) (“A statement is not\\nhearsay if . . . [t]he statement is offered against a party and\\nis . . . a statement by a coconspirator of a party during the\\ncourse and in furtherance of the conspiracy.” (alterations in\\noriginal)). The district court held the United States’ motion\\nin abeyance pending trial, reasoning:\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 17\\n\\n Although a final decision on this point must\\n await trial evidence, prior proceedings in this\\n case cause the court to believe that the United\\n States will succeed in producing evidence of\\n a joint venture or concert of action between\\n the FLDS Church and the [Towns]. Subject\\n to the United States proving up its concerted\\n action contention, the court is prepared to\\n rule that out-of-court statements of FLDS\\n leaders in furtherance of concerted action\\n between the FLDS Church and [the Towns]\\n [are] not hearsay and [are] admissible.\\n\\nLater during trial, the court determined that the United States\\nhad satisfied its burden of establishing the existence of a\\nconspiracy. It instructed the jury that “the [United States]\\nhas made its case, as far as I am concerned, for purposes of\\nthe admission of the testimony, that there was a\\n[conspiracy].”\\n\\n Colorado City first argues that the district court clearly\\nerred by finding the existence of a conspiracy between the\\nChurch and the Towns. Its contention rests largely on the\\nclaim that “[w]hen the district court ‘tentatively’ ruled that\\nthe United States had established a conspiracy, the evidence\\nwas insufficient to show the existence of a conspiracy for\\npurposes of Rule 801(d)(2)(E).” That argument fails\\nbecause “[i]t is not controlling [] whether sufficient\\nindependent evidence connecting [the Church] with the\\nconspiracy existed at the time the trial judge made his first\\nruling under Rule 801(d)(2)(E).” United States v. Watkins,\\n600 F.2d 201, 204–05 (9th Cir. 1979). “In ascertaining\\nwhether the foundation has been established, we can,\\ntherefore, consider all the evidence independent of the\\nchallenged statements, regardless of the order of proof.”\\n\\f18 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nUnited States v. Miranda-Uriarte, 649 F.2d 1345, 1351 (9th\\nCir. 1981). The evidence must “be considered in a light most\\nfavorable to the government.” Id.\\n\\n Here, the United States presented extensive evidence at\\ntrial that supported the existence of a conspiracy between the\\nChurch and the Towns. That evidence included testimony\\nthat: officials from the Towns attended meetings in which\\nFLDS leaders instructed them on how to handle legal issues\\nin a way that advanced the Church’s interests; Jeffs\\nexcommunicated the Towns’ leaders who did not follow his\\norders; FLDS leaders determined who would occupy the\\nTowns’ government positions such as mayor, city council\\nmembers, and police officers; the Marshal’s Office was\\nwillfully blind to FLDS members’ illegal activities;\\nmembers of the Marshal’s Office helped Jeffs evade capture\\nby the FBI while he was a fugitive; and several of the Towns’\\nofficials “spied” on residents who the Church considered\\n“out of conformance with [FLDS] regulations.” Taken\\ntogether, this evidence is sufficient to establish that the\\nTowns conspired with FLDS members to advance the\\nChurch’s illicit objectives. The district court did not clearly\\nerr in making that finding.\\n\\n Colorado City also argues that the district court legally\\nerred by misinterpreting the co-conspirator exception to the\\nrule against hearsay. That argument, however, is belied by\\nthe record. The district court repeatedly acknowledged that\\na statement must be made in furtherance of a conspiracy to\\nqualify under Rule 801(d)(2)(E). Moreover, when\\ninstructing the jury, the court accurately stated the standard\\nfor statements to fall under the co-conspirator exception:\\n“one, certain individuals worked with the defendants toward\\na common goal in a joint venture or in a concerted effort and,\\ntwo, that those individuals made out-of-court statements\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 19\\n\\nduring and in furtherance of that effort.” Accordingly, we\\nreject Colorado City’s argument that the court misconstrued\\nthe rules of evidence.\\n\\n Lastly, contrary to Colorado City’s argument, the district\\ncourt did not abuse its discretion by admitting over twenty\\nstatements by FLDS leaders. Several of the statements,\\nincluding several transcriptions of Jeffs’ dictations and\\ntelephone calls, were properly admitted under the co-\\nconspirator exception. These statements “catalogued and\\nanalyzed factors relevant” to the alleged conspiracy. United\\nStates v. Schmit, 881 F.2d 608, 612 (9th Cir. 1989). They\\ninclude statements by Jeffs recounting instructions he gave\\nto FLDS members to perform underage marriages,\\ndescribing the appointment of FLDS members to leadership\\npositions in the Towns, and stating that FLDS members had\\ngone into hiding “to not be served [] legal papers.” That\\n“some portions of the statement[s] may have been ‘idle\\nchatter’ or [‘]casual admissions of culpability’ [does] not\\nrender” the statements inadmissible. Id. That is especially\\ntrue because, during trial, the Towns moved to exclude “the\\nstatement[s] as a whole” rather than “particular passages in\\nthe statement[s].” Id.\\n\\n While certain other statements admitted by the district\\ncourt and challenged by Colorado City do not fall under the\\nco-conspirator exception in Rule 801(d)(2)(E), we hold that\\nthe district court did not err in admitting them because they\\nwere otherwise admissible. See United States v. Alexander,\\n48 F.3d 1477, 1487 (9th Cir. 1995) (upholding a district\\ncourt's admission of evidence on a ground “different from\\nthe reason given by the district court”).\\n\\n Several of those statements were admissible because\\nthey were not hearsay. These included instructions by Jeffs\\nto FLDS members to not communicate with, and to\\n\\f20 UNITED STATES V. TOWN OF COLORADO CITY\\n\\notherwise avoid, “apostates”—residents who were once\\nFLDS members, but who had left the Church. See United\\nStates v. Chung, 659 F.3d 815, 833 (9th Cir. 2011)\\n(“Instructions to an individual to do something are . . . not\\nhearsay . . . because they are declarations of fact and\\ntherefore are not capable of being true or false.” (alterations\\nin original) (quoting United States v. Reilly, 33 F.3d 1396,\\n1410 (3d Cir. 1994))). Still others, such as Jeffs’ statement\\nthat he prayed for the destruction of Arizona and Utah, were\\nintroduced for their effect on the listener. See United States\\nv. Payne, 944 F.2d 1458, 1472 (9th Cir. 1991).\\n\\n Other challenged statements, such as Jeffs’ statement\\nabout “[t]he attack of [] enemies upon [the FLDS\\ncommunity],” were admissible under the business records\\nexception to the hearsay rule. Fed. R. Evid. 803(6). Despite\\nthe Towns’ argument, Ranger John Nick Hanna was\\nqualified to testify about Jeffs’ dictations. Rule 803(6) “only\\nrequires [testimony by] ‘someone with knowledge’ about the\\nrecord-keeping, not necessarily . . . someone with\\nknowledge about how the reports were made or maintained,”\\nABS Entm’t, Inc. v. CBS Corp., 908 F.3d 405, 426 (9th Cir.\\n2018), and Hanna had spent over four years studying the\\ndocuments about which he testified. Colorado City has not\\nnot shown that the dictations do not meet the trustworthiness\\nstandard of the business records exception.\\n\\n Even if the district court erroneously admitted some\\nhearsay statements, reversal is not warranted. Because the\\njury rendered only an advisory verdict on the United States’\\n§ 12601 claim, our review is limited to “the findings of the\\ncourt as if there had been no verdict from an advisory jury.”\\nAshland v. Ling-Temco-Vought, Inc., 711 F.2d 1431, 1438\\n(9th Cir. 1983) (quoting 9 Charles Alan Wright & Arthur R.\\nMiller, Federal Practice and Procedure § 2335 (3d ed.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 21\\n\\n1971)). Because the judge ultimately ruled on the United\\nStates’ § 12601 claim, the court “ha[d] discretion to receive\\nevidence that might be inadmissible in a jury trial.”\\nHollinger v. United States, 651 F.2d 636, 640 (9th Cir.\\n1981). We conclude that because of the overwhelming\\nevidence that Colorado City deprived non-FLDS residents\\nof their constitutional rights, “it is more probable than not”\\nthat the court would have reached the same verdict on the\\nUnited States’ § 12601 claim even if the challenged\\nstatements had been excluded. Nationwide Life Ins. Co. v.\\nRichards, 541 F.3d 903, 911 (9th Cir. 2008).\\n\\n CONCLUSION\\n\\n In holding that the Towns violated § 12601, the district\\ncourt correctly interpreted the statute and did not err in\\nadmitting the statements of FLDS leaders. We need not\\naddress Colorado City’s arguments about the district court’s\\nFourth Amendment-related factual findings because, even if\\nthose arguments are correct, the error was harmless.\\n\\n AFFIRMED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4433267/\", \"author_raw\": \"M. SMITH, Circuit Judge:\"}]}"},"judge1":{"kind":"string","value":"MICHAEL DALY HAWKINS"},"judge2":{"kind":"string","value":"MILAN D SMITH JR"},"judge3":{"kind":"string","value":"BARBARA M G LYNN"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{\"MICHAEL DALY HAWKINS\": \"MILAN D SMITH, JR , Circuit\", \"MILAN D SMITH JR\": \"MILAN D SMITH, JR , Circuit\", \"BARBARA M G LYNN\": \", District\"}"},"opinions_count":{"kind":"number","value":1,"string":"1"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":0,"string":"0"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4656014/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Town of Colorado City\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-26\nOpinion type: majority\nHas dissenting opinion: No\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITED STATES OF AMERICA, No. 17-16472\n Plaintiff-Appellee,\n D.C. No.\n v. 3:12-cv-08123-\n HRH\n TOWN OF COLORADO CITY,\n ARIZONA; TWIN CITY WATER\n AUTHORITY, INC., OPINION\n Defendants-Appellants.\n\n Appeal from the United States District Court\n for the District of Arizona\n H. Russel Holland, District Judge, Presiding\n\n Argued and Submitted April 18, 2019\n San Francisco, California\n\n Filed August 26, 2019\n\n Before: MICHAEL DALY HAWKINS and MILAN D.\n SMITH, JR., Circuit Judges, and BARBARA M. G.\n LYNN, * District Judge.\n\n Opinion by Judge Milan D. Smith, Jr.\n\n\n *\n The Honorable Barbara M. G. Lynn, Chief United States District\nJudge for the Northern District of Texas, sitting by designation.\n\f2 UNITED STATES V. TOWN OF COLORADO CITY\n\n SUMMARY**\n\n\n Civil Rights\n\n The panel affirmed the district court’s judgment in favor\nof the United States in its action against the Town of\nColorado City, Arizona brought under the Violent Crime\nControl and Law Enforcement Act of 1994, 34 U.S.C.\n§ 12601, which prohibits any governmental authority from\nengaging in a pattern or practice of conduct by law\nenforcement officers or government agents that deprives\npersons of rights, privileges, or immunities secured or\nprotected by the Constitution or laws of the United States.\n\n The United States brought a civil action against the\nmunicipal defendants and their utility providers alleging a\npattern or practice of discrimination against residents who\nwere not members of Fundamentalist Church of Jesus Christ\nof Latter-Day Saints. The essential allegation of the United\nStates was that defendants functioned as an arm of the\nChurch and conspired with Church leaders to use municipal\nresources to advance Church interests.\n\n The panel held that, in holding that defendants violated\n§ 12601, the district court correctly interpreted the statute to\nallow for respondeat superior liability. The panel rejected\nthe assertion that § 12601 requires the United States to\ndemonstrate that the Towns instituted an official municipal\npolicy of violating residents’ constitutional rights and\ntherefore declined to extend the holding in Monell v.\nDepartment of Social Services, 436 U.S. 658 (1978) to\nclaims pursuant to § 12601. The panel held that, had\nCongress wished to eliminate respondeat superior liability\nunder § 12601, it could have easily done so with explicit\n\f UNITED STATES V. TOWN OF COLORADO CITY 3\n\nstatutory language. Its decision not to do so suggested that\nit intended for § 12601, like most civil rights statutes, to\nallow for respondeat superior liability.\n\n The panel held that it was not necessary to address\nColorado City’s arguments about the district court’s Fourth\nAmendment-related factual findings because, even if those\narguments were correct, the error was harmless. The panel\nfurther held that the district court did not err in admitting the\nstatements of Church leaders under the co-conspirator\nexception to the rule against hearsay. The panel held that\ntaken together, the evidence was sufficient to establish that\ndefendants conspired with Church members to advance the\nChurch’s illicit objectives. While certain other statements\nadmitted by the district court did not fall under the co-\nconspirator exception, the district court did not err in\nadmitting them because they were otherwise admissible.\nThe panel concluded that because of the overwhelming\nevidence that Colorado City deprived non-Church residents\nof their constitutional rights, it was more probable than not\nthat the court would have reached the same verdict on the\nUnited States’ § 12601 claim even if the challenged\nstatements had been excluded.\n\n\n COUNSEL\n\nJeffrey C. Matura (argued) and Melissa England, Barrett &\nMatura P.C., Scottsdale, Arizona; R. Blake Hamilton,\nDurham Jones & Pinegar P.C., Salt Lake City, Utah; for\nDefendants-Appellants.\n\f4 UNITED STATES V. TOWN OF COLORADO CITY\n\nChristine A. Monta (argued) and Thomas E. Chandler,\nAttorneys; John M. Gore, Acting Assistant Attorney\nGeneral; Appellate Section, Civil Rights Division, United\nStates Department of Justice, Washington, D.C.; for\nPlaintiff-Appellee.\n\n\n OPINION\n\nM. SMITH, Circuit Judge:\n\n When the United States suspected the Town of Colorado\nCity, Arizona (Colorado City) and Hildale City, Utah\n(collectively the Towns) of engaging in a pattern or practice\nof violating the constitutional rights of residents who were\nnot members of the Fundamentalist Church of Jesus Christ\nof Latter-Day Saints (FLDS or the Church), it sued the\nTowns pursuant to the Violent Crime Control and Law\nEnforcement Act of 1994, 34 U.S.C. § 12601 (formerly\ncodified at 42 U.S.C. § 14141). 1 After a 44-day trial, a jury\nreturned an advisory verdict finding the Towns liable. The\ndistrict court handed down a judgment holding that the\nTowns violated § 12601, and granted injunctive relief\nagainst the Towns.\n\n Colorado City2 appeals the district court’s decision on\nthree grounds, all of which fail. The district court correctly\ninterpreted § 12601 when it concluded that the statute does\n\n 1\n The United States also sued the Towns pursuant to the Fair\nHousing Act, 42 U.S.C. § 3601 et seq., but this appeal does not concern\nthat claim.\n 2\n Although Hildale City also appealed the district court’s decision,\nit has since withdrawn from this proceeding.\n\f UNITED STATES V. TOWN OF COLORADO CITY 5\n\nnot require an official municipal policy of violating\nconstitutional rights in order for the United States to prevail.\nColorado City’s arguments about the district court’s factual\nfindings, even if they are correct, do not entitle it to relief\nbecause the district court’s judgment is supported on other\ngrounds. The district court did not err in admitting several\nstatements that Colorado City contends were hearsay. We\naffirm the district court.\n\n FACTUAL AND PROCEDURAL BACKGROUND\n\n Straddling the Utah and Arizona border, the Short Creek\nCommunity is a religious settlement composed of the\nTowns. Most residents are FLDS members and follow the\nteachings of Warren Jeffs, whom they sustain as a prophet\nand leader of the Church. Since becoming the head of the\nChurch in 2002, Jeffs has promulgated a strict set of rules for\nFLDS members, such as prohibitions on: vacations, toys,\nattendance at public schools, and displays of affection\nbetween husbands and wives.\n\n The United States brought a civil action against the\nTowns and their municipal utility providers, Twin City\nWater Authority and Twin City Power, alleging a pattern or\npractice of discrimination against non-FLDS residents. The\nessential allegation of the United States was that the Towns\nfunctioned as an arm of the Church and conspired with\nFLDS leaders to use the Towns’ municipal resources to\nadvance Church interests. The complaint stated a claim\nagainst the Towns pursuant to § 12601 for violating the\nEstablishment Clause of the First Amendment, the Fourth\nAmendment’s prohibition on unreasonable searches and\nseizures, and the Equal Protection Clause of the Fourteenth\nAmendment. Because § 12601 does not provide a right to a\njury trial, the parties agreed that a jury would render only an\nadvisory verdict.\n\f6 UNITED STATES V. TOWN OF COLORADO CITY\n\n At trial, the United States argued that FLDS leaders\nselected the Towns’ leaders and members of the Colorado\nCity Marshal’s Office (Marshals), which served as the police\ndepartment for the Towns. The United States offered\ntestimony that the FLDS “ran the [Towns’] government” and\nthat the Towns’ government “was a part of the [C]hurch.” It\nalso offered evidence demonstrating that FLDS leaders\ninstructed local government officials on how to perform their\njobs in a way that advanced the Church’s interests.\nMarshals, for example, ignored violations of the law—such\nas underage marriage, unlicensed drug distribution, and food\nstamp fraud—by FLDS members.\n\n The Marshal’s Office worked closely with FLDS\nleaders. Marshals helped FLDS leaders evade service of\nprocess by the FBI, and ran computer checks of license\nplates of unfamiliar cars, when asked to do so by FLDS\nleaders. Cooperation between the Church and the Towns\neven extended to sharing tangible resources. For example,\nthe Marshal’s Office provided equipment such as tasers and\nnight-vision binoculars to Church Security, the FLDS’s\nprivate security force.\n\n The Marshal’s Office also helped Jeffs after he became\na fugitive. Less than three years after Jeffs became head of\nthe Church, the United States secured a warrant for his arrest\non charges of sexual misconduct with children. The FBI\nsought the help of the Marshal’s Office to locate Jeffs, but\nthe Marshals did not cooperate; instead, they hindered the\nFBI’s investigation and helped Jeffs hide for over a year.\nThe Marshals also provided Jeffs with financial assistance\nand information on the activities of federal law enforcement\nto help him evade capture. The Marshals even helped\ndestroy evidence of the crimes for which Jeffs was accused\nby burglarizing a former FLDS member’s business.\n\f UNITED STATES V. TOWN OF COLORADO CITY 7\n\n The United States also presented evidence that members\nof the Marshal’s Office discriminated against non-FLDS\nresidents. It contended that the Marshals failed to provide\neffective police protection to residents who were not FLDS\nmembers. One non-FLDS resident testified, for example,\nthat a Marshal drove to his home, walked out of his car, and\n“just came over and grabbed my arm and [] bent it up around\nmy back.” Although the resident explained that he had a\nlegal right to occupy the property and presented an\noccupancy agreement, he was charged with trespassing.\n\n The jury returned an advisory verdict finding the Towns\nliable under § 12601. After an evidentiary hearing, the\ndistrict court issued a judgment holding the Towns liable\nunder § 12601 for engaging in a pattern or practice of\nviolating the First, Fourth, and Fourteenth Amendment\nrights of their residents. The court determined that the\nMarshal’s Office “fostered excessive government\nentanglement with religion” in an effort to “endors[e],\nfavor[], or promot[e] the FLDS Church at the expense of\nnon-FLDS residents.” The court also concluded that\nmembers of the Marshal’s Office “selectively enforce[ed]\nthe law based upon religion” and arrested several residents\nwho were not FLDS members without probable cause. The\ndistrict court ordered injunctive relief requiring the Towns\nto, among other things, work with a court-appointed monitor\nto institute national guidelines for constitutional policing.\n\n Although the Towns appealed the district court’s finding\nof liability under § 12601, Hildale City has since withdrawn\nfrom this appeal. Accordingly, we address only Colorado\nCity’s arguments.\n\n STANDARD OF REVIEW AND JURISDICTION\n\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\n\f8 UNITED STATES V. TOWN OF COLORADO CITY\n\n Whether the district court correctly interpreted 34 U.S.C.\n§ 12601 is a legal question that we review de novo.\nGoTo.com, Inc. v. Walt Disney Co., 202 F.3d 1199, 1204\n(9th Cir. 2000). We review the court’s factual findings for\nclear error. United States v. Christensen, 828 F.3d 763, 815\n(9th Cir. 2015). We review de novo the court’s\ninterpretation of the hearsay rule, but review the court’s\nadmission of evidence for abuse of discretion. United States\nv. Morales, 720 F.3d 1194, 1199 (9th Cir. 2013).\n\n ANALYSIS\n\nI. 34 U.S.C. § 12601\n\n The principal dispute in this case concerns the proper\ninterpretation of 34 U.S.C. § 12601. That statute prohibits\n\n any governmental authority, or any agent\n thereof, or any person acting on behalf of a\n governmental authority, [from engaging] in a\n pattern or practice of conduct by law\n enforcement officers or by officials or\n employees of any governmental agency . . .\n that deprives persons of rights, privileges, or\n immunities secured or protected by the\n Constitution or laws of the United States.\n\n Colorado City argues that the district court erred by\nconstruing the statute as imposing liability on governments\nfor patterns of constitutional violations committed by their\nofficers and agents. It asserts that § 12601 requires the\nUnited States to demonstrate that the Towns “instituted an\nofficial municipal policy” of violating residents’\nconstitutional rights. The United States, on the other hand,\ncontends that the statute “imposes liability on municipalities\nfor patterns of constitutional violations [that] their law\n\f UNITED STATES V. TOWN OF COLORADO CITY 9\n\nenforcement officers commit, without requiring an\nadditional showing that the municipality’s policy or custom\ncaused those violations.” This issue—whether § 12601\nimposes respondeat superior liability3—is one of first\nimpression in our circuit. Cf. United States v. County. of\nMaricopa, 889 F.3d 648, 653 (9th Cir. 2018) (finding it\nunnecessary to decide “whether the language of § 12601\nimposes liability on the basis of general agency principles”).\n\n Colorado City relies on the premise that, by including\n“pattern or practice” in § 12601, Congress used “language\nwith a well-defined meaning [] developed under [Monell v.\nDepartment of Social Services, 436 U.S. 658 (1978)] for\nmunicipal liability.” That contention, however, confuses the\nrelationship between general liability rules in civil rights\nstatutes and the Supreme Court’s decision in Monell.\n\n “[T]he general rule regarding actions under civil rights\nstatutes is that respondeat superior applies.” Bonner v.\nLewis, 857 F.2d 559, 566 (9th Cir. 1988). In Monell, the\nCourt carved out an exception to this general rule by holding\nthat a municipality may not be held liable pursuant to\n42 U.S.C. § 1983 for the actions of its subordinates. Instead,\nto establish municipal liability, a plaintiff must show that a\nlocal government’s “policy or custom” led to the plaintiff’s\ninjury. Monell, 436 U.S. at 694. In reaching its holding, the\nCourt relied on “the language of § 1983, read against the\nbackground of the [statute’s] legislative history.” Id. at 691.\nBecause § 1983 imposes liability only where a state actor,\n“under color of some official policy, ‘causes’ an employee\n\n 3\n Respondeat superior is “[t]he doctrine holding an employer or\nprincipal liable for the employee’s or agent’s wrongful acts committed\nwithin the scope of employment or agency.” Black’s Law Dictionary\n(11th ed. 2019).\n\f10 UNITED STATES V. TOWN OF COLORADO CITY\n\nto violate another’s constitutional rights,” the Court\nreasoned that Congress did not intend to impose vicarious\nliability on municipalities “solely on the basis of the\nexistence of an employer-employee relationship with a\ntortfeasor.” Id. at 692. Moreover, in the Civil Rights Act of\n1871—the predecessor statute to § 1983—Congress “did not\nintend municipalities to be held liable unless action pursuant\nto official municipal policy of some nature caused a\nconstitutional tort.” Id. at 691.\n\n Monell’s holding remains the exception to the general\nrule. 4 We have declined to bar respondeat superior liability\nin other contexts. In Bonner, for example, we held that\nrespondeat superior liability applies to claims pursuant to\n§ 504 of the Rehabilitation Act of 1973 because “[t]he\napplication of respondeat superior . . . [is] entirely\nconsistent with the policy of that statute, which is to\neliminate discrimination against the handicapped.” 857 F.2d\nat 566–67 (quoting Patton v. Cumpson, 498 F. Supp. 933,\n943 (S.D.N.Y. 1980)). And, in Duvall v. County of Kitsap,\nwe held that respondeat superior liability applies to claims\nbrought pursuant to Title II of the Americans with\nDisabilities Act, 42 U.S.C. § 12132. 260 F.3d 1124, 1141\n(9th Cir. 2001).\n\n We likewise decline to extend Monell’s holding to\nclaims pursuant to § 12601. Several features of the statutory\ntext lead us to that conclusion. See Esquivel-Quintana v.\n\n\n 4\n The Supreme Court has held that respondeat superior liability is\nalso unavailable against local governments pursuant to 42 U.S.C. § 1981.\nJett v. Dallas Indep. Sch. Dist., 491 U.S. 701, 733 (1989). That decision\nwas based on the fact that “the express cause of action for damages\ncreated by § 1983 constitutes the exclusive federal remedy for violation\nof the rights guaranteed in § 1981.” Id.\n\f UNITED STATES V. TOWN OF COLORADO CITY 11\n\nSessions, 137 S. Ct. 1562, 1568 (2017) (“We begin, as\nalways, with the text.”).\n\n First, § 12601, unlike § 1983, does not include the words\n“under color of any law, statute, ordinance, regulation,\ncustom or usage.” That difference is important because, by\nincluding “custom” in § 1983, Congress expressly\ncontemplated imposing liability on actors who violated\nconstitutional rights under an official policy. The absence of\nthat language from § 12601, therefore, suggests that\nCongress did not intend to limit liability to those acting\nunder an official law or policy. Instead, the plain text of\n§ 12601 shows that any government agent who engages in a\npattern or practice of conduct that deprives persons of their\nconstitutional rights violates § 12601.\n\n Second, § 12601 does not limit liability to those who\n“cause [citizens or persons] to be subjected” to a deprivation\nof their constitutional rights. The Monell Court interpreted\nthat language, which appears in § 1983, as imposing liability\n“on a government that, under color of some official policy,\n‘causes’ an employee to violate another’s constitutional\nrights.” Monell, 436 U.S. at 692. The lack of that causal\nphrase in § 12601 suggests that Congress did not intend to\nlimit local governments’ liability to situations when “the\naction that is alleged to be unconstitutional implements or\nexecutes a policy statement, ordinance, regulation, or\ndecision officially adopted and promulgated by that body’s\nofficers.” Id. at 690. Taken together, these statutory clues\npersuade us that Congress intended to allow for respondeat\nsuperior liability against local governments pursuant to\n§ 12601.\n\n In arguing that the statutory text supports its position,\nColorado City relies on the fact that the phrase “pattern or\npractice” appears in both § 1983 and § 12601. That phrase,\n\f12 UNITED STATES V. TOWN OF COLORADO CITY\n\nit claims, “refers to the same language necessary to show a\n‘custom’ under Monell.”\n\n We acknowledge that Congress used “pattern or\npractice” in both statutes, and are mindful that “[a] basic\nprinciple of interpretation is that courts ought to interpret\nsimilar language in the same way, unless context indicates\nthat they should do otherwise.” Shirk v. United States ex rel.\nDep’t of Interior, 773 F.3d 999, 1004 (9th Cir. 2014). That\nprinciple, however, does not necessarily support Colorado\nCity’s argument, for Congress has also used “pattern or\npractice” literally, rather than as a term of art, in several\nstatutes. See, e.g., 42 U.S.C. § 2000e-6(a) (authorizing the\nAttorney General to pursue injunctive relief in cases alleging\na pattern or practice of employment discrimination);\n42 U.S.C. § 3614(a) (authorizing the Attorney General to\nbring civil action in cases involving a pattern or practice of\nFair Housing Act violations); 42 U.S.C. § 10101(e)\n(authorizing courts to find a pattern or practice of voting\nrights deprivations). Under those statutes, the United States\nmust demonstrate only that the conduct alleged “was not an\nisolated or accidental or peculiar event.” United States v.\nIronworkers Local 86, 443 F.2d 544, 552 (9th Cir. 1971). It\nneed not show the existence of an official policy or custom.\n\n For this reason, Congress’s use of “pattern or practice”\nin § 12601 does not support the weight that Colorado City\nwishes to place upon it. Congress could have used the phrase\nto refer to an official policy or custom, as in § 1983, but it\nalso could have used the phrase to refer to a regular event, as\nin the statutes cited above.\n\n Our interpretation of the statute aligns with our\nrecognition that although “[§] 12601 shares important\nsimilarities with § 1983[,] . . . . the language of § 12601 goes\neven further than § 1983.” County of Maricopa, 889 F.3d\n\f UNITED STATES V. TOWN OF COLORADO CITY 13\n\nat 653. Had Congress wished to eliminate respondeat\nsuperior liability under § 12601, it could have easily done so\nwith explicit statutory language. See McNary v. Haitian\nRefugee Ctr., Inc., 498 U.S. 479, 494 (1991). Its decision\nnot to do so suggests that it intended for § 12601, like most\ncivil rights statutes, to allow for respondeat superior\nliability.\n\n Unable to muster support for its position in the statutory\ntext, Colorado City urges us to examine § 12601’s legislative\nhistory. But the Supreme Court has admonished that\n“legislative history is not the law.” Epic Sys. Corp. v. Lewis,\n138 S. Ct. 1612, 1631 (2018). That principle is particularly\nsalient in a case where the legislative history “is virtually\nnon-existent.” Marshall Miller, Police Brutality, 17 Yale L.\n& Pol’y Rev. 149, 167 (1998); cf. Azar v. Allina Health\nServs., 139 S. Ct. 1804, 1814 (2019) (declining to accord\nweight to the Medicare Act’s legislative history when “the\nlegislative history presented . . . is ambiguous at best”).\n\n Even if this case warranted consideration of it, the\nlegislative history that Colorado City relies on does not\nsupport its argument. Colorado City focuses on the\nlegislative history of § 12601’s predecessor bill, the Police\nAccountability Act of 1991 (PAA). H.R. 2972, 102d Cong.\n(1991). That history suggests that Congress enacted the\nPAA to “close [a] gap in the law”—the Justice Department’s\ninability “to address systematic patterns or practices of\npolice misconduct.” H.R. Rep. No. 102-242, pt. 1, at 137,\n138 (1991). The House Judiciary Committee did not define\n“pattern or practice,” but did mention “[t]wo cases [that]\nillustrate both the need for this authority and how it will\nwork.” Id. One case involved police officers from Mason\nCounty, Washington who beat “citizens in four incidents”\ndue to “the lack of training.” Id. at 138–39. The other case,\n\f14 UNITED STATES V. TOWN OF COLORADO CITY\n\nwhich arose from Goldsboro, North Carolina, involved “a\nyoung black man who was strangled to death by city police\nofficers.” Id. at 139. Congress did not suggest that either\nincident arose from an official policy or custom.\nNonetheless, Colorado City contends that this history—\nalthough history of a different and superseded law—\ndemonstrates that Congress did not intend to impose\nrespondeat superior liability under § 12601.\n\n We disagree. At best, the PAA’s legislative history\nsupports the argument that Congress passed § 12601’s\npredecessor bill to allow the United States to prosecute\nmunicipalities when local police departments violate\nconstitutional rights—whether or not those violations arose\nfrom an official policy or custom. Our construction of the\nstatute, which allows local governments to be held liable\nwhen their agents engage in a pattern or practice of conduct\nthat deprives persons of their constitutional rights, accords\nwith that statutory purpose.\n\n When interpreting legislation, our role “is to apply the\nstatute as it is written—even if we think some other approach\nmight ‘accor[d] with good policy.’” Burrage v. United\nStates, 571 U.S. 204, 218 (2014) (alteration in original)\n(quoting Comm’r v. Lundy, 516 U.S. 235, 252 (1996)).\nSection 12601 provides a civil cause of action to the United\nStates Attorney General when a local government’s agents\n“engage in a pattern or practice of conduct . . . that deprives\npersons of rights, privileges, or immunities secured or\nprotected by the Constitution or laws of the United States.”\n34 U.S.C. § 12601. Because the statutory language does not\ndemonstrate that Congress intended to exclude local\ngovernments from respondeat superior liability, we hold\nthat § 12601 imposes liability based on general agency\n\f UNITED STATES V. TOWN OF COLORADO CITY 15\n\nprinciples. Accordingly, the district court did not err in its\nconstruction of § 12601. 5\n\nII. Factual Findings\n\n Colorado City next argues that the district court made\nthree mistakes in its factual findings related to the Towns’\nalleged violations of the Fourth Amendment. First, it argues\nthat the court erroneously included legal conclusions in its\nfindings of fact, in violation of Federal Rule of Civil\nProcedure 52(a). Second, it contends that the district court\nimproperly adopted the government’s proposed findings of\nfact and failed to independently evaluate the evidence.\nThird, it argues that the court improperly made summary\nfindings in its written judgment.\n\n We need not address these alleged errors because they\ndo not affect the district court’s holding that Colorado City\nviolated § 12601. As the district court stated in its judgment,\na finding “of a pattern or practice of violating any one of the\nthree constitutional amendments in question”—the First,\nFourth, and Fourteenth Amendments—“would entitle [the\nUnited States] to relief.” The district court held that the\nTowns violated all three constitutional amendments.\n\n 5\n We reject Colorado City’s argument that our interpretation of\n34 U.S.C. § 12601 violates § 5 of the Fourteenth Amendment. It argues\nthat the district court violated the Supreme Court’s decision in City of\nCanton v. Harris because it “interpreted [§ 12601] to impose vicarious\nliability upon the Towns without requiring the United States to show that\nthe Towns were responsible for the alleged misconduct.” 489 U.S. 378\n(1989). Not so. The Court’s decision in City of Canton, which stated\nthat permitting respondeat superior liability against local governments\nunder § 1983 “would implicate serious questions of federalism,” was\nlimited to that statute. Id. at 392. The Court did not hold that it was\nunconstitutional to permit respondeat superior liability against local\ngovernments in any context.\n\f16 UNITED STATES V. TOWN OF COLORADO CITY\n\nBecause Colorado City’s arguments relate only to the court’s\nfindings of fact as to the Fourth Amendment violation, the\ncourt’s judgment stands even if Colorado City is correct.\nAccordingly, any such purported error by the district court\nwas harmless.\n\n Colorado City urges us to nevertheless address the merits\nof its argument because “the district court’s erroneous\nfactual findings exposed the Towns to unfounded liability\nthrough lawsuits that Isaac Wyler, Patrick Pipkin, and\nAndrew Chatwin filed for unlawful arrest.” Those lawsuits,\nhowever, are irrelevant to our harmless error analysis\nbecause the relevant question is “whether the [] verdict\nactually rendered in this trial” was attributable to the district\ncourt’s error. Sullivan v. Louisiana, 508 U.S. 275, 279\n(1993). Thus, the fact that other parties may have sued the\nTowns does not affect our conclusion.\n\nIII. Admission of FLDS Leaders’ Statements\n\n Finally, we turn to Colorado City’s argument that the\ndistrict court erred when it admitted the statements of\nvarious FLDS leaders under the co-conspirator exception to\nthe rule against hearsay.\n\n Before trial, the United States moved to admit several\nFLDS leaders’ statements under the co-conspirator\nexception. Fed. R. Evid. 801(d)(2)(E); see Bourjaily v.\nUnited States, 483 U.S. 171, 173 (1987) (“A statement is not\nhearsay if . . . [t]he statement is offered against a party and\nis . . . a statement by a coconspirator of a party during the\ncourse and in furtherance of the conspiracy.” (alterations in\noriginal)). The district court held the United States’ motion\nin abeyance pending trial, reasoning:\n\f UNITED STATES V. TOWN OF COLORADO CITY 17\n\n Although a final decision on this point must\n await trial evidence, prior proceedings in this\n case cause the court to believe that the United\n States will succeed in producing evidence of\n a joint venture or concert of action between\n the FLDS Church and the [Towns]. Subject\n to the United States proving up its concerted\n action contention, the court is prepared to\n rule that out-of-court statements of FLDS\n leaders in furtherance of concerted action\n between the FLDS Church and [the Towns]\n [are] not hearsay and [are] admissible.\n\nLater during trial, the court determined that the United States\nhad satisfied its burden of establishing the existence of a\nconspiracy. It instructed the jury that “the [United States]\nhas made its case, as far as I am concerned, for purposes of\nthe admission of the testimony, that there was a\n[conspiracy].”\n\n Colorado City first argues that the district court clearly\nerred by finding the existence of a conspiracy between the\nChurch and the Towns. Its contention rests largely on the\nclaim that “[w]hen the district court ‘tentatively’ ruled that\nthe United States had established a conspiracy, the evidence\nwas insufficient to show the existence of a conspiracy for\npurposes of Rule 801(d)(2)(E).” That argument fails\nbecause “[i]t is not controlling [] whether sufficient\nindependent evidence connecting [the Church] with the\nconspiracy existed at the time the trial judge made his first\nruling under Rule 801(d)(2)(E).” United States v. Watkins,\n600 F.2d 201, 204–05 (9th Cir. 1979). “In ascertaining\nwhether the foundation has been established, we can,\ntherefore, consider all the evidence independent of the\nchallenged statements, regardless of the order of proof.”\n\f18 UNITED STATES V. TOWN OF COLORADO CITY\n\nUnited States v. Miranda-Uriarte, 649 F.2d 1345, 1351 (9th\nCir. 1981). The evidence must “be considered in a light most\nfavorable to the government.” Id.\n\n Here, the United States presented extensive evidence at\ntrial that supported the existence of a conspiracy between the\nChurch and the Towns. That evidence included testimony\nthat: officials from the Towns attended meetings in which\nFLDS leaders instructed them on how to handle legal issues\nin a way that advanced the Church’s interests; Jeffs\nexcommunicated the Towns’ leaders who did not follow his\norders; FLDS leaders determined who would occupy the\nTowns’ government positions such as mayor, city council\nmembers, and police officers; the Marshal’s Office was\nwillfully blind to FLDS members’ illegal activities;\nmembers of the Marshal’s Office helped Jeffs evade capture\nby the FBI while he was a fugitive; and several of the Towns’\nofficials “spied” on residents who the Church considered\n“out of conformance with [FLDS] regulations.” Taken\ntogether, this evidence is sufficient to establish that the\nTowns conspired with FLDS members to advance the\nChurch’s illicit objectives. The district court did not clearly\nerr in making that finding.\n\n Colorado City also argues that the district court legally\nerred by misinterpreting the co-conspirator exception to the\nrule against hearsay. That argument, however, is belied by\nthe record. The district court repeatedly acknowledged that\na statement must be made in furtherance of a conspiracy to\nqualify under Rule 801(d)(2)(E). Moreover, when\ninstructing the jury, the court accurately stated the standard\nfor statements to fall under the co-conspirator exception:\n“one, certain individuals worked with the defendants toward\na common goal in a joint venture or in a concerted effort and,\ntwo, that those individuals made out-of-court statements\n\f UNITED STATES V. TOWN OF COLORADO CITY 19\n\nduring and in furtherance of that effort.” Accordingly, we\nreject Colorado City’s argument that the court misconstrued\nthe rules of evidence.\n\n Lastly, contrary to Colorado City’s argument, the district\ncourt did not abuse its discretion by admitting over twenty\nstatements by FLDS leaders. Several of the statements,\nincluding several transcriptions of Jeffs’ dictations and\ntelephone calls, were properly admitted under the co-\nconspirator exception. These statements “catalogued and\nanalyzed factors relevant” to the alleged conspiracy. United\nStates v. Schmit, 881 F.2d 608, 612 (9th Cir. 1989). They\ninclude statements by Jeffs recounting instructions he gave\nto FLDS members to perform underage marriages,\ndescribing the appointment of FLDS members to leadership\npositions in the Towns, and stating that FLDS members had\ngone into hiding “to not be served [] legal papers.” That\n“some portions of the statement[s] may have been ‘idle\nchatter’ or [‘]casual admissions of culpability’ [does] not\nrender” the statements inadmissible. Id. That is especially\ntrue because, during trial, the Towns moved to exclude “the\nstatement[s] as a whole” rather than “particular passages in\nthe statement[s].” Id.\n\n While certain other statements admitted by the district\ncourt and challenged by Colorado City do not fall under the\nco-conspirator exception in Rule 801(d)(2)(E), we hold that\nthe district court did not err in admitting them because they\nwere otherwise admissible. See United States v. Alexander,\n48 F.3d 1477, 1487 (9th Cir. 1995) (upholding a district\ncourt's admission of evidence on a ground “different from\nthe reason given by the district court”).\n\n Several of those statements were admissible because\nthey were not hearsay. These included instructions by Jeffs\nto FLDS members to not communicate with, and to\n\f20 UNITED STATES V. TOWN OF COLORADO CITY\n\notherwise avoid, “apostates”—residents who were once\nFLDS members, but who had left the Church. See United\nStates v. Chung, 659 F.3d 815, 833 (9th Cir. 2011)\n(“Instructions to an individual to do something are . . . not\nhearsay . . . because they are declarations of fact and\ntherefore are not capable of being true or false.” (alterations\nin original) (quoting United States v. Reilly, 33 F.3d 1396,\n1410 (3d Cir. 1994))). Still others, such as Jeffs’ statement\nthat he prayed for the destruction of Arizona and Utah, were\nintroduced for their effect on the listener. See United States\nv. Payne, 944 F.2d 1458, 1472 (9th Cir. 1991).\n\n Other challenged statements, such as Jeffs’ statement\nabout “[t]he attack of [] enemies upon [the FLDS\ncommunity],” were admissible under the business records\nexception to the hearsay rule. Fed. R. Evid. 803(6). Despite\nthe Towns’ argument, Ranger John Nick Hanna was\nqualified to testify about Jeffs’ dictations. Rule 803(6) “only\nrequires [testimony by] ‘someone with knowledge’ about the\nrecord-keeping, not necessarily . . . someone with\nknowledge about how the reports were made or maintained,”\nABS Entm’t, Inc. v. CBS Corp., 908 F.3d 405, 426 (9th Cir.\n2018), and Hanna had spent over four years studying the\ndocuments about which he testified. Colorado City has not\nnot shown that the dictations do not meet the trustworthiness\nstandard of the business records exception.\n\n Even if the district court erroneously admitted some\nhearsay statements, reversal is not warranted. Because the\njury rendered only an advisory verdict on the United States’\n§ 12601 claim, our review is limited to “the findings of the\ncourt as if there had been no verdict from an advisory jury.”\nAshland v. Ling-Temco-Vought, Inc., 711 F.2d 1431, 1438\n(9th Cir. 1983) (quoting 9 Charles Alan Wright & Arthur R.\nMiller, Federal Practice and Procedure § 2335 (3d ed.\n\f UNITED STATES V. TOWN OF COLORADO CITY 21\n\n1971)). Because the judge ultimately ruled on the United\nStates’ § 12601 claim, the court “ha[d] discretion to receive\nevidence that might be inadmissible in a jury trial.”\nHollinger v. United States, 651 F.2d 636, 640 (9th Cir.\n1981). We conclude that because of the overwhelming\nevidence that Colorado City deprived non-FLDS residents\nof their constitutional rights, “it is more probable than not”\nthat the court would have reached the same verdict on the\nUnited States’ § 12601 claim even if the challenged\nstatements had been excluded. Nationwide Life Ins. Co. v.\nRichards, 541 F.3d 903, 911 (9th Cir. 2008).\n\n CONCLUSION\n\n In holding that the Towns violated § 12601, the district\ncourt correctly interpreted the statute and did not err in\nadmitting the statements of FLDS leaders. We need not\naddress Colorado City’s arguments about the district court’s\nFourth Amendment-related factual findings because, even if\nthose arguments are correct, the error was harmless.\n\n AFFIRMED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Town of Colorado City\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-26\\nOpinion type: majority\\nHas dissenting opinion: No\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\n UNITED STATES OF AMERICA, No. 17-16472\\n Plaintiff-Appellee,\\n D.C. No.\\n v. 3:12-cv-08123-\\n HRH\\n TOWN OF COLORADO CITY,\\n ARIZONA; TWIN CITY WATER\\n AUTHORITY, INC., OPINION\\n Defendants-Appellants.\\n\\n Appeal from the United States District Court\\n for the District of Arizona\\n H. Russel Holland, District Judge, Presiding\\n\\n Argued and Submitted April 18, 2019\\n San Francisco, California\\n\\n Filed August 26, 2019\\n\\n Before: MICHAEL DALY HAWKINS and MILAN D.\\n SMITH, JR., Circuit Judges, and BARBARA M. G.\\n LYNN, * District Judge.\\n\\n Opinion by Judge Milan D. Smith, Jr.\\n\\n\\n *\\n The Honorable Barbara M. G. Lynn, Chief United States District\\nJudge for the Northern District of Texas, sitting by designation.\\n\\f2 UNITED STATES V. TOWN OF COLORADO CITY\\n\\n SUMMARY**\\n\\n\\n Civil Rights\\n\\n The panel affirmed the district court’s judgment in favor\\nof the United States in its action against the Town of\\nColorado City, Arizona brought under the Violent Crime\\nControl and Law Enforcement Act of 1994, 34 U.S.C.\\n§ 12601, which prohibits any governmental authority from\\nengaging in a pattern or practice of conduct by law\\nenforcement officers or government agents that deprives\\npersons of rights, privileges, or immunities secured or\\nprotected by the Constitution or laws of the United States.\\n\\n The United States brought a civil action against the\\nmunicipal defendants and their utility providers alleging a\\npattern or practice of discrimination against residents who\\nwere not members of Fundamentalist Church of Jesus Christ\\nof Latter-Day Saints. The essential allegation of the United\\nStates was that defendants functioned as an arm of the\\nChurch and conspired with Church leaders to use municipal\\nresources to advance Church interests.\\n\\n The panel held that, in holding that defendants violated\\n§ 12601, the district court correctly interpreted the statute to\\nallow for respondeat superior liability. The panel rejected\\nthe assertion that § 12601 requires the United States to\\ndemonstrate that the Towns instituted an official municipal\\npolicy of violating residents’ constitutional rights and\\ntherefore declined to extend the holding in Monell v.\\nDepartment of Social Services, 436 U.S. 658 (1978) to\\nclaims pursuant to § 12601. The panel held that, had\\nCongress wished to eliminate respondeat superior liability\\nunder § 12601, it could have easily done so with explicit\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 3\\n\\nstatutory language. Its decision not to do so suggested that\\nit intended for § 12601, like most civil rights statutes, to\\nallow for respondeat superior liability.\\n\\n The panel held that it was not necessary to address\\nColorado City’s arguments about the district court’s Fourth\\nAmendment-related factual findings because, even if those\\narguments were correct, the error was harmless. The panel\\nfurther held that the district court did not err in admitting the\\nstatements of Church leaders under the co-conspirator\\nexception to the rule against hearsay. The panel held that\\ntaken together, the evidence was sufficient to establish that\\ndefendants conspired with Church members to advance the\\nChurch’s illicit objectives. While certain other statements\\nadmitted by the district court did not fall under the co-\\nconspirator exception, the district court did not err in\\nadmitting them because they were otherwise admissible.\\nThe panel concluded that because of the overwhelming\\nevidence that Colorado City deprived non-Church residents\\nof their constitutional rights, it was more probable than not\\nthat the court would have reached the same verdict on the\\nUnited States’ § 12601 claim even if the challenged\\nstatements had been excluded.\\n\\n\\n COUNSEL\\n\\nJeffrey C. Matura (argued) and Melissa England, Barrett &\\nMatura P.C., Scottsdale, Arizona; R. Blake Hamilton,\\nDurham Jones & Pinegar P.C., Salt Lake City, Utah; for\\nDefendants-Appellants.\\n\\f4 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nChristine A. Monta (argued) and Thomas E. Chandler,\\nAttorneys; John M. Gore, Acting Assistant Attorney\\nGeneral; Appellate Section, Civil Rights Division, United\\nStates Department of Justice, Washington, D.C.; for\\nPlaintiff-Appellee.\\n\\n\\n OPINION\\n\\nM. SMITH, Circuit Judge:\\n\\n When the United States suspected the Town of Colorado\\nCity, Arizona (Colorado City) and Hildale City, Utah\\n(collectively the Towns) of engaging in a pattern or practice\\nof violating the constitutional rights of residents who were\\nnot members of the Fundamentalist Church of Jesus Christ\\nof Latter-Day Saints (FLDS or the Church), it sued the\\nTowns pursuant to the Violent Crime Control and Law\\nEnforcement Act of 1994, 34 U.S.C. § 12601 (formerly\\ncodified at 42 U.S.C. § 14141). 1 After a 44-day trial, a jury\\nreturned an advisory verdict finding the Towns liable. The\\ndistrict court handed down a judgment holding that the\\nTowns violated § 12601, and granted injunctive relief\\nagainst the Towns.\\n\\n Colorado City2 appeals the district court’s decision on\\nthree grounds, all of which fail. The district court correctly\\ninterpreted § 12601 when it concluded that the statute does\\n\\n 1\\n The United States also sued the Towns pursuant to the Fair\\nHousing Act, 42 U.S.C. § 3601 et seq., but this appeal does not concern\\nthat claim.\\n 2\\n Although Hildale City also appealed the district court’s decision,\\nit has since withdrawn from this proceeding.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 5\\n\\nnot require an official municipal policy of violating\\nconstitutional rights in order for the United States to prevail.\\nColorado City’s arguments about the district court’s factual\\nfindings, even if they are correct, do not entitle it to relief\\nbecause the district court’s judgment is supported on other\\ngrounds. The district court did not err in admitting several\\nstatements that Colorado City contends were hearsay. We\\naffirm the district court.\\n\\n FACTUAL AND PROCEDURAL BACKGROUND\\n\\n Straddling the Utah and Arizona border, the Short Creek\\nCommunity is a religious settlement composed of the\\nTowns. Most residents are FLDS members and follow the\\nteachings of Warren Jeffs, whom they sustain as a prophet\\nand leader of the Church. Since becoming the head of the\\nChurch in 2002, Jeffs has promulgated a strict set of rules for\\nFLDS members, such as prohibitions on: vacations, toys,\\nattendance at public schools, and displays of affection\\nbetween husbands and wives.\\n\\n The United States brought a civil action against the\\nTowns and their municipal utility providers, Twin City\\nWater Authority and Twin City Power, alleging a pattern or\\npractice of discrimination against non-FLDS residents. The\\nessential allegation of the United States was that the Towns\\nfunctioned as an arm of the Church and conspired with\\nFLDS leaders to use the Towns’ municipal resources to\\nadvance Church interests. The complaint stated a claim\\nagainst the Towns pursuant to § 12601 for violating the\\nEstablishment Clause of the First Amendment, the Fourth\\nAmendment’s prohibition on unreasonable searches and\\nseizures, and the Equal Protection Clause of the Fourteenth\\nAmendment. Because § 12601 does not provide a right to a\\njury trial, the parties agreed that a jury would render only an\\nadvisory verdict.\\n\\f6 UNITED STATES V. TOWN OF COLORADO CITY\\n\\n At trial, the United States argued that FLDS leaders\\nselected the Towns’ leaders and members of the Colorado\\nCity Marshal’s Office (Marshals), which served as the police\\ndepartment for the Towns. The United States offered\\ntestimony that the FLDS “ran the [Towns’] government” and\\nthat the Towns’ government “was a part of the [C]hurch.” It\\nalso offered evidence demonstrating that FLDS leaders\\ninstructed local government officials on how to perform their\\njobs in a way that advanced the Church’s interests.\\nMarshals, for example, ignored violations of the law—such\\nas underage marriage, unlicensed drug distribution, and food\\nstamp fraud—by FLDS members.\\n\\n The Marshal’s Office worked closely with FLDS\\nleaders. Marshals helped FLDS leaders evade service of\\nprocess by the FBI, and ran computer checks of license\\nplates of unfamiliar cars, when asked to do so by FLDS\\nleaders. Cooperation between the Church and the Towns\\neven extended to sharing tangible resources. For example,\\nthe Marshal’s Office provided equipment such as tasers and\\nnight-vision binoculars to Church Security, the FLDS’s\\nprivate security force.\\n\\n The Marshal’s Office also helped Jeffs after he became\\na fugitive. Less than three years after Jeffs became head of\\nthe Church, the United States secured a warrant for his arrest\\non charges of sexual misconduct with children. The FBI\\nsought the help of the Marshal’s Office to locate Jeffs, but\\nthe Marshals did not cooperate; instead, they hindered the\\nFBI’s investigation and helped Jeffs hide for over a year.\\nThe Marshals also provided Jeffs with financial assistance\\nand information on the activities of federal law enforcement\\nto help him evade capture. The Marshals even helped\\ndestroy evidence of the crimes for which Jeffs was accused\\nby burglarizing a former FLDS member’s business.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 7\\n\\n The United States also presented evidence that members\\nof the Marshal’s Office discriminated against non-FLDS\\nresidents. It contended that the Marshals failed to provide\\neffective police protection to residents who were not FLDS\\nmembers. One non-FLDS resident testified, for example,\\nthat a Marshal drove to his home, walked out of his car, and\\n“just came over and grabbed my arm and [] bent it up around\\nmy back.” Although the resident explained that he had a\\nlegal right to occupy the property and presented an\\noccupancy agreement, he was charged with trespassing.\\n\\n The jury returned an advisory verdict finding the Towns\\nliable under § 12601. After an evidentiary hearing, the\\ndistrict court issued a judgment holding the Towns liable\\nunder § 12601 for engaging in a pattern or practice of\\nviolating the First, Fourth, and Fourteenth Amendment\\nrights of their residents. The court determined that the\\nMarshal’s Office “fostered excessive government\\nentanglement with religion” in an effort to “endors[e],\\nfavor[], or promot[e] the FLDS Church at the expense of\\nnon-FLDS residents.” The court also concluded that\\nmembers of the Marshal’s Office “selectively enforce[ed]\\nthe law based upon religion” and arrested several residents\\nwho were not FLDS members without probable cause. The\\ndistrict court ordered injunctive relief requiring the Towns\\nto, among other things, work with a court-appointed monitor\\nto institute national guidelines for constitutional policing.\\n\\n Although the Towns appealed the district court’s finding\\nof liability under § 12601, Hildale City has since withdrawn\\nfrom this appeal. Accordingly, we address only Colorado\\nCity’s arguments.\\n\\n STANDARD OF REVIEW AND JURISDICTION\\n\\n We have jurisdiction pursuant to 28 U.S.C. § 1291.\\n\\f8 UNITED STATES V. TOWN OF COLORADO CITY\\n\\n Whether the district court correctly interpreted 34 U.S.C.\\n§ 12601 is a legal question that we review de novo.\\nGoTo.com, Inc. v. Walt Disney Co., 202 F.3d 1199, 1204\\n(9th Cir. 2000). We review the court’s factual findings for\\nclear error. United States v. Christensen, 828 F.3d 763, 815\\n(9th Cir. 2015). We review de novo the court’s\\ninterpretation of the hearsay rule, but review the court’s\\nadmission of evidence for abuse of discretion. United States\\nv. Morales, 720 F.3d 1194, 1199 (9th Cir. 2013).\\n\\n ANALYSIS\\n\\nI. 34 U.S.C. § 12601\\n\\n The principal dispute in this case concerns the proper\\ninterpretation of 34 U.S.C. § 12601. That statute prohibits\\n\\n any governmental authority, or any agent\\n thereof, or any person acting on behalf of a\\n governmental authority, [from engaging] in a\\n pattern or practice of conduct by law\\n enforcement officers or by officials or\\n employees of any governmental agency . . .\\n that deprives persons of rights, privileges, or\\n immunities secured or protected by the\\n Constitution or laws of the United States.\\n\\n Colorado City argues that the district court erred by\\nconstruing the statute as imposing liability on governments\\nfor patterns of constitutional violations committed by their\\nofficers and agents. It asserts that § 12601 requires the\\nUnited States to demonstrate that the Towns “instituted an\\nofficial municipal policy” of violating residents’\\nconstitutional rights. The United States, on the other hand,\\ncontends that the statute “imposes liability on municipalities\\nfor patterns of constitutional violations [that] their law\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 9\\n\\nenforcement officers commit, without requiring an\\nadditional showing that the municipality’s policy or custom\\ncaused those violations.” This issue—whether § 12601\\nimposes respondeat superior liability3—is one of first\\nimpression in our circuit. Cf. United States v. County. of\\nMaricopa, 889 F.3d 648, 653 (9th Cir. 2018) (finding it\\nunnecessary to decide “whether the language of § 12601\\nimposes liability on the basis of general agency principles”).\\n\\n Colorado City relies on the premise that, by including\\n“pattern or practice” in § 12601, Congress used “language\\nwith a well-defined meaning [] developed under [Monell v.\\nDepartment of Social Services, 436 U.S. 658 (1978)] for\\nmunicipal liability.” That contention, however, confuses the\\nrelationship between general liability rules in civil rights\\nstatutes and the Supreme Court’s decision in Monell.\\n\\n “[T]he general rule regarding actions under civil rights\\nstatutes is that respondeat superior applies.” Bonner v.\\nLewis, 857 F.2d 559, 566 (9th Cir. 1988). In Monell, the\\nCourt carved out an exception to this general rule by holding\\nthat a municipality may not be held liable pursuant to\\n42 U.S.C. § 1983 for the actions of its subordinates. Instead,\\nto establish municipal liability, a plaintiff must show that a\\nlocal government’s “policy or custom” led to the plaintiff’s\\ninjury. Monell, 436 U.S. at 694. In reaching its holding, the\\nCourt relied on “the language of § 1983, read against the\\nbackground of the [statute’s] legislative history.” Id. at 691.\\nBecause § 1983 imposes liability only where a state actor,\\n“under color of some official policy, ‘causes’ an employee\\n\\n 3\\n Respondeat superior is “[t]he doctrine holding an employer or\\nprincipal liable for the employee’s or agent’s wrongful acts committed\\nwithin the scope of employment or agency.” Black’s Law Dictionary\\n(11th ed. 2019).\\n\\f10 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nto violate another’s constitutional rights,” the Court\\nreasoned that Congress did not intend to impose vicarious\\nliability on municipalities “solely on the basis of the\\nexistence of an employer-employee relationship with a\\ntortfeasor.” Id. at 692. Moreover, in the Civil Rights Act of\\n1871—the predecessor statute to § 1983—Congress “did not\\nintend municipalities to be held liable unless action pursuant\\nto official municipal policy of some nature caused a\\nconstitutional tort.” Id. at 691.\\n\\n Monell’s holding remains the exception to the general\\nrule. 4 We have declined to bar respondeat superior liability\\nin other contexts. In Bonner, for example, we held that\\nrespondeat superior liability applies to claims pursuant to\\n§ 504 of the Rehabilitation Act of 1973 because “[t]he\\napplication of respondeat superior . . . [is] entirely\\nconsistent with the policy of that statute, which is to\\neliminate discrimination against the handicapped.” 857 F.2d\\nat 566–67 (quoting Patton v. Cumpson, 498 F. Supp. 933,\\n943 (S.D.N.Y. 1980)). And, in Duvall v. County of Kitsap,\\nwe held that respondeat superior liability applies to claims\\nbrought pursuant to Title II of the Americans with\\nDisabilities Act, 42 U.S.C. § 12132. 260 F.3d 1124, 1141\\n(9th Cir. 2001).\\n\\n We likewise decline to extend Monell’s holding to\\nclaims pursuant to § 12601. Several features of the statutory\\ntext lead us to that conclusion. See Esquivel-Quintana v.\\n\\n\\n 4\\n The Supreme Court has held that respondeat superior liability is\\nalso unavailable against local governments pursuant to 42 U.S.C. § 1981.\\nJett v. Dallas Indep. Sch. Dist., 491 U.S. 701, 733 (1989). That decision\\nwas based on the fact that “the express cause of action for damages\\ncreated by § 1983 constitutes the exclusive federal remedy for violation\\nof the rights guaranteed in § 1981.” Id.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 11\\n\\nSessions, 137 S. Ct. 1562, 1568 (2017) (“We begin, as\\nalways, with the text.”).\\n\\n First, § 12601, unlike § 1983, does not include the words\\n“under color of any law, statute, ordinance, regulation,\\ncustom or usage.” That difference is important because, by\\nincluding “custom” in § 1983, Congress expressly\\ncontemplated imposing liability on actors who violated\\nconstitutional rights under an official policy. The absence of\\nthat language from § 12601, therefore, suggests that\\nCongress did not intend to limit liability to those acting\\nunder an official law or policy. Instead, the plain text of\\n§ 12601 shows that any government agent who engages in a\\npattern or practice of conduct that deprives persons of their\\nconstitutional rights violates § 12601.\\n\\n Second, § 12601 does not limit liability to those who\\n“cause [citizens or persons] to be subjected” to a deprivation\\nof their constitutional rights. The Monell Court interpreted\\nthat language, which appears in § 1983, as imposing liability\\n“on a government that, under color of some official policy,\\n‘causes’ an employee to violate another’s constitutional\\nrights.” Monell, 436 U.S. at 692. The lack of that causal\\nphrase in § 12601 suggests that Congress did not intend to\\nlimit local governments’ liability to situations when “the\\naction that is alleged to be unconstitutional implements or\\nexecutes a policy statement, ordinance, regulation, or\\ndecision officially adopted and promulgated by that body’s\\nofficers.” Id. at 690. Taken together, these statutory clues\\npersuade us that Congress intended to allow for respondeat\\nsuperior liability against local governments pursuant to\\n§ 12601.\\n\\n In arguing that the statutory text supports its position,\\nColorado City relies on the fact that the phrase “pattern or\\npractice” appears in both § 1983 and § 12601. That phrase,\\n\\f12 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nit claims, “refers to the same language necessary to show a\\n‘custom’ under Monell.”\\n\\n We acknowledge that Congress used “pattern or\\npractice” in both statutes, and are mindful that “[a] basic\\nprinciple of interpretation is that courts ought to interpret\\nsimilar language in the same way, unless context indicates\\nthat they should do otherwise.” Shirk v. United States ex rel.\\nDep’t of Interior, 773 F.3d 999, 1004 (9th Cir. 2014). That\\nprinciple, however, does not necessarily support Colorado\\nCity’s argument, for Congress has also used “pattern or\\npractice” literally, rather than as a term of art, in several\\nstatutes. See, e.g., 42 U.S.C. § 2000e-6(a) (authorizing the\\nAttorney General to pursue injunctive relief in cases alleging\\na pattern or practice of employment discrimination);\\n42 U.S.C. § 3614(a) (authorizing the Attorney General to\\nbring civil action in cases involving a pattern or practice of\\nFair Housing Act violations); 42 U.S.C. § 10101(e)\\n(authorizing courts to find a pattern or practice of voting\\nrights deprivations). Under those statutes, the United States\\nmust demonstrate only that the conduct alleged “was not an\\nisolated or accidental or peculiar event.” United States v.\\nIronworkers Local 86, 443 F.2d 544, 552 (9th Cir. 1971). It\\nneed not show the existence of an official policy or custom.\\n\\n For this reason, Congress’s use of “pattern or practice”\\nin § 12601 does not support the weight that Colorado City\\nwishes to place upon it. Congress could have used the phrase\\nto refer to an official policy or custom, as in § 1983, but it\\nalso could have used the phrase to refer to a regular event, as\\nin the statutes cited above.\\n\\n Our interpretation of the statute aligns with our\\nrecognition that although “[§] 12601 shares important\\nsimilarities with § 1983[,] . . . . the language of § 12601 goes\\neven further than § 1983.” County of Maricopa, 889 F.3d\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 13\\n\\nat 653. Had Congress wished to eliminate respondeat\\nsuperior liability under § 12601, it could have easily done so\\nwith explicit statutory language. See McNary v. Haitian\\nRefugee Ctr., Inc., 498 U.S. 479, 494 (1991). Its decision\\nnot to do so suggests that it intended for § 12601, like most\\ncivil rights statutes, to allow for respondeat superior\\nliability.\\n\\n Unable to muster support for its position in the statutory\\ntext, Colorado City urges us to examine § 12601’s legislative\\nhistory. But the Supreme Court has admonished that\\n“legislative history is not the law.” Epic Sys. Corp. v. Lewis,\\n138 S. Ct. 1612, 1631 (2018). That principle is particularly\\nsalient in a case where the legislative history “is virtually\\nnon-existent.” Marshall Miller, Police Brutality, 17 Yale L.\\n& Pol’y Rev. 149, 167 (1998); cf. Azar v. Allina Health\\nServs., 139 S. Ct. 1804, 1814 (2019) (declining to accord\\nweight to the Medicare Act’s legislative history when “the\\nlegislative history presented . . . is ambiguous at best”).\\n\\n Even if this case warranted consideration of it, the\\nlegislative history that Colorado City relies on does not\\nsupport its argument. Colorado City focuses on the\\nlegislative history of § 12601’s predecessor bill, the Police\\nAccountability Act of 1991 (PAA). H.R. 2972, 102d Cong.\\n(1991). That history suggests that Congress enacted the\\nPAA to “close [a] gap in the law”—the Justice Department’s\\ninability “to address systematic patterns or practices of\\npolice misconduct.” H.R. Rep. No. 102-242, pt. 1, at 137,\\n138 (1991). The House Judiciary Committee did not define\\n“pattern or practice,” but did mention “[t]wo cases [that]\\nillustrate both the need for this authority and how it will\\nwork.” Id. One case involved police officers from Mason\\nCounty, Washington who beat “citizens in four incidents”\\ndue to “the lack of training.” Id. at 138–39. The other case,\\n\\f14 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nwhich arose from Goldsboro, North Carolina, involved “a\\nyoung black man who was strangled to death by city police\\nofficers.” Id. at 139. Congress did not suggest that either\\nincident arose from an official policy or custom.\\nNonetheless, Colorado City contends that this history—\\nalthough history of a different and superseded law—\\ndemonstrates that Congress did not intend to impose\\nrespondeat superior liability under § 12601.\\n\\n We disagree. At best, the PAA’s legislative history\\nsupports the argument that Congress passed § 12601’s\\npredecessor bill to allow the United States to prosecute\\nmunicipalities when local police departments violate\\nconstitutional rights—whether or not those violations arose\\nfrom an official policy or custom. Our construction of the\\nstatute, which allows local governments to be held liable\\nwhen their agents engage in a pattern or practice of conduct\\nthat deprives persons of their constitutional rights, accords\\nwith that statutory purpose.\\n\\n When interpreting legislation, our role “is to apply the\\nstatute as it is written—even if we think some other approach\\nmight ‘accor[d] with good policy.’” Burrage v. United\\nStates, 571 U.S. 204, 218 (2014) (alteration in original)\\n(quoting Comm’r v. Lundy, 516 U.S. 235, 252 (1996)).\\nSection 12601 provides a civil cause of action to the United\\nStates Attorney General when a local government’s agents\\n“engage in a pattern or practice of conduct . . . that deprives\\npersons of rights, privileges, or immunities secured or\\nprotected by the Constitution or laws of the United States.”\\n34 U.S.C. § 12601. Because the statutory language does not\\ndemonstrate that Congress intended to exclude local\\ngovernments from respondeat superior liability, we hold\\nthat § 12601 imposes liability based on general agency\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 15\\n\\nprinciples. Accordingly, the district court did not err in its\\nconstruction of § 12601. 5\\n\\nII. Factual Findings\\n\\n Colorado City next argues that the district court made\\nthree mistakes in its factual findings related to the Towns’\\nalleged violations of the Fourth Amendment. First, it argues\\nthat the court erroneously included legal conclusions in its\\nfindings of fact, in violation of Federal Rule of Civil\\nProcedure 52(a). Second, it contends that the district court\\nimproperly adopted the government’s proposed findings of\\nfact and failed to independently evaluate the evidence.\\nThird, it argues that the court improperly made summary\\nfindings in its written judgment.\\n\\n We need not address these alleged errors because they\\ndo not affect the district court’s holding that Colorado City\\nviolated § 12601. As the district court stated in its judgment,\\na finding “of a pattern or practice of violating any one of the\\nthree constitutional amendments in question”—the First,\\nFourth, and Fourteenth Amendments—“would entitle [the\\nUnited States] to relief.” The district court held that the\\nTowns violated all three constitutional amendments.\\n\\n 5\\n We reject Colorado City’s argument that our interpretation of\\n34 U.S.C. § 12601 violates § 5 of the Fourteenth Amendment. It argues\\nthat the district court violated the Supreme Court’s decision in City of\\nCanton v. Harris because it “interpreted [§ 12601] to impose vicarious\\nliability upon the Towns without requiring the United States to show that\\nthe Towns were responsible for the alleged misconduct.” 489 U.S. 378\\n(1989). Not so. The Court’s decision in City of Canton, which stated\\nthat permitting respondeat superior liability against local governments\\nunder § 1983 “would implicate serious questions of federalism,” was\\nlimited to that statute. Id. at 392. The Court did not hold that it was\\nunconstitutional to permit respondeat superior liability against local\\ngovernments in any context.\\n\\f16 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nBecause Colorado City’s arguments relate only to the court’s\\nfindings of fact as to the Fourth Amendment violation, the\\ncourt’s judgment stands even if Colorado City is correct.\\nAccordingly, any such purported error by the district court\\nwas harmless.\\n\\n Colorado City urges us to nevertheless address the merits\\nof its argument because “the district court’s erroneous\\nfactual findings exposed the Towns to unfounded liability\\nthrough lawsuits that Isaac Wyler, Patrick Pipkin, and\\nAndrew Chatwin filed for unlawful arrest.” Those lawsuits,\\nhowever, are irrelevant to our harmless error analysis\\nbecause the relevant question is “whether the [] verdict\\nactually rendered in this trial” was attributable to the district\\ncourt’s error. Sullivan v. Louisiana, 508 U.S. 275, 279\\n(1993). Thus, the fact that other parties may have sued the\\nTowns does not affect our conclusion.\\n\\nIII. Admission of FLDS Leaders’ Statements\\n\\n Finally, we turn to Colorado City’s argument that the\\ndistrict court erred when it admitted the statements of\\nvarious FLDS leaders under the co-conspirator exception to\\nthe rule against hearsay.\\n\\n Before trial, the United States moved to admit several\\nFLDS leaders’ statements under the co-conspirator\\nexception. Fed. R. Evid. 801(d)(2)(E); see Bourjaily v.\\nUnited States, 483 U.S. 171, 173 (1987) (“A statement is not\\nhearsay if . . . [t]he statement is offered against a party and\\nis . . . a statement by a coconspirator of a party during the\\ncourse and in furtherance of the conspiracy.” (alterations in\\noriginal)). The district court held the United States’ motion\\nin abeyance pending trial, reasoning:\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 17\\n\\n Although a final decision on this point must\\n await trial evidence, prior proceedings in this\\n case cause the court to believe that the United\\n States will succeed in producing evidence of\\n a joint venture or concert of action between\\n the FLDS Church and the [Towns]. Subject\\n to the United States proving up its concerted\\n action contention, the court is prepared to\\n rule that out-of-court statements of FLDS\\n leaders in furtherance of concerted action\\n between the FLDS Church and [the Towns]\\n [are] not hearsay and [are] admissible.\\n\\nLater during trial, the court determined that the United States\\nhad satisfied its burden of establishing the existence of a\\nconspiracy. It instructed the jury that “the [United States]\\nhas made its case, as far as I am concerned, for purposes of\\nthe admission of the testimony, that there was a\\n[conspiracy].”\\n\\n Colorado City first argues that the district court clearly\\nerred by finding the existence of a conspiracy between the\\nChurch and the Towns. Its contention rests largely on the\\nclaim that “[w]hen the district court ‘tentatively’ ruled that\\nthe United States had established a conspiracy, the evidence\\nwas insufficient to show the existence of a conspiracy for\\npurposes of Rule 801(d)(2)(E).” That argument fails\\nbecause “[i]t is not controlling [] whether sufficient\\nindependent evidence connecting [the Church] with the\\nconspiracy existed at the time the trial judge made his first\\nruling under Rule 801(d)(2)(E).” United States v. Watkins,\\n600 F.2d 201, 204–05 (9th Cir. 1979). “In ascertaining\\nwhether the foundation has been established, we can,\\ntherefore, consider all the evidence independent of the\\nchallenged statements, regardless of the order of proof.”\\n\\f18 UNITED STATES V. TOWN OF COLORADO CITY\\n\\nUnited States v. Miranda-Uriarte, 649 F.2d 1345, 1351 (9th\\nCir. 1981). The evidence must “be considered in a light most\\nfavorable to the government.” Id.\\n\\n Here, the United States presented extensive evidence at\\ntrial that supported the existence of a conspiracy between the\\nChurch and the Towns. That evidence included testimony\\nthat: officials from the Towns attended meetings in which\\nFLDS leaders instructed them on how to handle legal issues\\nin a way that advanced the Church’s interests; Jeffs\\nexcommunicated the Towns’ leaders who did not follow his\\norders; FLDS leaders determined who would occupy the\\nTowns’ government positions such as mayor, city council\\nmembers, and police officers; the Marshal’s Office was\\nwillfully blind to FLDS members’ illegal activities;\\nmembers of the Marshal’s Office helped Jeffs evade capture\\nby the FBI while he was a fugitive; and several of the Towns’\\nofficials “spied” on residents who the Church considered\\n“out of conformance with [FLDS] regulations.” Taken\\ntogether, this evidence is sufficient to establish that the\\nTowns conspired with FLDS members to advance the\\nChurch’s illicit objectives. The district court did not clearly\\nerr in making that finding.\\n\\n Colorado City also argues that the district court legally\\nerred by misinterpreting the co-conspirator exception to the\\nrule against hearsay. That argument, however, is belied by\\nthe record. The district court repeatedly acknowledged that\\na statement must be made in furtherance of a conspiracy to\\nqualify under Rule 801(d)(2)(E). Moreover, when\\ninstructing the jury, the court accurately stated the standard\\nfor statements to fall under the co-conspirator exception:\\n“one, certain individuals worked with the defendants toward\\na common goal in a joint venture or in a concerted effort and,\\ntwo, that those individuals made out-of-court statements\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 19\\n\\nduring and in furtherance of that effort.” Accordingly, we\\nreject Colorado City’s argument that the court misconstrued\\nthe rules of evidence.\\n\\n Lastly, contrary to Colorado City’s argument, the district\\ncourt did not abuse its discretion by admitting over twenty\\nstatements by FLDS leaders. Several of the statements,\\nincluding several transcriptions of Jeffs’ dictations and\\ntelephone calls, were properly admitted under the co-\\nconspirator exception. These statements “catalogued and\\nanalyzed factors relevant” to the alleged conspiracy. United\\nStates v. Schmit, 881 F.2d 608, 612 (9th Cir. 1989). They\\ninclude statements by Jeffs recounting instructions he gave\\nto FLDS members to perform underage marriages,\\ndescribing the appointment of FLDS members to leadership\\npositions in the Towns, and stating that FLDS members had\\ngone into hiding “to not be served [] legal papers.” That\\n“some portions of the statement[s] may have been ‘idle\\nchatter’ or [‘]casual admissions of culpability’ [does] not\\nrender” the statements inadmissible. Id. That is especially\\ntrue because, during trial, the Towns moved to exclude “the\\nstatement[s] as a whole” rather than “particular passages in\\nthe statement[s].” Id.\\n\\n While certain other statements admitted by the district\\ncourt and challenged by Colorado City do not fall under the\\nco-conspirator exception in Rule 801(d)(2)(E), we hold that\\nthe district court did not err in admitting them because they\\nwere otherwise admissible. See United States v. Alexander,\\n48 F.3d 1477, 1487 (9th Cir. 1995) (upholding a district\\ncourt's admission of evidence on a ground “different from\\nthe reason given by the district court”).\\n\\n Several of those statements were admissible because\\nthey were not hearsay. These included instructions by Jeffs\\nto FLDS members to not communicate with, and to\\n\\f20 UNITED STATES V. TOWN OF COLORADO CITY\\n\\notherwise avoid, “apostates”—residents who were once\\nFLDS members, but who had left the Church. See United\\nStates v. Chung, 659 F.3d 815, 833 (9th Cir. 2011)\\n(“Instructions to an individual to do something are . . . not\\nhearsay . . . because they are declarations of fact and\\ntherefore are not capable of being true or false.” (alterations\\nin original) (quoting United States v. Reilly, 33 F.3d 1396,\\n1410 (3d Cir. 1994))). Still others, such as Jeffs’ statement\\nthat he prayed for the destruction of Arizona and Utah, were\\nintroduced for their effect on the listener. See United States\\nv. Payne, 944 F.2d 1458, 1472 (9th Cir. 1991).\\n\\n Other challenged statements, such as Jeffs’ statement\\nabout “[t]he attack of [] enemies upon [the FLDS\\ncommunity],” were admissible under the business records\\nexception to the hearsay rule. Fed. R. Evid. 803(6). Despite\\nthe Towns’ argument, Ranger John Nick Hanna was\\nqualified to testify about Jeffs’ dictations. Rule 803(6) “only\\nrequires [testimony by] ‘someone with knowledge’ about the\\nrecord-keeping, not necessarily . . . someone with\\nknowledge about how the reports were made or maintained,”\\nABS Entm’t, Inc. v. CBS Corp., 908 F.3d 405, 426 (9th Cir.\\n2018), and Hanna had spent over four years studying the\\ndocuments about which he testified. Colorado City has not\\nnot shown that the dictations do not meet the trustworthiness\\nstandard of the business records exception.\\n\\n Even if the district court erroneously admitted some\\nhearsay statements, reversal is not warranted. Because the\\njury rendered only an advisory verdict on the United States’\\n§ 12601 claim, our review is limited to “the findings of the\\ncourt as if there had been no verdict from an advisory jury.”\\nAshland v. Ling-Temco-Vought, Inc., 711 F.2d 1431, 1438\\n(9th Cir. 1983) (quoting 9 Charles Alan Wright & Arthur R.\\nMiller, Federal Practice and Procedure § 2335 (3d ed.\\n\\f UNITED STATES V. TOWN OF COLORADO CITY 21\\n\\n1971)). Because the judge ultimately ruled on the United\\nStates’ § 12601 claim, the court “ha[d] discretion to receive\\nevidence that might be inadmissible in a jury trial.”\\nHollinger v. United States, 651 F.2d 636, 640 (9th Cir.\\n1981). We conclude that because of the overwhelming\\nevidence that Colorado City deprived non-FLDS residents\\nof their constitutional rights, “it is more probable than not”\\nthat the court would have reached the same verdict on the\\nUnited States’ § 12601 claim even if the challenged\\nstatements had been excluded. Nationwide Life Ins. Co. v.\\nRichards, 541 F.3d 903, 911 (9th Cir. 2008).\\n\\n CONCLUSION\\n\\n In holding that the Towns violated § 12601, the district\\ncourt correctly interpreted the statute and did not err in\\nadmitting the statements of FLDS leaders. We need not\\naddress Colorado City’s arguments about the district court’s\\nFourth Amendment-related factual findings because, even if\\nthose arguments are correct, the error was harmless.\\n\\n AFFIRMED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}},{"rowIdx":4099,"cells":{"id":{"kind":"number","value":4656015,"string":"4,656,015"},"name":{"kind":"string","value":""},"name_abbreviation":{"kind":"string","value":"United States v. Davion Fitzgerald"},"decision_date":{"kind":"string","value":"2019-08-26"},"docket_number":{"kind":"string","value":"18-10116"},"court":{"kind":"string","value":"U.S. Court of Appeals for the Ninth Circuit"},"casebody":{"kind":"string","value":"{\"judges\": \"Before: William A. Fletcher, Paul J. Watford, and Andrew D. Hurwitz, Circuit Judges.\", \"parties\": \"\", \"opinions\": [{\"author\": \"PER CURIAM\", \"type\": \"010combined\", \"text\": \"FOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 18-10116\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:17-cr-00295-\\n JCM-NJK-1\\nDAVION FITZGERALD,\\n Defendant-Appellee. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Nevada\\n James C. Mahan, District Judge, Presiding\\n\\n Argued and Submitted March 12, 2019\\n San Francisco, California\\n\\n Filed August 26, 2019\\n\\n Before: William A. Fletcher, Paul J. Watford,\\n and Andrew D. Hurwitz, Circuit Judges.\\n\\n Per Curiam Opinion;\\n Dissent by Judge W. Fletcher\\n\\n2 UNITED STATES V. FITZGERALD\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n Vacating a sentence and remanding for resentencing, the\\npanel held that the defendant’s prior Nevada conviction for\\nattempted battery with substantial bodily harm in violation\\nof Nev. Rev. Stat. §§ 200.481(2)(b) and 193.330 qualifies as\\na felony conviction for a crime of violence under U.S.S.G.\\n§ 2K2.1.\\n\\n Applying United States v. Johnson, 920 F.3d 628 (9th\\nCir. 2019), and observing that the state court treated the\\ndefendant’s conviction as a felony rather than a\\nmisdemeanor, the panel rejected the defendant’s contention\\nthat the conviction is not a felony conviction because it is a\\nwobbler.\\n\\n The panel held that the defendant’s Nevada conviction\\nqualifies as a crime of violence under the elements clause of\\nU.S.S.G. § 4B1.2(a)(1). In so holding, the panel addressed\\nNevada’s definition of “substantial bodily harm,” which\\nincludes “prolonged physical pain,” and concluded that it is\\nnot evident that there’s a realistic probability that a\\ndefendant could be convicted of Nevada attempted battery\\nwith substantial bodily harm without the attempted use of\\nviolent force.\\n\\n Dissenting, Judge W. Fletcher wrote that because\\n“prolonged physical pain,” as the Nevada Supreme Court\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. FITZGERALD 3\\n\\nhas explained, may be caused by simple touching, and\\nbecause the definition of “substantial bodily harm” is\\nindivisible, attempted battery with substantial bodily harm\\nunder §§ 193.330 and 200.481 does not qualify as a crime of\\nviolence under the elements clause.\\n\\n\\n COUNSEL\\n\\nElizabeth O. White (argued), Appellate Chief; Dayle\\nElieson, United States Attorney; United States Attorney’s\\nOffice, Reno, Nevada; for Plaintiff-Appellant.\\n\\nAmy B. Cleary (argued), Assistant Federal Public Defender;\\nRene L. Valladares, Federal Public Defender; Office of the\\nFederal Public Defender, Las Vegas, Nevada; for\\nDefendant-Appellee.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n Davion Fitzgerald pleaded guilty to unlawful possession\\nof a firearm in violation of 18 U.S.C. §§ 922(g)(1) and\\n924(a)(2). At sentencing, the government requested an\\nenhancement under § 2K2.1(a)(4)(A) of the 2016\\nSentencing Guidelines, which provides for an increase to a\\nbase offense level of 20 if the defendant has a prior “felony\\nconviction of . . . a crime of violence.” The government\\nbased its request on Fitzgerald’s prior Nevada conviction for\\nattempted battery with substantial bodily harm in violation\\nof Nev. Rev. Stat. §§ 200.481(2)(b) and 193.330. The\\ndistrict court declined to apply the enhancement, concluding\\nthat Fitzgerald’s Nevada conviction qualified neither as a\\n\\n4 UNITED STATES V. FITZGERALD\\n\\n“felony conviction” nor a “crime of violence.” The\\ngovernment has appealed from the sentence imposed. We\\ndisagree with the district court on both counts, and therefore\\nvacate Fitzgerald’s sentence.\\n\\n I\\n\\n Fitzgerald first argues that his Nevada conviction is not\\na “felony conviction” because it is a “wobbler.” That is,\\nunder state law, it may be treated as either a felony or a\\nmisdemeanor. See Nev. Rev. Stat. § 193.330(1)(a)(4);\\nUnited States v. Bridgeforth, 441 F.3d 864, 870 (9th Cir.\\n2006). The Sentencing Guidelines define a “felony\\nconviction” as “a prior adult federal or state conviction for\\nan offense punishable by death or imprisonment for a term\\nexceeding one year, regardless of whether such offense is\\nspecifically designated as a felony and regardless of the\\nactual sentence imposed.” U.S.S.G. § 2K2.1 cmt. n.1.\\n“Despite this clear admonition, our binding circuit precedent\\nrequires us, where wobblers are concerned, to ignore the\\nmaximum sentence allowed by statute and instead adopt the\\ndesignation that [the State] gives to the offense.” United\\nStates v. Johnson, 920 F.3d 628, 634 n.3 (9th Cir. 2019).\\nBecause “a state court’s subsequent treatment of a wobbler\\nis controlling,” Bridgeforth, 441 F.3d at 872, we must\\nexamine how Fitzgerald was actually punished. Here, it is\\nclear that the state court treated his conviction as a felony.\\n\\n Fitzgerald argues that our precedents on this point did\\nnot survive Moncrieffe v. Holder, 569 U.S. 184 (2013), but\\nwe recently rejected that very argument. See Johnson,\\n920 F.3d at 637–38. Fitzgerald’s Nevada conviction\\ntherefore qualifies as a “felony conviction” for purposes of\\nU.S.S.G. § 2K2.1.\\n\\n UNITED STATES V. FITZGERALD 5\\n\\n II\\n\\n Fitzgerald next contends that his Nevada conviction does\\nnot qualify as a “crime of violence.” The commentary to\\n§ 2K2.1 defines “crime of violence” by cross-reference to\\n§ 4B1.2, which reads:\\n\\n (a) The term “crime of violence” means any\\n offense under federal or state law, punishable\\n by imprisonment for a term exceeding one\\n year, that—\\n\\n (1) has as an element the use, attempted use,\\n or threatened use of physical force against the\\n person of another, or\\n\\n (2) is murder, voluntary manslaughter,\\n kidnapping, aggravated assault, a forcible sex\\n offense, robbery, arson, extortion, or the use\\n or unlawful possession of a firearm described\\n in 26 U.S.C. § 5845(a) or explosive material\\n as defined in 18 U.S.C. § 841(c).\\n\\nU.S.S.G. § 4B1.2(a). The first of the numbered clauses is\\nknown as the “elements” clause, the second as the\\n“enumerated offenses” clause. “We use the categorical\\napproach to determine whether a state crime qualifies as a\\ncrime of violence for Guidelines purposes.” United States v.\\nMolinar, 881 F.3d 1064, 1067 (9th Cir. 2017). Because we\\nconclude that Fitzgerald’s conviction qualifies as a crime of\\nviolence under the elements clause, we do not address the\\nenumerated offenses clause.\\n\\n Under the elements clause, we ask whether the Nevada\\ncrime of attempted battery with substantial bodily harm “has\\nas an element the use, attempted use, or threatened use of\\n\\n6 UNITED STATES V. FITZGERALD\\n\\nphysical force against the person of another.” U.S.S.G.\\n§ 4B1.2(a)(1). Because this language is identical to that\\ninterpreted by the Supreme Court in Johnson v. United\\nStates, 559 U.S. 133, 136 (2010), we apply Johnson’s\\ndefinition of “physical force”: “violent force—that is, force\\ncapable of causing physical pain or injury to another\\nperson.” Id. at 140; see Molinar, 881 F.3d at 1068 & n.3.\\nThat threshold requires, at the very least, more than “a mere\\nunwanted touching.” Johnson, 559 U.S. at 142.\\n\\n In Nevada, a person can commit simple battery with\\nnothing more than an offensive touching. Nev. Rev. Stat.\\n§ 200.481(1)(a); Hobbs v. State, 251 P.3d 177, 179 (Nev.\\n2011). But battery with substantial bodily harm, as the name\\nsuggests, requires that the battery result in “substantial\\nbodily harm to the victim.” Nev. Rev. Stat. § 200.481(2)(b).\\nAnd attempted battery with substantial bodily harm—\\nFitzgerald’s crime of conviction—requires that the\\ndefendant act with the specific intent both to commit battery\\nand to bring about substantial bodily harm. See Nev. Rev.\\nStat. § 193.330(1); Tanksley v. State, 944 P.2d 240, 243\\n(Nev. 1997). 1\\n\\n Fitzgerald’s offense, therefore, is defined by the bodily\\ninjury the defendant intends to produce, not by the actual\\nlevel of force used. In United States v. Castleman, 572 U.S.\\n157 (2014), the Supreme Court explicitly declined to decide\\n“[w]hether or not the causation of bodily injury necessarily\\nentails violent force.” Id. at 167. But our court has held that,\\n\\n 1\\n Because “Nevada’s definition of attempt is coextensive with the\\nfederal definition,” United States v. Sarbia, 367 F.3d 1079, 1086 (9th\\nCir. 2004), there is no possibility that Nevada attempt sweeps more\\nbroadly than § 4B1.2’s reference to “attempted use . . . of physical\\nforce.”\\n\\n UNITED STATES V. FITZGERALD 7\\n\\nin general, “in the context of assault statutes, bodily injury\\nentails the use of violent, physical force.” United States v.\\nCalvillo-Palacios, 860 F.3d 1285, 1291 (9th Cir. 2017).\\n\\n Our rule is premised on a straightforward inference that\\nit takes Johnson-level force to produce bodily injury. See id.\\nat 1290. However, as we acknowledged in Calvillo-\\nPalacios, the validity of that inference depends on how a\\nstate defines “bodily injury” or, in this case, “substantial\\nbodily harm.” See id. at 1291–92. To take an extreme\\nexample, a state statute that defined bodily injury as merely\\n“an offensive touching” would not require violent force\\nunder Johnson. 559 U.S. at 142.\\n\\n Fitzgerald argues that we are dealing with such a statute\\nin this case. Nevada defines “substantial bodily harm” as\\neither “(1) [b]odily injury which creates a substantial risk of\\ndeath or which causes serious, permanent disfigurement or\\nprotracted loss or impairment of the function of any bodily\\nmember or organ; or (2) [p]rolonged physical pain.” Nev.\\nRev. Stat. § 0.060. The government does not argue that the\\nstatute is divisible, and Fitzgerald does not contest that a\\nconviction under the first definition would necessarily\\nrequire the use of violent force. So only the second\\ndefinition—“prolonged physical pain”—matters for our\\npurposes.\\n\\n In Collins v. State, 203 P.3d 90 (Nev. 2009), the Supreme\\nCourt of Nevada considered the meaning of the phrase\\n“prolonged physical pain.” In sustaining the statute against\\na vagueness challenge, the court held that “[t]he term ‘pain’\\nhas multiple meanings, ranging from mild discomfort or dull\\ndistress to acute often unbearable agony . . . and cannot be\\ndefined further.” Id. at 92 (internal quotation marks\\nomitted). For pain to be “prolonged,” the court stated, there\\nmust be “at least some physical suffering that lasts longer\\n\\n8 UNITED STATES V. FITZGERALD\\n\\nthan the pain immediately resulting from the wrongful act.”\\nId. at 93. “In a battery,” the court further explained, “the\\nwrongdoer would not be liable for ‘prolonged physical pain’\\nfor the touching itself,” but “would be liable for any lasting\\nphysical pain resulting from the touching.” Id. at 93 n.3.\\n\\n Fitzgerald argues that, because “substantial bodily harm”\\ncan mean only “mild discomfort” lasting “longer than the\\npain immediately resulting from the wrongful act,” the\\ninference that violent force is required to inflict such harm is\\nnot warranted. He points out that the Collins court used\\n“touching the skin of a person who has suffered third degree\\nburns” as an example of an act that would cause “exquisite\\npain.” Id. at 92 (citation omitted). Therefore, he argues,\\nNevada battery with substantial bodily harm can be\\ncommitted with only a mere touch (and attempted with only\\nan attempted touch), which Johnson teaches cannot count as\\nviolent force. 559 U.S. at 142.\\n\\n Supreme Court precedent, however, “requires more than\\nthe application of legal imagination to a state statute’s\\nlanguage.” Gonzales v. Duenas-Alvarez, 549 U.S. 183, 193\\n(2007). Fitzgerald “must demonstrate a realistic probability,\\nnot a theoretical possibility, that Nevada would apply its\\nstatute to conduct that falls outside the definition of a crime\\nof violence.” United States v. Guizar-Rodriguez, 900 F.3d\\n1044, 1052 (9th Cir. 2018) (internal quotation marks\\nomitted). Ordinarily, a defendant “must at least point to his\\nown case or other cases in which the state courts in fact did\\napply the statute in the overbroad manner for which he\\nargues.” Id. at 1048 (internal quotation marks omitted).\\nOtherwise, the overbreadth of the state statute must be\\n“evident from its text” or “evident from state court\\nprecedents interpreting that text.” Id. at 1052 (internal\\nquotation marks omitted).\\n\\n UNITED STATES V. FITZGERALD 9\\n\\n We do not think that Collins makes it “evident” that the\\nstatute sweeps as broadly as Fitzgerald contends. The\\ndecision neither holds nor states that substantial bodily harm\\ncan be caused by a mere touch (or by anything less than\\n“violent force”). Although Collins uses a touch as an\\nexample of an act that may cause pain, 203 P.3d at 92, it\\nnever says that the pain would count as “prolonged.” Indeed,\\nit clarifies that a batterer is not “liable for ‘prolonged\\nphysical pain’ for the touching itself,” but only “for any\\nlasting physical pain resulting from the touching.” Id. at 93\\nn.3. These statements make the answer to the question we\\nconfront here—whether a defendant could realistically be\\nconvicted of attempted battery with substantial bodily harm\\nwithout the attempted use of violent force—far from evident.\\n\\n For Fitzgerald to prevail, he would have to show that a\\ndefendant could realistically be convicted of attempted\\nbattery with substantial bodily harm for trying, with the\\nintent to cause lasting discomfort, merely to touch his victim\\n(or use other nonviolent force). That odd hypothetical\\nstrikes us as an exercise of “legal imagination.” Duenas-\\nAlvarez, 549 U.S. at 193. When someone intends to inflict\\nprolonged pain, even relatively minor pain, it is highly\\nimprobable that they would choose to do so through the use\\nof nonviolent force, which could easily fail to accomplish\\ntheir goal. And for the same reason, it is equally improbable\\nthat Nevada prosecutors would be able to secure convictions\\nfor attempted battery with substantial bodily harm in cases\\nin which the defendant tried to use only a touch or other\\nnonviolent force. Even if the possibility of such a conviction\\nis not theoretically foreclosed by Collins, that decision does\\nnot make evident that such a conviction is a “realistic\\nprobability.” Fitzgerald’s argument therefore falls short.\\n\\n * * *\\n\\n10 UNITED STATES V. FITZGERALD\\n\\n We conclude that Fitzgerald’s Nevada conviction for\\nattempted battery with substantial bodily harm qualifies as a\\nfelony conviction for a crime of violence under U.S.S.G.\\n§ 2K2.1. We therefore vacate Fitzgerald’s sentence and\\nremand for resentencing consistent with this opinion.\\n\\n VACATED AND REMANDED.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4433268/\", \"author_raw\": \"PER CURIAM\"}, {\"author\": \"W. FLETCHER. Circuit Judge, dissenting\", \"type\": \"dissent\", \"text\": \"W. FLETCHER. Circuit Judge, dissenting:\\n\\n I respectfully dissent.\\n\\n The majority holds that attempted battery with\\nsubstantial bodily harm qualifies as a “crime of violence”\\nunder the “elements clause” of U.S.S.G. § 4B1.2(a)(1)\\nbecause Nev. Rev. Stat. § 200.481 “has as an element the\\nuse, attempted use, or threatened use of physical force.”\\nMaj. Op. at 10. I disagree.\\n\\n 1. Elements Clause\\n\\n In Johnson v. United States, 559 U.S. 133, 140 (2010),\\nthe Supreme Court held that for a conviction under a state\\nstatute to qualify as a “violent felony” under the Armed\\nCareer Criminal Act (“ACCA”), the “physical force”\\nrequired under the statute must be “violent force” or “force\\ncapable of causing physical pain or injury to another\\nperson.” “The mere potential for some trivial pain or slight\\ninjury will not suffice. Rather, ‘violent’ force must be\\n‘substantial’ and ‘strong.’” United States v. Walton,\\n881 F.3d 768, 773 (9th Cir. 2018) (citing Johnson, 559 U.S.\\nat 140). Although Johnson construed the term “violent\\nfelony” under the ACCA, we have applied Johnson’s\\ndefinition of “physical force” to the elements clause of the\\n\\n UNITED STATES V. FITZGERALD 11\\n\\nphrase “crime of violence” in the Guidelines. See United\\nStates v. Molinar, 881 F.3d 1064, 1068 (9th Cir. 2017).\\n\\n Under the categorical approach, we must consider\\n“whether every violation of the [Nevada] statute necessarily\\ninvolves violent force.” Solorio-Ruiz v. Sessions, 881 F.3d\\n733, 737 (9th Cir. 2018) (emphasis in original). We look to\\nthe text of the statute and state court decisions interpreting\\nthe statute’s terms, treating “state cases examining ‘the outer\\ncontours of the conduct criminalized by the state statute’ as\\n‘particularly important.’” Id. (quoting United States v.\\nStrickland, 860 F.3d 1224, 1226–27 (9th Cir. 2017)).\\n\\n Under Nevada law, simple battery is not a “crime of\\nviolence” under Johnson’s definition of “physical force.”\\nSee United States v. Guizar-Rodriguez, 900 F.3d 1044, 1049\\n(9th Cir. 2018). As the Nevada Supreme Court has held, the\\namount of force required for simple battery in Nevada is “the\\nintentional and unwanted exertion of force upon another,\\nhowever slight.” Hobbs v. State, 251 P.3d 177, 180 (2011)\\n(holding that the act of spitting on another is a battery).\\n“[N]onharmful and nonviolent force suffices . . . [the] force\\nneed not be violent or severe and need not cause bodily pain\\nor bodily harm.” Id. at 179.\\n\\n However, Fitzgerald was not convicted of simple\\nbattery. He was convicted of attempted battery with intent\\nto cause substantial bodily harm. See Nev. Rev. Stat.\\n§§ 193.330, 200.481. Under Nevada law, “substantial\\nbodily harm” means (1) “Bodily injury which creates a\\nsubstantial risk of death or which causes serious, permanent\\ndisfigurement or protracted loss or impairment of the\\nfunction of any bodily member or organ;” or (2) “Prolonged\\nphysical pain.” Nev. Rev. Stat. § 0.060. The parties agree\\nthat the first definition of substantial bodily harm under\\nNevada law requires violent force as understood in Johnson.\\n\\n12 UNITED STATES V. FITZGERALD\\n\\nTherefore, the only question is whether the amount of force\\nrequired to cause “prolonged physical pain” always involves\\nthe violent physical force that Johnson requires.\\n\\n The Nevada Supreme Court has answered this question.\\nThe Court has told us that the amount of force required to\\ncause “prolonged physical pain” does not always involve the\\nviolent physical force Johnson requires. In Collins v. State,\\n203 P.3d 90, 92–93 (Nev. 2009), the Nevada Supreme Court\\ndefined “prolonged physical pain” as “some physical\\nsuffering or injury that lasts longer than the pain\\nimmediately resulting from the wrongful act.” The court\\nwrote that “physical pain” ranges from “mild discomfort or\\ndull distress to acute often unbearable agony. . . . the term\\n‘pain’ is necessarily subjective and cannot be defined\\nfurther.” Id. at 92 (internal quotations and citations omitted).\\nFor example, “touching the skin of a person who has\\nsuffered third-degree burns will cause exquisite pain, while\\nthe forceful striking of a gymnast in the solar plexus may\\ncause him no discomfort at all.” Id. (citing Matter of Philip\\nA., 49 N.Y.2d 198 (Ct. App. 1980)).\\n\\n Because “prolonged physical pain” may be caused by\\nsimple touching—as in the Nevada Supreme Court’s\\nexample, by touching a person suffering from third-degree\\nburns—a conviction for battery causing substantial bodily\\nharm can be sustained through “the merest touching.” See\\nJohnson, 559 U.S. at 139–43. Battery causing substantial\\nbodily harm may therefore be effectuated under Nevada law\\nby using—or attempting to use—a level of physical force\\nthat is insufficient under Johnson. Because the Nevada\\nSupreme Court has told us that every violation of Nev. Rev.\\nStat. § 200.481 does not necessarily involve violent physical\\nforce, I would hold that the statute is overbroad and does not\\ncategorically qualify as a “crime of violence” under the\\n\\n UNITED STATES V. FITZGERALD 13\\n\\nelements clause. And because the definition of “substantial\\nbodily harm” is indivisible, attempted battery with\\nsubstantial bodily harm under Nev. Rev. Stat. §§ 193.330\\nand 200.481 does not qualify as a crime of violence under\\nthe elements clause.\\n\\n The majority rejects this straightforward reading of state\\nlaw as “the application of legal imagination to [the] state\\nstatute’s language.” Maj. Op. at 8 (quoting Gonzales v.\\nDuenas-Alvarez, 549 U.S. 183, 193 (2007)). But the crime\\nI just described was not imagined or abstracted from the bare\\ntext of the statute. Instead, it comes directly from the Nevada\\nSupreme Court’s discussion of its own law. This is precisely\\nthe kind of “state case[] examin[ing] the outer contours of\\nthe conduct criminalized by the state statute” we are\\nsupposed to treat as “particularly important” in deciding\\nwhether a state crime involves the use of violent force.\\nWalton, 881 F.3d at 771–72 (quoting United States v.\\nStrickland, 860 F.3d 1224, 1226–27 (9th Cir. 2017)).\\n\\n Fitzgerald has thus shown “that a defendant could\\nrealistically be convicted of attempted battery with\\nsubstantial bodily harm for trying, with the intent to cause\\nlasting discomfort, merely to touch his victim (or use other\\nnonviolent force).” Maj. Op. at 9. I would conclude that his\\nconviction does not qualify as a crime of violence under the\\nelements clause.\\n\\n 2. Enumerated Offenses Clause\\n\\n Because the panel majority holds that attempted battery\\nwith substantial bodily harm under Nev. Rev. Stat.\\n§ 200.481 qualifies as a “crime of violence” under the\\n“elements clause” of U.S.S.G. § 4B1.2(a)(1), it did not need\\nto reach the question whether § 200.481 qualifies as a “crime\\nof violence” under the “enumerated offenses clause” of\\n\\n14 UNITED STATES V. FITZGERALD\\n\\nU.S.S.G. § 4B1.2(a)(2). Because I disagree with the panel\\nmajority on the “elements clause” question, I need to reach\\nthe “enumerated offenses” question. The question is\\nwhether a conviction under § 200.481 is a conviction for\\naggravated assault.\\n\\n Under the categorical approach, we compare the\\nelements of Nev. Rev. Stat. § 200.481 with the generic\\nfederal definition of “aggravated assault” to determine if\\nthey are a categorical match. The generic federal definition\\nof aggravated assault requires “proof of an aggravating\\nfactor.” United States v. Dominguez-Maroyoqui, 748 F.3d\\n918, 920 (9th Cir. 2014). We have held that “(1) intent to\\ncause serious bodily injury, and (2) use of a deadly weapon\\nto attempt to cause bodily injury (serious or not), are both\\ngeneric aggravating factors.” United States v. Gomez-\\nHernandez, 680 F.3d 1171, 1178 (9th Cir. 2012). It is\\nbeyond dispute that attempted battery with substantial bodily\\nharm under Nevada law does not require use of a deadly\\nweapon. The only serious question is whether “substantial\\nbodily harm” under Nevada law is broader than “serious\\nbodily injury” under generic aggravated assault. I conclude\\nthat it is.\\n\\n We determine the generic federal definition of “serious\\nbodily injury” by “survey[ing] a number of sources—\\nincluding state statutes, the Model Penal Code, federal law,\\nand criminal law treatises.” United States v. Garcia-\\nJimenez, 807 F.3d 1079, 1084 (9th Cir. 2015). “Most often,\\n‘[t]he generic definition of an offense roughly corresponds\\nto the definitions of the offense in a majority of the States’\\ncriminal codes.’” Id. (citation omitted).\\n\\n The Model Penal Code defines “serious bodily injury”\\nfor purposes of aggravated assault as “bodily injury which\\ncreates a substantial risk of death or which causes serious,\\n\\n UNITED STATES V. FITZGERALD 15\\n\\npermanent disfigurement, or protracted loss or impairment\\nof the function of any bodily member or organ.” Model\\nPenal Code § 210.0. A majority of states (approximately 31)\\nuse almost precisely this same definition.\\n\\n Only seven states (including Nevada) incorporate any\\nform of physical pain into their definitions of “substantial\\nbodily harm” or “serious bodily injury.” As noted above,\\nNevada defines “substantial bodily harm” as:\\n\\n 1. Bodily injury which creates a substantial\\n risk of death or which causes serious,\\n permanent disfigurement or protracted\\n loss or impairment of the function of any\\n bodily member or organ; or\\n\\n 2. Prolonged physical pain.\\n\\nNev. Rev. Stat. § 0.060. Nevada’s first definition is a\\ncategorical match for the Model Penal Code’s definition and\\nthe definition used in at least 31 other states. But the second\\ndefinition is not. A review of state statutes indicates that\\nonly Nevada, North Carolina, North Dakota, Ohio,\\nTennessee, Utah, and Wyoming incorporate any form of\\nphysical pain into their definitions. I therefore conclude that\\nthe generic federal definition of “serious bodily injury” does\\nnot include “prolonged physical pain” as that term is defined\\nin Nevada law. As a result, Nevada’s offense is not a\\ncategorical match for the generic federal definition of\\naggravated assault and is not a “crime of violence” under the\\nenumerated offense clause of U.S.S.G. § 4B1.2(a)(2).\\n\\n ***\\n\\n Attempted battery with substantial bodily harm under\\nNevada law is not a crime of violence under either the\\n\\n16 UNITED STATES V. FITZGERALD\\n\\nelements clause or the enumerated offenses clause. I\\nrespectfully dissent and would affirm the district court.\", \"resource_uri\": \"https://www.courtlistener.com/api/rest/v4/opinions/4433268/\", \"author_raw\": \"W. FLETCHER. Circuit Judge, dissenting\"}]}"},"judge1":{"kind":"string","value":"WILLIAM A FLETCHER"},"judge2":{"kind":"string","value":"PAUL J WATFORD"},"judge3":{"kind":"string","value":"ANDREW D HURWITZ"},"exactly_three_caps":{"kind":"number","value":1,"string":"1"},"extra_judges":{"kind":"string","value":""},"judge_designations":{"kind":"string","value":"{}"},"opinions_count":{"kind":"number","value":2,"string":"2"},"majority_count":{"kind":"number","value":0,"string":"0"},"dissent_count":{"kind":"number","value":1,"string":"1"},"concurrence_count":{"kind":"number","value":0,"string":"0"},"other_count":{"kind":"number","value":1,"string":"1"},"primary_citation":{"kind":"string","value":""},"reporter":{"kind":"string","value":""},"volume":{"kind":"null"},"resource_url":{"kind":"string","value":"https://www.courtlistener.com/api/rest/v4/clusters/4656015/"},"precedential_status":{"kind":"string","value":"Published"},"010combined_count":{"kind":"number","value":1,"string":"1"},"020lead_count":{"kind":"number","value":0,"string":"0"},"030concurrence_count":{"kind":"number","value":0,"string":"0"},"035concurrenceinpart_count":{"kind":"number","value":0,"string":"0"},"040dissent_count":{"kind":"number","value":0,"string":"0"},"year":{"kind":"number","value":2019,"string":"2,019"},"month":{"kind":"number","value":8,"string":"8"},"messages":{"kind":"list like","value":[{"content":"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\n\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\nobject that encodes the following variables:\n\n- Petitioner (name, description, and general category)\n- Respondent (name, description, and general category)\n- Disposition of Case\n- Winning Party\n- Issue Area\n- Legal Provisions\n- Decision Direction\n- Lower Court Direction\n- Declaration of Unconstitutionality\n- Method (panel, en banc, or rehearing)\n- Who Appealed (original plaintiff vs. defendant)\n- Dissent Basis (procedural vs. merits — only if dissent exists)\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\n\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\nrules as closely as possible, treating the **Court of Appeals as \"the Court\"** and assuming\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\n\nTo encode these variables, start by determining the petitioner and respondent, following the\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\n\n==================================================\n1. INPUT FORMAT\n==================================================\n\nYou will receive input in two sections:\n\n[CASE_METADATA]\nCase name: <case name as in the caption>\nCourt: <name of the Court of Appeals, e.g. \"United States Court of Appeals for the Ninth Circuit\">\nDate decided: <YYYY-MM-DD or best approximation>\nOpinion type: <e.g. \"majority\", \"per curiam\", \"on-the-merits\", \"rehearing\", \"en banc\", etc.>\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\nAdditional notes: <any other metadata or summary, if provided>\n\n[CASE_TEXT]\n<full or excerpted opinion text here>\n[END_CASE]\n\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\nsubstantive discussion in the opinion.\n\n==================================================\n2. OUTPUT FORMAT\n==================================================\n\nYou must output ONLY a single JSON object with exactly these keys:\n\n- \"petitioner\"\n- \"respondent\"\n- \"case_disposition\"\n- \"winning_party\"\n- \"issue_area\"\n- \"legal_provisions\"\n- \"decision_direction\"\n- \"lower_court_direction\"\n- \"declaration_of_unconstitutionality\"\n- \"method\"\n- \"who_appealed\"\n- \"dissent_basis\" (only if the case has a dissenting opinion)\n- \"concurrence_type\" (only if the case has a concurring opinion)\n- \"concurrence_basis\" (only if the case has a concurring opinion)\n- \"justification\"\n\nSchema:\n\n- \"petitioner\": {\n \"name\": string, // short party name from caption; usually the appellant\n \"description\": string, // role and identity, e.g. \"federal criminal defendant appealing conviction\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"respondent\": {\n \"name\": string,\n \"description\": string, // e.g. \"United States as appellee\", \"private employer\", \"federal agency\"\n \"category\": integer, // 1–7 general litigant type (see Section 3.1)\n \"category_label\": string // label for the category code\n }\n\n- \"case_disposition\": {\n \"code\": integer, // must be one of the codes below\n \"label\": string\n }\n\n- \"winning_party\": {\n \"code\": integer, // 0, 1, or 2 as defined below\n \"label\": string\n }\n\n- \"issue_area\": {\n \"code\": integer, // 1–14 from the list below\n \"label\": string // short description of the main issue area\n }\n\n- \"issue_area_alt\": {\n \"code\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\n \"label\": string\n }\n\n- \"legal_provisions\": [\n {\n \"area_code\": integer, // 1–9 (law area codes below)\n \"area_label\": string,\n \"provision_text\": string // e.g. \"Fourth Amendment, search and seizure\", \"18 U.S.C. § 924(c)\"\n },\n ...\n ]\n\n- \"decision_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"lower_court_direction\": {\n \"code\": integer, // 1, 2, or 3 as defined below\n \"label\": string\n }\n\n- \"declaration_of_unconstitutionality\": {\n \"code\": integer, // 0 or 1 (see Section 3.8)\n \"label\": string // \"no\" or \"yes\"\n }\n\n- \"method\": {\n \"code\": integer, // 1–4 (see Section 3.9)\n \"label\": string\n }\n\n- \"who_appealed\": {\n \"code\": integer, // 1–4 (see Section 3.10)\n \"label\": string\n }\n\n- \"dissent_basis\": { // ONLY include if \"Has dissenting opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"concurrence_type\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"code\": integer, // 1–3 (see Section 3.12)\n \"label\": string\n }\n\n- \"concurrence_basis\": { // ONLY include if \"Has concurring opinion: Yes\" in metadata\n \"procedural\": integer, // 0–3 (see Section 3.11)\n \"merits\": integer // 0–3 (see Section 3.11)\n }\n\n- \"justification\": string // 2–4 sentences summarizing why you chose these codes.\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\n // (c) how that combination implied the decision_direction.code you chose, and\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\n\nYour entire response must be exactly one JSON object in this schema and nothing else.\n\n==================================================\n3. VARIABLE INTERPRETATION\n==================================================\n\n3.1 Petitioner and Respondent\n\n- Use the case caption, opinion type, and text to determine who is appealing.\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\n- Respondent = the party opposing the appeal (typically the appellee).\n- The \"name\" should be a reasonably short label from the caption.\n- The \"description\" should describe their role and identity (e.g., \"federal prisoner challenging sentence\",\n \"state official in civil-rights suit\", \"private corporation defendant\", \"federal agency\").\n\nFor each party, also assign a \"category\" code using this 7-category typology:\n\n1 = private business or corporation (including partnerships, sole proprietorships)\n2 = nonprofit organization, interest group, or union (as an institutional party)\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\n4 = sub-state government (city, county, municipality, school district, local official)\n5 = state government (state, governor, state agency, state official in official capacity)\n6 = tribal government or Indian nation\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\n\nChoose the category based on who the party IS, not what kind of case it is. When a government\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\n\n\n3.2 Disposition of Case\n\nCodes (Songer-style numbering):\n\n1 = affirmed (includes modified and affirmed)\n2 = reversed\n3 = reversed and remanded\n4 = vacated and remanded\n5 = affirmed in part, reversed (or vacated) in part\n6 = affirmed in part, reversed (or vacated) in part, and remanded\n7 = vacated\n8 = petition denied or appeal dismissed\n9 = stay, petition, or motion granted\n10 = certification to or from a lower court\n11 = affirmed, vacated, and remanded (without reversal)\n0 = no disposition\n\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\n or original authority (the word \"remand\" must appear in the opinion text).\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\n or vacatur (7) of the prior judgment.\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\"no disposition\").\n\n\n3.3 Winning Party\n\nCodes:\n\n0 = no favorable disposition for petitioning party apparent\n1 = petitioning party received a favorable disposition\n2 = favorable disposition for petitioning party unclear\n\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\n- A \"favorable disposition\" is one that materially improves the petitioner's position relative to the lower court judgment\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\n regardless of the case disposition.\n\n\n3.4 Issue Area\n\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\n\n1 = Criminal Procedure\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\n\n2 = Civil Rights\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\n\n3 = First Amendment\n (e.g., freedom of speech, press, association, religion, and related expression claims)\n\n4 = Due Process\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\n\n5 = Privacy\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\n\n6 = Attorneys\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\n\n7 = Unions\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\n\n8 = Economic Activity\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\n\n9 = Judicial Power\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\n\n10 = Federalism\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\n\n11 = Interstate Relations\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\n\n12 = Federal Taxation\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\n\n13 = Miscellaneous\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\n\n14 = Private Action\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\n\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\nIf multiple areas appear, choose the one most emphasized in the reasoning.\n\nIMPORTANT — AVOID MISCELLANEOUS (13):\n\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\napproximately. Before choosing 13, reconsider the following common mappings:\n\n - Medical malpractice, products liability, personal injury, breach of contract,\n admiralty, property disputes → 14 (Private Action)\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\n - Attorneys' fees, bar discipline → 6 (Attorneys)\n - Federal preemption of state law → 10 (Federalism)\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\n\nIf you DO choose issue_area.code = 13, you MUST also fill in \"issue_area_alt\" with your\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\nset \"issue_area_alt\" to the same code and label as \"issue_area\".\n\nISSUE AREA TIEBREAKING RULES:\n\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\n\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\n a civil-rights statute was violated.\n\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\n termination of benefits, property seizure in a civil case).\n\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\n primarily about whether it has the power to hear the case — i.e., the court resolves a\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\n\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\n statute preempts state regulation, use 10. If the holding is about the substance of a\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\n use 8.\n\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\n procedural fairness or property/liberty deprivation without an underlying discrimination\n allegation.\n\n\n3.5 Legal Provisions\n\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \"legal_provisions\".\n\nLaw Area codes (SCDB varLawArea):\n\n1 = Constitution (original text)\n An Article or Section of the original Constitution (not an Amendment).\n Examples: Article I (congressional power), Article II (executive power),\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\n and Credit Clause, Contracts Clause.\n\n2 = Constitutional Amendment\n Any Amendment to the Constitution, including the Bill of Rights.\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\n punishment), Fourteenth Amendment (equal protection, due process).\n\n3 = Federal Statute\n A statute enacted by Congress, cited by title and section of the U.S. Code or\n by its popular name. This is the most common category.\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\n\n4 = Court Rules\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\n Procedure, as well as local circuit rules.\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\n (prejudicial evidence).\n\n5 = Other\n Common law doctrines, equitable principles, or other legal authorities not fitting\n the categories above.\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\n of laches, implied contracts.\n\n6 = Infrequently litigated statutes\n Federal statutes that appear rarely in appellate litigation and are not well-known\n by popular name. Use this when the statute is obscure or highly specialized.\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\n obscure provisions of the Internal Revenue Code.\n\n8 = State or local law or regulation\n A state statute, state constitutional provision, local ordinance, or state/local\n regulation. Use this when the court applies or interprets state or local law rather\n than federal law.\n Examples: state tort law, state contract law, state criminal statute, local zoning\n ordinance, state environmental regulation.\n\n9 = No Legal Provision\n The opinion does not meaningfully rely on any identifiable legal provision. Very\n rare — use only when the court's analysis is purely factual or procedural without\n citing specific legal authority.\n\nFor each entry:\n- `area_code`: choose the single best area for the provision.\n- `area_label`: the corresponding label (e.g., \"Constitution\", \"Federal Statute\").\n- `provision_text`: identify the specific provision (e.g., \"Fourth Amendment, search and seizure\",\n \"42 U.S.C. § 1983\", \"Federal Rule of Civil Procedure 12(b)(6)\", \"state tort law\").\n\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\n\nDistinguishing area codes:\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\n- Use 5 for common law doctrines that are not codified in any statute or rule.\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\n\n\n3.6 Decision Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\nCRITICAL: The codes \"conservative\" and \"liberal\" are MECHANICAL LABELS defined by the SCDB\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\nTaxation case, regardless of the policy context. A decision upholding a government regulation\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\nbusiness. Do NOT let your general understanding of politics override these rules.\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\n\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\n\nissue_area.code | benefiting side | direction\n----------------|------------------------------------------------------------|----------\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\n | regulation, or small entities against large business |\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\n10 (Federalism) | federal authority over state authority | 2 (liberal)\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\n | with one of the patterns above |\n\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\nissue_area, and LAST look up the decision_direction from the table above by matching the\nissue_area.code with the side that benefited. Do NOT reason about direction from first\nprinciples — use the table.\n\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\n\n\n3.6.1 Worked Examples\n\nThe following three examples demonstrate correct mechanical application of the direction\nlookup table, including cases where the SCDB coding may seem counterintuitive:\n\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\n\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\n\n Coding:\n issue_area.code = 12 (Federal Taxation)\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\n says \"taxpayer → 1 (conservative)\"\n\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\n codes this as conservative (1) because the taxpayer benefited.\n\n\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\n\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\n\n Coding:\n issue_area.code = 1 (Criminal Procedure)\n winning_party.code = 0 (no favorable disposition for petitioner)\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\n row for code 1 says \"prosecution → 1 (conservative)\"\n\n\nEXAMPLE C — Economic Activity (issue_area.code = 8)\n\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\n affirms, upholding the regulation as within the agency's statutory authority.\n\n Coding:\n issue_area.code = 8 (Economic Activity)\n winning_party.code = 0 (no favorable disposition for petitioner/company)\n decision_direction = 2 (liberal) ← government regulation upheld against business;\n lookup table row for code 8 says \"govt regulation → 2 (liberal)\"\n\n NOTE: Even if one might view upholding a regulation as a \"pro-business\" outcome (e.g., it\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\n over a business challenge as liberal (2).\n\n\n3.6.2 Internal Consistency Check\n\nBefore you output the JSON, you MUST run the following consistency check:\n\n1. Identify issue_area.code.\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\n4. Verify that your decision_direction.code matches the table entry.\n5. If it does not match, CORRECT decision_direction.code to match the table.\n\n\n3.7 Lower Court Direction\n\nCodes:\n\n1 = conservative\n2 = liberal\n3 = unspecifiable\n\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\ncourt's judgment and the appellate court's disposition.\n\nUse this rule of thumb:\n\n1. First, identify the appellate disposition (case_disposition.code).\n2. Next, identify the decision_direction.code for the Court of Appeals.\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\n\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \"affirmed\"\n or 8 \"petition denied or appeal dismissed\"), then the lower court's direction is the SAME as decision_direction.code.\n\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \"reversed\",\n 3 \"reversed and remanded\", 4 \"vacated and remanded\", or 7 \"vacated\"), then the lower court's direction is the\n OPPOSITE of decision_direction.code.\n\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\n the lower court's direction should usually be the SAME as decision_direction.code.\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\n\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\n the lower court if possible; if not, use 3 (unspecifiable).\n\nWhen you set lower_court_direction, remember:\n- \"SAME\" means lower_court_direction.code = decision_direction.code.\n- \"OPPOSITE\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\n\n3.8 Declaration of Unconstitutionality\n\nCodes:\n\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\n1 = the court DID declare a statute, regulation, or government action unconstitutional\n\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\nregulations, state statutes, state constitutional provisions, and local ordinances.\n\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\nhold that something is unconstitutional. If the court upholds the constitutionality of the\nchallenged provision, code 0.\n\n\n3.9 Method\n\nCodes:\n\n1 = standard panel decision (typically 3 judges, first hearing)\n2 = panel rehearing (same panel reconsidering its prior decision)\n3 = en banc (full court or en banc panel rehearing a case)\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\n\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\nstatements like \"before the court en banc,\" \"on rehearing,\" \"upon reconsideration,\" or\n\"per curiam.\" If the metadata indicates more than 3 judges participated, this strongly\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\n\n\n3.10 Who Appealed\n\nCodes:\n\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\n3 = federal government on behalf of the original plaintiff\n4 = federal government on behalf of the original defendant\n\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\nhistory. The \"original plaintiff\" is the party who filed suit in the lower court; the \"original\ndefendant\" is the party who was sued. In criminal cases, the original plaintiff is typically the\ngovernment (prosecution) and the original defendant is the accused.\n\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\nordering (appellant is typically listed first).\n\n\n3.11 Dissent/Concurrence Basis (Conditional)\n\nIf the case metadata states \"Has dissenting opinion: Yes\" or \"Has concurring opinion: Yes\",\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\n\nFor each dimension (procedural / merits), use one of these codes:\n\n0 = the dissent/concurrence exists but does NOT address this type of issue\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\n3 = addresses an issue NOT discussed by the majority opinion\n\n\"dissent_basis\" (include ONLY if \"Has dissenting opinion: Yes\"):\n - \"procedural\": Does the dissent address procedural questions (jurisdiction, standing,\n standard of review, procedural default, timeliness, etc.)?\n - \"merits\": Does the dissent address the substantive legal merits (statutory interpretation,\n constitutional analysis, factual sufficiency, etc.)?\n\n\"concurrence_basis\" (include ONLY if \"Has concurring opinion: Yes\"):\n - \"procedural\": Does the concurrence address procedural questions?\n - \"merits\": Does the concurrence address the substantive legal merits?\n\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\n(the longest or most substantive).\n\nIf the metadata says \"Has dissenting opinion: No\", OMIT \"dissent_basis\" entirely from your output.\nIf the metadata says \"Has concurring opinion: No\", OMIT \"concurrence_type\" and \"concurrence_basis\" entirely from your output.\n\n\n3.12 Concurrence Type (Conditional)\n\nIf the case metadata states \"Has concurring opinion: Yes\", classify the concurrence as one of\nthe following types:\n\nCodes:\n\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\n\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\n like \"concurring in the result,\" \"concurring in the judgment,\" \"concurring in the result\n only,\" or \"I concur in the judgment but write separately because I disagree with the\n majority's analysis.\" The key indicator: the judge explicitly declines to join the majority\n opinion or distances themselves from the majority's reasoning.\n\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\n the majority opinion or only the result.\n\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\nwrites a special concurrence is qualitatively different from one where all judges agree on\nboth outcome and reasoning.\n\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\nmost substantive). If one is a regular concurrence and another is a special concurrence,\ncode based on the special concurrence (code 2), as it is more analytically informative.\n\n\n==================================================\n4. GENERAL RULES\n==================================================\n\n- Use metadata (case name, court, date, opinion type) to:\n - distinguish main merits opinions from rehearing or procedural opinions,\n - interpret how the appellate disposition relates to the prior judgment,\n - infer whether the case is criminal, civil rights, tax, etc.\n- If some information is missing, use the appropriate \"unspecifiable\" or \"no disposition\" code\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\n- Carefully follow the steps laid out in the introduction in order. In particular,\n decide who won the case (winning_party). Then, decide what the issue_area is.\n Only choose decision_direction once you have completed those steps, and follow the exact\n directions in Section 3.6.\n- In the justification, explicitly state who won, why (according to the disposition), and\n how that winner and the issue_area code together implied the direction code you chose.\n Reference the DIRECTION LOOKUP TABLE row you applied.\n- ALWAYS choose exactly one code for each coded variable where required.\n- NEVER output any keys other than the ones defined in the schema above.\n- Include \"dissent_basis\", \"concurrence_type\", and \"concurrence_basis\" ONLY when the metadata\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\n- NEVER output any numeric codes outside the permitted lists above.\n\n\n==================================================\n5. FINAL REQUIREMENT\n==================================================\n\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\n","role":"system"},{"content":"[CASE_METADATA]\nCase name: United States v. Davion Fitzgerald\nCourt: U.S. Court of Appeals for the Ninth Circuit\nDate decided: 2019-08-26\nOpinion type: majority\nHas dissenting opinion: Yes\nHas concurring opinion: No\nAdditional notes: \n\n[CASE_TEXT]\nFOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 18-10116\n Plaintiff-Appellant,\n D.C. No.\n v. 2:17-cr-00295-\n JCM-NJK-1\nDAVION FITZGERALD,\n Defendant-Appellee. OPINION\n\n Appeal from the United States District Court\n for the District of Nevada\n James C. Mahan, District Judge, Presiding\n\n Argued and Submitted March 12, 2019\n San Francisco, California\n\n Filed August 26, 2019\n\n Before: William A. Fletcher, Paul J. Watford,\n and Andrew D. Hurwitz, Circuit Judges.\n\n Per Curiam Opinion;\n Dissent by Judge W. Fletcher\n\n2 UNITED STATES V. FITZGERALD\n\n SUMMARY *\n\n\n Criminal Law\n\n Vacating a sentence and remanding for resentencing, the\npanel held that the defendant’s prior Nevada conviction for\nattempted battery with substantial bodily harm in violation\nof Nev. Rev. Stat. §§ 200.481(2)(b) and 193.330 qualifies as\na felony conviction for a crime of violence under U.S.S.G.\n§ 2K2.1.\n\n Applying United States v. Johnson, 920 F.3d 628 (9th\nCir. 2019), and observing that the state court treated the\ndefendant’s conviction as a felony rather than a\nmisdemeanor, the panel rejected the defendant’s contention\nthat the conviction is not a felony conviction because it is a\nwobbler.\n\n The panel held that the defendant’s Nevada conviction\nqualifies as a crime of violence under the elements clause of\nU.S.S.G. § 4B1.2(a)(1). In so holding, the panel addressed\nNevada’s definition of “substantial bodily harm,” which\nincludes “prolonged physical pain,” and concluded that it is\nnot evident that there’s a realistic probability that a\ndefendant could be convicted of Nevada attempted battery\nwith substantial bodily harm without the attempted use of\nviolent force.\n\n Dissenting, Judge W. Fletcher wrote that because\n“prolonged physical pain,” as the Nevada Supreme Court\n\n *\n This summary constitutes no part of the opinion of the court. It\nhas been prepared by court staff for the convenience of the reader.\n\n UNITED STATES V. FITZGERALD 3\n\nhas explained, may be caused by simple touching, and\nbecause the definition of “substantial bodily harm” is\nindivisible, attempted battery with substantial bodily harm\nunder §§ 193.330 and 200.481 does not qualify as a crime of\nviolence under the elements clause.\n\n\n COUNSEL\n\nElizabeth O. White (argued), Appellate Chief; Dayle\nElieson, United States Attorney; United States Attorney’s\nOffice, Reno, Nevada; for Plaintiff-Appellant.\n\nAmy B. Cleary (argued), Assistant Federal Public Defender;\nRene L. Valladares, Federal Public Defender; Office of the\nFederal Public Defender, Las Vegas, Nevada; for\nDefendant-Appellee.\n\n\n OPINION\n\nPER CURIAM:\n\n Davion Fitzgerald pleaded guilty to unlawful possession\nof a firearm in violation of 18 U.S.C. §§ 922(g)(1) and\n924(a)(2). At sentencing, the government requested an\nenhancement under § 2K2.1(a)(4)(A) of the 2016\nSentencing Guidelines, which provides for an increase to a\nbase offense level of 20 if the defendant has a prior “felony\nconviction of . . . a crime of violence.” The government\nbased its request on Fitzgerald’s prior Nevada conviction for\nattempted battery with substantial bodily harm in violation\nof Nev. Rev. Stat. §§ 200.481(2)(b) and 193.330. The\ndistrict court declined to apply the enhancement, concluding\nthat Fitzgerald’s Nevada conviction qualified neither as a\n\n4 UNITED STATES V. FITZGERALD\n\n“felony conviction” nor a “crime of violence.” The\ngovernment has appealed from the sentence imposed. We\ndisagree with the district court on both counts, and therefore\nvacate Fitzgerald’s sentence.\n\n I\n\n Fitzgerald first argues that his Nevada conviction is not\na “felony conviction” because it is a “wobbler.” That is,\nunder state law, it may be treated as either a felony or a\nmisdemeanor. See Nev. Rev. Stat. § 193.330(1)(a)(4);\nUnited States v. Bridgeforth, 441 F.3d 864, 870 (9th Cir.\n2006). The Sentencing Guidelines define a “felony\nconviction” as “a prior adult federal or state conviction for\nan offense punishable by death or imprisonment for a term\nexceeding one year, regardless of whether such offense is\nspecifically designated as a felony and regardless of the\nactual sentence imposed.” U.S.S.G. § 2K2.1 cmt. n.1.\n“Despite this clear admonition, our binding circuit precedent\nrequires us, where wobblers are concerned, to ignore the\nmaximum sentence allowed by statute and instead adopt the\ndesignation that [the State] gives to the offense.” United\nStates v. Johnson, 920 F.3d 628, 634 n.3 (9th Cir. 2019).\nBecause “a state court’s subsequent treatment of a wobbler\nis controlling,” Bridgeforth, 441 F.3d at 872, we must\nexamine how Fitzgerald was actually punished. Here, it is\nclear that the state court treated his conviction as a felony.\n\n Fitzgerald argues that our precedents on this point did\nnot survive Moncrieffe v. Holder, 569 U.S. 184 (2013), but\nwe recently rejected that very argument. See Johnson,\n920 F.3d at 637–38. Fitzgerald’s Nevada conviction\ntherefore qualifies as a “felony conviction” for purposes of\nU.S.S.G. § 2K2.1.\n\n UNITED STATES V. FITZGERALD 5\n\n II\n\n Fitzgerald next contends that his Nevada conviction does\nnot qualify as a “crime of violence.” The commentary to\n§ 2K2.1 defines “crime of violence” by cross-reference to\n§ 4B1.2, which reads:\n\n (a) The term “crime of violence” means any\n offense under federal or state law, punishable\n by imprisonment for a term exceeding one\n year, that—\n\n (1) has as an element the use, attempted use,\n or threatened use of physical force against the\n person of another, or\n\n (2) is murder, voluntary manslaughter,\n kidnapping, aggravated assault, a forcible sex\n offense, robbery, arson, extortion, or the use\n or unlawful possession of a firearm described\n in 26 U.S.C. § 5845(a) or explosive material\n as defined in 18 U.S.C. § 841(c).\n\nU.S.S.G. § 4B1.2(a). The first of the numbered clauses is\nknown as the “elements” clause, the second as the\n“enumerated offenses” clause. “We use the categorical\napproach to determine whether a state crime qualifies as a\ncrime of violence for Guidelines purposes.” United States v.\nMolinar, 881 F.3d 1064, 1067 (9th Cir. 2017). Because we\nconclude that Fitzgerald’s conviction qualifies as a crime of\nviolence under the elements clause, we do not address the\nenumerated offenses clause.\n\n Under the elements clause, we ask whether the Nevada\ncrime of attempted battery with substantial bodily harm “has\nas an element the use, attempted use, or threatened use of\n\n6 UNITED STATES V. FITZGERALD\n\nphysical force against the person of another.” U.S.S.G.\n§ 4B1.2(a)(1). Because this language is identical to that\ninterpreted by the Supreme Court in Johnson v. United\nStates, 559 U.S. 133, 136 (2010), we apply Johnson’s\ndefinition of “physical force”: “violent force—that is, force\ncapable of causing physical pain or injury to another\nperson.” Id. at 140; see Molinar, 881 F.3d at 1068 & n.3.\nThat threshold requires, at the very least, more than “a mere\nunwanted touching.” Johnson, 559 U.S. at 142.\n\n In Nevada, a person can commit simple battery with\nnothing more than an offensive touching. Nev. Rev. Stat.\n§ 200.481(1)(a); Hobbs v. State, 251 P.3d 177, 179 (Nev.\n2011). But battery with substantial bodily harm, as the name\nsuggests, requires that the battery result in “substantial\nbodily harm to the victim.” Nev. Rev. Stat. § 200.481(2)(b).\nAnd attempted battery with substantial bodily harm—\nFitzgerald’s crime of conviction—requires that the\ndefendant act with the specific intent both to commit battery\nand to bring about substantial bodily harm. See Nev. Rev.\nStat. § 193.330(1); Tanksley v. State, 944 P.2d 240, 243\n(Nev. 1997). 1\n\n Fitzgerald’s offense, therefore, is defined by the bodily\ninjury the defendant intends to produce, not by the actual\nlevel of force used. In United States v. Castleman, 572 U.S.\n157 (2014), the Supreme Court explicitly declined to decide\n“[w]hether or not the causation of bodily injury necessarily\nentails violent force.” Id. at 167. But our court has held that,\n\n 1\n Because “Nevada’s definition of attempt is coextensive with the\nfederal definition,” United States v. Sarbia, 367 F.3d 1079, 1086 (9th\nCir. 2004), there is no possibility that Nevada attempt sweeps more\nbroadly than § 4B1.2’s reference to “attempted use . . . of physical\nforce.”\n\n UNITED STATES V. FITZGERALD 7\n\nin general, “in the context of assault statutes, bodily injury\nentails the use of violent, physical force.” United States v.\nCalvillo-Palacios, 860 F.3d 1285, 1291 (9th Cir. 2017).\n\n Our rule is premised on a straightforward inference that\nit takes Johnson-level force to produce bodily injury. See id.\nat 1290. However, as we acknowledged in Calvillo-\nPalacios, the validity of that inference depends on how a\nstate defines “bodily injury” or, in this case, “substantial\nbodily harm.” See id. at 1291–92. To take an extreme\nexample, a state statute that defined bodily injury as merely\n“an offensive touching” would not require violent force\nunder Johnson. 559 U.S. at 142.\n\n Fitzgerald argues that we are dealing with such a statute\nin this case. Nevada defines “substantial bodily harm” as\neither “(1) [b]odily injury which creates a substantial risk of\ndeath or which causes serious, permanent disfigurement or\nprotracted loss or impairment of the function of any bodily\nmember or organ; or (2) [p]rolonged physical pain.” Nev.\nRev. Stat. § 0.060. The government does not argue that the\nstatute is divisible, and Fitzgerald does not contest that a\nconviction under the first definition would necessarily\nrequire the use of violent force. So only the second\ndefinition—“prolonged physical pain”—matters for our\npurposes.\n\n In Collins v. State, 203 P.3d 90 (Nev. 2009), the Supreme\nCourt of Nevada considered the meaning of the phrase\n“prolonged physical pain.” In sustaining the statute against\na vagueness challenge, the court held that “[t]he term ‘pain’\nhas multiple meanings, ranging from mild discomfort or dull\ndistress to acute often unbearable agony . . . and cannot be\ndefined further.” Id. at 92 (internal quotation marks\nomitted). For pain to be “prolonged,” the court stated, there\nmust be “at least some physical suffering that lasts longer\n\n8 UNITED STATES V. FITZGERALD\n\nthan the pain immediately resulting from the wrongful act.”\nId. at 93. “In a battery,” the court further explained, “the\nwrongdoer would not be liable for ‘prolonged physical pain’\nfor the touching itself,” but “would be liable for any lasting\nphysical pain resulting from the touching.” Id. at 93 n.3.\n\n Fitzgerald argues that, because “substantial bodily harm”\ncan mean only “mild discomfort” lasting “longer than the\npain immediately resulting from the wrongful act,” the\ninference that violent force is required to inflict such harm is\nnot warranted. He points out that the Collins court used\n“touching the skin of a person who has suffered third degree\nburns” as an example of an act that would cause “exquisite\npain.” Id. at 92 (citation omitted). Therefore, he argues,\nNevada battery with substantial bodily harm can be\ncommitted with only a mere touch (and attempted with only\nan attempted touch), which Johnson teaches cannot count as\nviolent force. 559 U.S. at 142.\n\n Supreme Court precedent, however, “requires more than\nthe application of legal imagination to a state statute’s\nlanguage.” Gonzales v. Duenas-Alvarez, 549 U.S. 183, 193\n(2007). Fitzgerald “must demonstrate a realistic probability,\nnot a theoretical possibility, that Nevada would apply its\nstatute to conduct that falls outside the definition of a crime\nof violence.” United States v. Guizar-Rodriguez, 900 F.3d\n1044, 1052 (9th Cir. 2018) (internal quotation marks\nomitted). Ordinarily, a defendant “must at least point to his\nown case or other cases in which the state courts in fact did\napply the statute in the overbroad manner for which he\nargues.” Id. at 1048 (internal quotation marks omitted).\nOtherwise, the overbreadth of the state statute must be\n“evident from its text” or “evident from state court\nprecedents interpreting that text.” Id. at 1052 (internal\nquotation marks omitted).\n\n UNITED STATES V. FITZGERALD 9\n\n We do not think that Collins makes it “evident” that the\nstatute sweeps as broadly as Fitzgerald contends. The\ndecision neither holds nor states that substantial bodily harm\ncan be caused by a mere touch (or by anything less than\n“violent force”). Although Collins uses a touch as an\nexample of an act that may cause pain, 203 P.3d at 92, it\nnever says that the pain would count as “prolonged.” Indeed,\nit clarifies that a batterer is not “liable for ‘prolonged\nphysical pain’ for the touching itself,” but only “for any\nlasting physical pain resulting from the touching.” Id. at 93\nn.3. These statements make the answer to the question we\nconfront here—whether a defendant could realistically be\nconvicted of attempted battery with substantial bodily harm\nwithout the attempted use of violent force—far from evident.\n\n For Fitzgerald to prevail, he would have to show that a\ndefendant could realistically be convicted of attempted\nbattery with substantial bodily harm for trying, with the\nintent to cause lasting discomfort, merely to touch his victim\n(or use other nonviolent force). That odd hypothetical\nstrikes us as an exercise of “legal imagination.” Duenas-\nAlvarez, 549 U.S. at 193. When someone intends to inflict\nprolonged pain, even relatively minor pain, it is highly\nimprobable that they would choose to do so through the use\nof nonviolent force, which could easily fail to accomplish\ntheir goal. And for the same reason, it is equally improbable\nthat Nevada prosecutors would be able to secure convictions\nfor attempted battery with substantial bodily harm in cases\nin which the defendant tried to use only a touch or other\nnonviolent force. Even if the possibility of such a conviction\nis not theoretically foreclosed by Collins, that decision does\nnot make evident that such a conviction is a “realistic\nprobability.” Fitzgerald’s argument therefore falls short.\n\n * * *\n\n10 UNITED STATES V. FITZGERALD\n\n We conclude that Fitzgerald’s Nevada conviction for\nattempted battery with substantial bodily harm qualifies as a\nfelony conviction for a crime of violence under U.S.S.G.\n§ 2K2.1. We therefore vacate Fitzgerald’s sentence and\nremand for resentencing consistent with this opinion.\n\n VACATED AND REMANDED.\n[END_CASE]","role":"user"}],"string":"[\n {\n \"content\": \"You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\\nin the codebook step by step and use the precise definitions of terms presented in the codebook below.\\n\\nYour task: Given metadata and text for a single federal appellate case, output a single JSON\\nobject that encodes the following variables:\\n\\n- Petitioner (name, description, and general category)\\n- Respondent (name, description, and general category)\\n- Disposition of Case\\n- Winning Party\\n- Issue Area\\n- Legal Provisions\\n- Decision Direction\\n- Lower Court Direction\\n- Declaration of Unconstitutionality\\n- Method (panel, en banc, or rehearing)\\n- Who Appealed (original plaintiff vs. defendant)\\n- Dissent Basis (procedural vs. merits — only if dissent exists)\\n- Concurrence Type (regular vs. special/result-only — only if concurrence exists)\\n- Concurrence Basis (procedural vs. merits — only if concurrence exists)\\n\\nYou are working with U.S. Courts of Appeals, not the Supreme Court. Apply the SCDB coding\\nrules as closely as possible, treating the **Court of Appeals as \\\"the Court\\\"** and assuming\\nthat the **lower court is, e.g., a federal district court** whose decision is being reviewed.\\n\\nTo encode these variables, start by determining the petitioner and respondent, following the\\ninstructions in Section 3.1. Then, find the disposition of the case and code it according\\nto Section 3.2. Knowing the disposition will guide your next step: determining if the petitioner\\nwas the winning party, following the directions in Section 3.3. Next, use the overall opinion content\\nto code the issue area as described in Section 3.4. This step will also inform the legal provisions\\ncoding as laid out in Section 3.5. With the winning party, disposition, and issue area fixed,\\ncarefully and precisely follow the directions in Section 3.6 to assign a decision direction for the opinion.\\nBased on that decision direction and the disposition, you can assign a lower court direction as well,\\nfollowing Section 3.7. Finally, justify your decisions in the manner described below.\\n\\n==================================================\\n1. INPUT FORMAT\\n==================================================\\n\\nYou will receive input in two sections:\\n\\n[CASE_METADATA]\\nCase name: <case name as in the caption>\\nCourt: <name of the Court of Appeals, e.g. \\\"United States Court of Appeals for the Ninth Circuit\\\">\\nDate decided: <YYYY-MM-DD or best approximation>\\nOpinion type: <e.g. \\\"majority\\\", \\\"per curiam\\\", \\\"on-the-merits\\\", \\\"rehearing\\\", \\\"en banc\\\", etc.>\\nHas dissenting opinion: <Yes or No, based on whether the input data contains a separate dissent>\\nHas concurring opinion: <Yes or No, based on whether the input data contains a separate concurrence>\\nAdditional notes: <any other metadata or summary, if provided>\\n\\n[CASE_TEXT]\\n<full or excerpted opinion text here>\\n[END_CASE]\\n\\nUse both metadata and text to code the variables. If metadata and text conflict, prefer the\\nsubstantive discussion in the opinion.\\n\\n==================================================\\n2. OUTPUT FORMAT\\n==================================================\\n\\nYou must output ONLY a single JSON object with exactly these keys:\\n\\n- \\\"petitioner\\\"\\n- \\\"respondent\\\"\\n- \\\"case_disposition\\\"\\n- \\\"winning_party\\\"\\n- \\\"issue_area\\\"\\n- \\\"legal_provisions\\\"\\n- \\\"decision_direction\\\"\\n- \\\"lower_court_direction\\\"\\n- \\\"declaration_of_unconstitutionality\\\"\\n- \\\"method\\\"\\n- \\\"who_appealed\\\"\\n- \\\"dissent_basis\\\" (only if the case has a dissenting opinion)\\n- \\\"concurrence_type\\\" (only if the case has a concurring opinion)\\n- \\\"concurrence_basis\\\" (only if the case has a concurring opinion)\\n- \\\"justification\\\"\\n\\nSchema:\\n\\n- \\\"petitioner\\\": {\\n \\\"name\\\": string, // short party name from caption; usually the appellant\\n \\\"description\\\": string, // role and identity, e.g. \\\"federal criminal defendant appealing conviction\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"respondent\\\": {\\n \\\"name\\\": string,\\n \\\"description\\\": string, // e.g. \\\"United States as appellee\\\", \\\"private employer\\\", \\\"federal agency\\\"\\n \\\"category\\\": integer, // 1–7 general litigant type (see Section 3.1)\\n \\\"category_label\\\": string // label for the category code\\n }\\n\\n- \\\"case_disposition\\\": {\\n \\\"code\\\": integer, // must be one of the codes below\\n \\\"label\\\": string\\n }\\n\\n- \\\"winning_party\\\": {\\n \\\"code\\\": integer, // 0, 1, or 2 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"issue_area\\\": {\\n \\\"code\\\": integer, // 1–14 from the list below\\n \\\"label\\\": string // short description of the main issue area\\n }\\n\\n- \\\"issue_area_alt\\\": {\\n \\\"code\\\": integer, // 1–14; your best non-Misc alternative if issue_area.code == 13, else same as issue_area.code\\n \\\"label\\\": string\\n }\\n\\n- \\\"legal_provisions\\\": [\\n {\\n \\\"area_code\\\": integer, // 1–9 (law area codes below)\\n \\\"area_label\\\": string,\\n \\\"provision_text\\\": string // e.g. \\\"Fourth Amendment, search and seizure\\\", \\\"18 U.S.C. § 924(c)\\\"\\n },\\n ...\\n ]\\n\\n- \\\"decision_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"lower_court_direction\\\": {\\n \\\"code\\\": integer, // 1, 2, or 3 as defined below\\n \\\"label\\\": string\\n }\\n\\n- \\\"declaration_of_unconstitutionality\\\": {\\n \\\"code\\\": integer, // 0 or 1 (see Section 3.8)\\n \\\"label\\\": string // \\\"no\\\" or \\\"yes\\\"\\n }\\n\\n- \\\"method\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.9)\\n \\\"label\\\": string\\n }\\n\\n- \\\"who_appealed\\\": {\\n \\\"code\\\": integer, // 1–4 (see Section 3.10)\\n \\\"label\\\": string\\n }\\n\\n- \\\"dissent_basis\\\": { // ONLY include if \\\"Has dissenting opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"concurrence_type\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"code\\\": integer, // 1–3 (see Section 3.12)\\n \\\"label\\\": string\\n }\\n\\n- \\\"concurrence_basis\\\": { // ONLY include if \\\"Has concurring opinion: Yes\\\" in metadata\\n \\\"procedural\\\": integer, // 0–3 (see Section 3.11)\\n \\\"merits\\\": integer // 0–3 (see Section 3.11)\\n }\\n\\n- \\\"justification\\\": string // 2–4 sentences summarizing why you chose these codes.\\n // You MUST explicitly say: (a) who won and why, (b) what issue_area.code is,\\n // (c) how that combination implied the decision_direction.code you chose, and\\n // (d) why the lower_court_direction is the same as or opposite to decision_direction.\\n\\nYour entire response must be exactly one JSON object in this schema and nothing else.\\n\\n==================================================\\n3. VARIABLE INTERPRETATION\\n==================================================\\n\\n3.1 Petitioner and Respondent\\n\\n- Use the case caption, opinion type, and text to determine who is appealing.\\n- Petitioner = the party seeking relief in the Court of Appeals (typically the appellant).\\n- Respondent = the party opposing the appeal (typically the appellee).\\n- The \\\"name\\\" should be a reasonably short label from the caption.\\n- The \\\"description\\\" should describe their role and identity (e.g., \\\"federal prisoner challenging sentence\\\",\\n \\\"state official in civil-rights suit\\\", \\\"private corporation defendant\\\", \\\"federal agency\\\").\\n\\nFor each party, also assign a \\\"category\\\" code using this 7-category typology:\\n\\n1 = private business or corporation (including partnerships, sole proprietorships)\\n2 = nonprofit organization, interest group, or union (as an institutional party)\\n3 = federal government (including U.S. as party, federal agencies like DOJ, IRS, EPA, SEC, NLRB, etc.)\\n4 = sub-state government (city, county, municipality, school district, local official)\\n5 = state government (state, governor, state agency, state official in official capacity)\\n6 = tribal government or Indian nation\\n7 = natural person (individual, including pro se litigants and individual officials sued personally)\\n\\nChoose the category based on who the party IS, not what kind of case it is. When a government\\nofficial is sued in their official capacity, code them as the relevant government level (3, 4, or 5).\\nWhen sued in their personal/individual capacity, code them as 7 (natural person).\\n\\n\\n3.2 Disposition of Case\\n\\nCodes (Songer-style numbering):\\n\\n1 = affirmed (includes modified and affirmed)\\n2 = reversed\\n3 = reversed and remanded\\n4 = vacated and remanded\\n5 = affirmed in part, reversed (or vacated) in part\\n6 = affirmed in part, reversed (or vacated) in part, and remanded\\n7 = vacated\\n8 = petition denied or appeal dismissed\\n9 = stay, petition, or motion granted\\n10 = certification to or from a lower court\\n11 = affirmed, vacated, and remanded (without reversal)\\n0 = no disposition\\n\\n- Use the text and opinion type (majority, rehearing, etc.) to identify what the Court of Appeals actually did.\\n- Only choose options 3, 4, or 6 if the opinion contains explicit instructions to remand the matter to the lower court\\n or original authority (the word \\\"remand\\\" must appear in the opinion text).\\n- For rehearing petitions, be explicit about whether the petition is granted (9), denied (8), or leads to modification\\n or vacatur (7) of the prior judgment.\\n- If the Court takes no substantive action on the merits (e.g., purely procedural opinion), use 0 (\\\"no disposition\\\").\\n\\n\\n3.3 Winning Party\\n\\nCodes:\\n\\n0 = no favorable disposition for petitioning party apparent\\n1 = petitioning party received a favorable disposition\\n2 = favorable disposition for petitioning party unclear\\n\\n- Treat the **petitioner** as the party seeking relief in the Court of Appeals.\\n- A \\\"favorable disposition\\\" is one that materially improves the petitioner's position relative to the lower court judgment\\n (e.g., reversal, vacatur, remand requiring relief, sentence reduction).\\n- For mixed outcomes (e.g., affirmed in part and reversed in part), assess whether, on balance, the petitioner is better off.\\n If clearly better off, use 1; if clearly not better off, use 0; if ambiguous, use 2.\\n- Note that an outcome favoring the petitioning party may be either an affirmation (if the lower court had ruled in favor of the petitioner)\\n OR a reversal (if the lower court had ruled against the petitioner). You should focus on whether the petitioner is better off,\\n regardless of the case disposition.\\n\\n\\n3.4 Issue Area\\n\\nChoose exactly one code and label from this list, based on the main legal/policy issue in the appeal:\\n\\n1 = Criminal Procedure\\n (e.g., search and seizure, arrest, interrogation, right to counsel, trial and sentencing rules, habeas)\\n\\n2 = Civil Rights\\n (e.g., equal protection, discrimination based on race or other protected traits, voting rights, school desegregation)\\n\\n3 = First Amendment\\n (e.g., freedom of speech, press, association, religion, and related expression claims)\\n\\n4 = Due Process\\n (e.g., non-criminal procedural fairness, property or liberty deprivations, economic regulation challenges framed as due process)\\n\\n5 = Privacy\\n (e.g., contraception, abortion, family relationships, intimate or sexual privacy, personal autonomy)\\n\\n6 = Attorneys\\n (e.g., attorney's fees, sanctions, bar admission and discipline, representation issues)\\n\\n7 = Unions\\n (e.g., labor relations, union membership and internal affairs, collective bargaining, union security)\\n\\n8 = Economic Activity\\n (e.g., antitrust, securities, regulation of business, commercial law, bankruptcy, financial regulation)\\n\\n9 = Judicial Power\\n (e.g., jurisdiction, standing, mootness, justiciability, sovereign immunity, standards of review, appellate procedure)\\n\\n10 = Federalism\\n (e.g., conflicts between federal and state authority, preemption, Tenth Amendment issues, state sovereign immunity vs federal power)\\n\\n11 = Interstate Relations\\n (e.g., disputes between states, boundary and water disputes, interstate compacts and obligations)\\n\\n12 = Federal Taxation\\n (e.g., federal income, corporate, estate, excise, and other federal tax issues)\\n\\n13 = Miscellaneous\\n (cases that genuinely do not fit any other issue area — see IMPORTANT note below)\\n\\n14 = Private Action\\n (e.g., private law disputes in property, contracts, torts, wills, estates, and similar matters)\\n\\nPick the **single best** issue area based on what the Court of Appeals treats as the central question.\\nIf multiple areas appear, choose the one most emphasized in the reasoning.\\n\\nIMPORTANT — AVOID MISCELLANEOUS (13):\\n\\nCode 13 should be used ONLY as a genuine last resort when NO other category applies even\\napproximately. Before choosing 13, reconsider the following common mappings:\\n\\n - Medical malpractice, products liability, personal injury, breach of contract,\\n admiralty, property disputes → 14 (Private Action)\\n - Workers' compensation (LHWCA, FELA, Longshore Act) → 7 (Unions) or 14 (Private Action)\\n - Bankruptcy, securities, antitrust, business regulation → 8 (Economic Activity)\\n - Social Security, disability, welfare benefits → 2 (Civil Rights)\\n - Attorneys' fees, bar discipline → 6 (Attorneys)\\n - Federal preemption of state law → 10 (Federalism)\\n - Procedural rehearing or jurisdictional questions → 9 (Judicial Power)\\n - Separation of powers, legislative veto → 13 (Miscellaneous) — this is a genuine use\\n\\nIf you DO choose issue_area.code = 13, you MUST also fill in \\\"issue_area_alt\\\" with your\\nbest-guess alternative code (the closest non-Misc category). If issue_area.code is not 13,\\nset \\\"issue_area_alt\\\" to the same code and label as \\\"issue_area\\\".\\n\\nISSUE AREA TIEBREAKING RULES:\\n\\nWhen a case implicates more than one issue area, apply these rules to select the primary one:\\n\\n- Criminal Procedure (1) vs. Civil Rights (2): If the court's holding concerns criminal\\n conviction, sentencing, or procedure itself (e.g., suppression of evidence, habeas review of\\n a conviction), use 1 even if discrimination or civil-rights statutes are mentioned. Use 2\\n only when the holding primarily addresses whether discriminatory treatment occurred or whether\\n a civil-rights statute was violated.\\n\\n- Criminal Procedure (1) vs. Due Process (4): If the party is a criminal defendant raising\\n due-process claims about trial procedure, sentencing, or conditions of confinement, use 1.\\n Use 4 only when the due-process claim arises in a non-criminal context (e.g., government\\n termination of benefits, property seizure in a civil case).\\n\\n- Judicial Power (9) vs. any other area: Many cases discuss jurisdiction, standing, or\\n justiciability before reaching the merits. Use 9 ONLY when the court's holding is\\n primarily about whether it has the power to hear the case — i.e., the court resolves a\\n jurisdictional or justiciability question and that resolution IS the outcome. If the court\\n establishes jurisdiction and then rules on the merits, use the substantive merits area instead.\\n\\n- Economic Activity (8) vs. Federalism (10): If the holding is about whether a federal\\n statute preempts state regulation, use 10. If the holding is about the substance of a\\n federal regulatory scheme (e.g., upholding or striking down an EPA rule on the merits),\\n use 8.\\n\\n- Civil Rights (2) vs. Due Process (4): If the claim involves alleged discrimination under\\n equal protection or a civil-rights statute, use 2. Use 4 only when the claim is about\\n procedural fairness or property/liberty deprivation without an underlying discrimination\\n allegation.\\n\\n\\n3.5 Legal Provisions\\n\\nFor each major legal provision the Court of Appeals applies or interprets, add an entry to \\\"legal_provisions\\\".\\n\\nLaw Area codes (SCDB varLawArea):\\n\\n1 = Constitution (original text)\\n An Article or Section of the original Constitution (not an Amendment).\\n Examples: Article I (congressional power), Article II (executive power),\\n Article III (judicial power), Supremacy Clause, Commerce Clause, Full Faith\\n and Credit Clause, Contracts Clause.\\n\\n2 = Constitutional Amendment\\n Any Amendment to the Constitution, including the Bill of Rights.\\n Examples: First Amendment (speech, religion), Fourth Amendment (search and\\n seizure), Fifth Amendment (due process, takings, self-incrimination), Sixth\\n Amendment (right to counsel, jury trial), Eighth Amendment (cruel and unusual\\n punishment), Fourteenth Amendment (equal protection, due process).\\n\\n3 = Federal Statute\\n A statute enacted by Congress, cited by title and section of the U.S. Code or\\n by its popular name. This is the most common category.\\n Examples: 42 U.S.C. § 1983 (civil rights), 28 U.S.C. § 1331 (federal question\\n jurisdiction), Title VII of the Civil Rights Act, Sherman Act § 1 (antitrust),\\n 18 U.S.C. § 924(c) (firearms), ERISA, CERCLA, ADA, FOIA, Sentencing Guidelines.\\n\\n4 = Court Rules\\n Federal Rules of Civil Procedure, Criminal Procedure, Evidence, or Appellate\\n Procedure, as well as local circuit rules.\\n Examples: Fed. R. Civ. P. 12(b)(6) (failure to state a claim), Fed. R. Civ. P.\\n 56 (summary judgment), Fed. R. Crim. P. 11 (plea agreements), Fed. R. Evid. 403\\n (prejudicial evidence).\\n\\n5 = Other\\n Common law doctrines, equitable principles, or other legal authorities not fitting\\n the categories above.\\n Examples: common law fraud, maritime law (general), equitable estoppel, doctrine\\n of laches, implied contracts.\\n\\n6 = Infrequently litigated statutes\\n Federal statutes that appear rarely in appellate litigation and are not well-known\\n by popular name. Use this when the statute is obscure or highly specialized.\\n Examples: Railway Labor Act, Packers and Stockyards Act, Suits in Admiralty Act,\\n obscure provisions of the Internal Revenue Code.\\n\\n8 = State or local law or regulation\\n A state statute, state constitutional provision, local ordinance, or state/local\\n regulation. Use this when the court applies or interprets state or local law rather\\n than federal law.\\n Examples: state tort law, state contract law, state criminal statute, local zoning\\n ordinance, state environmental regulation.\\n\\n9 = No Legal Provision\\n The opinion does not meaningfully rely on any identifiable legal provision. Very\\n rare — use only when the court's analysis is purely factual or procedural without\\n citing specific legal authority.\\n\\nFor each entry:\\n- `area_code`: choose the single best area for the provision.\\n- `area_label`: the corresponding label (e.g., \\\"Constitution\\\", \\\"Federal Statute\\\").\\n- `provision_text`: identify the specific provision (e.g., \\\"Fourth Amendment, search and seizure\\\",\\n \\\"42 U.S.C. § 1983\\\", \\\"Federal Rule of Civil Procedure 12(b)(6)\\\", \\\"state tort law\\\").\\n\\nInclude the one to three provisions central to the court's analysis. If the opinion does not meaningfully\\nrely on any legal provision (very rare), use a single entry with area_code 9 and a brief explanation.\\n\\nDistinguishing area codes:\\n- If the court interprets the CONSTITUTION ITSELF (e.g., Commerce Clause), use 1.\\n- If the court interprets a CONSTITUTIONAL AMENDMENT (e.g., Fourth Amendment), use 2.\\n- If the court interprets a FEDERAL STATUTE (the most common scenario), use 3.\\n- If the dispute is about a PROCEDURAL RULE (e.g., Rule 12(b)(6) motion), use 4.\\n- Use 8 for STATE LAW even if the case is in federal court on diversity jurisdiction.\\n- Use 5 for common law doctrines that are not codified in any statute or rule.\\n- Use 6 only for genuinely obscure federal statutes that would not be recognized by name.\\n\\n\\n3.6 Decision Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\n>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>\\nCRITICAL: The codes \\\"conservative\\\" and \\\"liberal\\\" are MECHANICAL LABELS defined by the SCDB\\ncodebook. They are NOT your own assessment of political ideology. You MUST apply them exactly\\nas the lookup table below specifies, even when the result seems counterintuitive. For example,\\na decision favoring a taxpayer over the IRS is ALWAYS coded conservative (1) in a Federal\\nTaxation case, regardless of the policy context. A decision upholding a government regulation\\nis ALWAYS coded liberal (2) in an Economic Activity case, even if the regulation benefits\\nbusiness. Do NOT let your general understanding of politics override these rules.\\n<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<\\n\\nDIRECTION LOOKUP TABLE — apply mechanically based on issue_area.code and who benefits:\\n\\nissue_area.code | benefiting side | direction\\n----------------|------------------------------------------------------------|----------\\n1 (Crim. Pro.) | accused / convicted person / criminal defendant | 2 (liberal)\\n1 (Crim. Pro.) | prosecution / law enforcement / government | 1 (conservative)\\n2 (Civ. Rts.) | civil rights or civil liberties claimant | 2 (liberal)\\n2 (Civ. Rts.) | government defendant / party opposing civil rights claim | 1 (conservative)\\n3 (1st Amend.) | person asserting First Amendment protections | 2 (liberal)\\n3 (1st Amend.) | government restricting speech, press, religion, etc. | 1 (conservative)\\n4 (Due Proc.) | individual challenging deprivation of liberty or property | 2 (liberal)\\n4 (Due Proc.) | government upholding deprivation | 1 (conservative)\\n5 (Privacy) | person asserting privacy / autonomy / reproductive choice | 2 (liberal)\\n5 (Privacy) | government restricting privacy or autonomy | 1 (conservative)\\n7 (Unions) | union / workers / collective bargaining interests | 2 (liberal)\\n7 (Unions) | employer / management / anti-union interests | 1 (conservative)\\n8 (Econ. Act.) | workers, consumers, injured parties, environment, govt | 2 (liberal)\\n | regulation, or small entities against large business |\\n8 (Econ. Act.) | employers, large business, property owners, deregulation | 1 (conservative)\\n9 (Jud. Power) | broader court access / standing / review | 2 (liberal)\\n9 (Jud. Power) | narrower court access / standing / review | 1 (conservative)\\n10 (Federalism) | federal authority over state authority | 2 (liberal)\\n10 (Federalism) | state sovereignty over federal power | 1 (conservative)\\n12 (Fed. Tax.) | United States / IRS / federal taxing authority | 2 (liberal)\\n12 (Fed. Tax.) | taxpayer | 1 (conservative)\\n11,14 | use 3 (unspecifiable) unless decision clearly aligns | 3 (unspecifiable)\\n | with one of the patterns above |\\n\\nPROCEDURE: You MUST FIRST determine who won the case (winning_party), THEN identify the\\nissue_area, and LAST look up the decision_direction from the table above by matching the\\nissue_area.code with the side that benefited. Do NOT reason about direction from first\\nprinciples — use the table.\\n\\nIf, after applying the table, you cannot determine which side benefited, use 3 (unspecifiable).\\n\\n\\n3.6.1 Worked Examples\\n\\nThe following three examples demonstrate correct mechanical application of the direction\\nlookup table, including cases where the SCDB coding may seem counterintuitive:\\n\\nEXAMPLE A — Federal Taxation (issue_area.code = 12)\\n\\n Facts: A taxpayer appeals an IRS assessment to the Court of Appeals. The court reverses the\\n Tax Court, holding that the taxpayer's deductions were proper and the assessment was incorrect.\\n\\n Coding:\\n issue_area.code = 12 (Federal Taxation)\\n winning_party.code = 1 (petitioner/taxpayer received favorable disposition)\\n decision_direction = 1 (conservative) ← taxpayer won; lookup table row for code 12\\n says \\\"taxpayer → 1 (conservative)\\\"\\n\\n NOTE: Even though the taxpayer prevailed against the government, the SCDB mechanically\\n codes this as conservative (1) because the taxpayer benefited.\\n\\n\\nEXAMPLE B — Criminal Procedure (issue_area.code = 1)\\n\\n Facts: A criminal defendant appeals his conviction, arguing that evidence was obtained\\n through an unlawful search. The Court of Appeals affirms, holding the search was lawful.\\n\\n Coding:\\n issue_area.code = 1 (Criminal Procedure)\\n winning_party.code = 0 (no favorable disposition for petitioner)\\n decision_direction = 1 (conservative) ← prosecution/government benefited; lookup table\\n row for code 1 says \\\"prosecution → 1 (conservative)\\\"\\n\\n\\nEXAMPLE C — Economic Activity (issue_area.code = 8)\\n\\n Facts: A chemical company challenges an EPA water-quality regulation. The Court of Appeals\\n affirms, upholding the regulation as within the agency's statutory authority.\\n\\n Coding:\\n issue_area.code = 8 (Economic Activity)\\n winning_party.code = 0 (no favorable disposition for petitioner/company)\\n decision_direction = 2 (liberal) ← government regulation upheld against business;\\n lookup table row for code 8 says \\\"govt regulation → 2 (liberal)\\\"\\n\\n NOTE: Even if one might view upholding a regulation as a \\\"pro-business\\\" outcome (e.g., it\\n creates regulatory certainty), the SCDB mechanically codes government regulation prevailing\\n over a business challenge as liberal (2).\\n\\n\\n3.6.2 Internal Consistency Check\\n\\nBefore you output the JSON, you MUST run the following consistency check:\\n\\n1. Identify issue_area.code.\\n2. Identify which side won in the case (petitioner vs respondent) and what kind of party they are.\\n3. Look up the corresponding row in the DIRECTION LOOKUP TABLE above.\\n4. Verify that your decision_direction.code matches the table entry.\\n5. If it does not match, CORRECT decision_direction.code to match the table.\\n\\n\\n3.7 Lower Court Direction\\n\\nCodes:\\n\\n1 = conservative\\n2 = liberal\\n3 = unspecifiable\\n\\nUse the same DIRECTION LOOKUP TABLE as for Decision Direction (see Section 3.6), but applied to\\nthe **lower court's outcome**. You MUST determine the lower court's direction by comparing the lower\\ncourt's judgment and the appellate court's disposition.\\n\\nUse this rule of thumb:\\n\\n1. First, identify the appellate disposition (case_disposition.code).\\n2. Next, identify the decision_direction.code for the Court of Appeals.\\n3. Last, infer who benefited in the lower court and assign lower_court_direction as follows:\\n\\n- If the Court of Appeals clearly **leaves the lower court judgment in place** (e.g., case_disposition.code = 1 \\\"affirmed\\\"\\n or 8 \\\"petition denied or appeal dismissed\\\"), then the lower court's direction is the SAME as decision_direction.code.\\n\\n- If the Court of Appeals clearly **undoes the lower court judgment** (e.g., case_disposition.code = 2 \\\"reversed\\\",\\n 3 \\\"reversed and remanded\\\", 4 \\\"vacated and remanded\\\", or 7 \\\"vacated\\\"), then the lower court's direction is the\\n OPPOSITE of decision_direction.code.\\n\\n- For mixed outcomes (case_disposition.code = 5 or 6), compare who won more in the lower court vs on appeal:\\n * If, on balance, the petitioner is better off after the appeal than under the lower court judgment, then\\n the lower court's direction should usually be the OPPOSITE of decision_direction.code.\\n * If, on balance, the petitioner is worse off or no better off than under the lower court judgment, then\\n the lower court's direction should usually be the SAME as decision_direction.code.\\n * If you cannot determine who benefited more from the lower court judgment even approximately, use 3 (unspecifiable).\\n\\n- For stays, motions, certifications, or purely procedural dispositions (codes 9, 10, 0), infer who benefited in\\n the lower court if possible; if not, use 3 (unspecifiable).\\n\\nWhen you set lower_court_direction, remember:\\n- \\\"SAME\\\" means lower_court_direction.code = decision_direction.code.\\n- \\\"OPPOSITE\\\" means: if decision_direction.code = 1 (conservative), then lower_court_direction.code = 2 (liberal);\\n if decision_direction.code = 2 (liberal), then lower_court_direction.code = 1 (conservative).\\n\\n3.8 Declaration of Unconstitutionality\\n\\nCodes:\\n\\n0 = the court did NOT strike down any statute, regulation, or government action as unconstitutional\\n1 = the court DID declare a statute, regulation, or government action unconstitutional\\n\\nCode 1 if the court holds that a law, regulation, executive action, or government practice violates\\nthe Constitution (whether facial or as-applied). This includes striking down acts of Congress, federal\\nregulations, state statutes, state constitutional provisions, and local ordinances.\\n\\nDo NOT code 1 merely because the court discusses constitutional issues. The court must actually\\nhold that something is unconstitutional. If the court upholds the constitutionality of the\\nchallenged provision, code 0.\\n\\n\\n3.9 Method\\n\\nCodes:\\n\\n1 = standard panel decision (typically 3 judges, first hearing)\\n2 = panel rehearing (same panel reconsidering its prior decision)\\n3 = en banc (full court or en banc panel rehearing a case)\\n4 = per curiam (unsigned opinion — may overlap with codes 1 or 3)\\n\\nUse the opinion text, metadata, and case caption to determine the method. Look for explicit\\nstatements like \\\"before the court en banc,\\\" \\\"on rehearing,\\\" \\\"upon reconsideration,\\\" or\\n\\\"per curiam.\\\" If the metadata indicates more than 3 judges participated, this strongly\\nsuggests en banc (code 3). If none of these indicators are present, use 1 (standard panel).\\n\\n\\n3.10 Who Appealed\\n\\nCodes:\\n\\n1 = original plaintiff (or the equivalent, e.g. the Title VII complainant, habeas petitioner)\\n2 = original defendant (or the equivalent, e.g. the employer in a Title VII suit)\\n3 = federal government on behalf of the original plaintiff\\n4 = federal government on behalf of the original defendant\\n\\nDetermine who initiated the appeal by examining the case caption, opinion text, and procedural\\nhistory. The \\\"original plaintiff\\\" is the party who filed suit in the lower court; the \\\"original\\ndefendant\\\" is the party who was sued. In criminal cases, the original plaintiff is typically the\\ngovernment (prosecution) and the original defendant is the accused.\\n\\nIf the opinion describes a cross-appeal, code based on who filed the PRIMARY appeal (the one\\nthe court addresses as the main appeal). If you cannot determine who appealed, use 1 as default\\nfor criminal defendants (who are most commonly appellants) or your best judgment from the caption\\nordering (appellant is typically listed first).\\n\\n\\n3.11 Dissent/Concurrence Basis (Conditional)\\n\\nIf the case metadata states \\\"Has dissenting opinion: Yes\\\" or \\\"Has concurring opinion: Yes\\\",\\ncode whether that separate opinion addresses procedural issues, merits issues, or both.\\n\\nFor each dimension (procedural / merits), use one of these codes:\\n\\n0 = the dissent/concurrence exists but does NOT address this type of issue\\n1 = addresses the SAME issue as the majority and DISAGREES with the majority's resolution\\n2 = addresses the SAME issue as the majority and AGREES with the majority's resolution\\n3 = addresses an issue NOT discussed by the majority opinion\\n\\n\\\"dissent_basis\\\" (include ONLY if \\\"Has dissenting opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the dissent address procedural questions (jurisdiction, standing,\\n standard of review, procedural default, timeliness, etc.)?\\n - \\\"merits\\\": Does the dissent address the substantive legal merits (statutory interpretation,\\n constitutional analysis, factual sufficiency, etc.)?\\n\\n\\\"concurrence_basis\\\" (include ONLY if \\\"Has concurring opinion: Yes\\\"):\\n - \\\"procedural\\\": Does the concurrence address procedural questions?\\n - \\\"merits\\\": Does the concurrence address the substantive legal merits?\\n\\nIf multiple dissenting or concurring opinions exist, code based on the PRIMARY one\\n(the longest or most substantive).\\n\\nIf the metadata says \\\"Has dissenting opinion: No\\\", OMIT \\\"dissent_basis\\\" entirely from your output.\\nIf the metadata says \\\"Has concurring opinion: No\\\", OMIT \\\"concurrence_type\\\" and \\\"concurrence_basis\\\" entirely from your output.\\n\\n\\n3.12 Concurrence Type (Conditional)\\n\\nIf the case metadata states \\\"Has concurring opinion: Yes\\\", classify the concurrence as one of\\nthe following types:\\n\\nCodes:\\n\\n1 = regular concurrence — the concurring judge JOINED the majority opinion AND wrote or joined\\n a separate concurrence to add additional reasoning, emphasize a point, or note a limitation.\\n The key indicator: the judge agrees with BOTH the outcome AND the reasoning of the majority.\\n\\n2 = special concurrence / concurrence in result only — the concurring judge agrees with the\\n OUTCOME (disposition) but does NOT join the majority opinion's reasoning. Look for language\\n like \\\"concurring in the result,\\\" \\\"concurring in the judgment,\\\" \\\"concurring in the result\\n only,\\\" or \\\"I concur in the judgment but write separately because I disagree with the\\n majority's analysis.\\\" The key indicator: the judge explicitly declines to join the majority\\n opinion or distances themselves from the majority's reasoning.\\n\\n3 = cannot determine — the opinion text does not make clear whether the concurring judge joined\\n the majority opinion or only the result.\\n\\nThis distinction matters because special concurrences (code 2) signal reasoning disagreement\\nwithin a nominally unanimous disposition. A case where all judges agree on the outcome but one\\nwrites a special concurrence is qualitatively different from one where all judges agree on\\nboth outcome and reasoning.\\n\\nIf multiple concurring opinions exist, code based on the PRIMARY concurrence (the longest or\\nmost substantive). If one is a regular concurrence and another is a special concurrence,\\ncode based on the special concurrence (code 2), as it is more analytically informative.\\n\\n\\n==================================================\\n4. GENERAL RULES\\n==================================================\\n\\n- Use metadata (case name, court, date, opinion type) to:\\n - distinguish main merits opinions from rehearing or procedural opinions,\\n - interpret how the appellate disposition relates to the prior judgment,\\n - infer whether the case is criminal, civil rights, tax, etc.\\n- If some information is missing, use the appropriate \\\"unspecifiable\\\" or \\\"no disposition\\\" code\\n (e.g., 3 for direction, 11 for case_disposition); do not invent facts.\\n- Carefully follow the steps laid out in the introduction in order. In particular,\\n decide who won the case (winning_party). Then, decide what the issue_area is.\\n Only choose decision_direction once you have completed those steps, and follow the exact\\n directions in Section 3.6.\\n- In the justification, explicitly state who won, why (according to the disposition), and\\n how that winner and the issue_area code together implied the direction code you chose.\\n Reference the DIRECTION LOOKUP TABLE row you applied.\\n- ALWAYS choose exactly one code for each coded variable where required.\\n- NEVER output any keys other than the ones defined in the schema above.\\n- Include \\\"dissent_basis\\\", \\\"concurrence_type\\\", and \\\"concurrence_basis\\\" ONLY when the metadata\\n indicates the corresponding opinion type exists. Omit them entirely otherwise.\\n- NEVER output any numeric codes outside the permitted lists above.\\n\\n\\n==================================================\\n5. FINAL REQUIREMENT\\n==================================================\\n\\nAfter reading [CASE_METADATA] and [CASE_TEXT], respond ONLY with the JSON object in the schema\\nabove. Do not include any extra commentary, markdown, or explanation outside the JSON.\\n\",\n \"role\": \"system\"\n },\n {\n \"content\": \"[CASE_METADATA]\\nCase name: United States v. Davion Fitzgerald\\nCourt: U.S. Court of Appeals for the Ninth Circuit\\nDate decided: 2019-08-26\\nOpinion type: majority\\nHas dissenting opinion: Yes\\nHas concurring opinion: No\\nAdditional notes: \\n\\n[CASE_TEXT]\\nFOR PUBLICATION\\n\\n UNITED STATES COURT OF APPEALS\\n FOR THE NINTH CIRCUIT\\n\\n\\nUNITED STATES OF AMERICA, No. 18-10116\\n Plaintiff-Appellant,\\n D.C. No.\\n v. 2:17-cr-00295-\\n JCM-NJK-1\\nDAVION FITZGERALD,\\n Defendant-Appellee. OPINION\\n\\n Appeal from the United States District Court\\n for the District of Nevada\\n James C. Mahan, District Judge, Presiding\\n\\n Argued and Submitted March 12, 2019\\n San Francisco, California\\n\\n Filed August 26, 2019\\n\\n Before: William A. Fletcher, Paul J. Watford,\\n and Andrew D. Hurwitz, Circuit Judges.\\n\\n Per Curiam Opinion;\\n Dissent by Judge W. Fletcher\\n\\n2 UNITED STATES V. FITZGERALD\\n\\n SUMMARY *\\n\\n\\n Criminal Law\\n\\n Vacating a sentence and remanding for resentencing, the\\npanel held that the defendant’s prior Nevada conviction for\\nattempted battery with substantial bodily harm in violation\\nof Nev. Rev. Stat. §§ 200.481(2)(b) and 193.330 qualifies as\\na felony conviction for a crime of violence under U.S.S.G.\\n§ 2K2.1.\\n\\n Applying United States v. Johnson, 920 F.3d 628 (9th\\nCir. 2019), and observing that the state court treated the\\ndefendant’s conviction as a felony rather than a\\nmisdemeanor, the panel rejected the defendant’s contention\\nthat the conviction is not a felony conviction because it is a\\nwobbler.\\n\\n The panel held that the defendant’s Nevada conviction\\nqualifies as a crime of violence under the elements clause of\\nU.S.S.G. § 4B1.2(a)(1). In so holding, the panel addressed\\nNevada’s definition of “substantial bodily harm,” which\\nincludes “prolonged physical pain,” and concluded that it is\\nnot evident that there’s a realistic probability that a\\ndefendant could be convicted of Nevada attempted battery\\nwith substantial bodily harm without the attempted use of\\nviolent force.\\n\\n Dissenting, Judge W. Fletcher wrote that because\\n“prolonged physical pain,” as the Nevada Supreme Court\\n\\n *\\n This summary constitutes no part of the opinion of the court. It\\nhas been prepared by court staff for the convenience of the reader.\\n\\n UNITED STATES V. FITZGERALD 3\\n\\nhas explained, may be caused by simple touching, and\\nbecause the definition of “substantial bodily harm” is\\nindivisible, attempted battery with substantial bodily harm\\nunder §§ 193.330 and 200.481 does not qualify as a crime of\\nviolence under the elements clause.\\n\\n\\n COUNSEL\\n\\nElizabeth O. White (argued), Appellate Chief; Dayle\\nElieson, United States Attorney; United States Attorney’s\\nOffice, Reno, Nevada; for Plaintiff-Appellant.\\n\\nAmy B. Cleary (argued), Assistant Federal Public Defender;\\nRene L. Valladares, Federal Public Defender; Office of the\\nFederal Public Defender, Las Vegas, Nevada; for\\nDefendant-Appellee.\\n\\n\\n OPINION\\n\\nPER CURIAM:\\n\\n Davion Fitzgerald pleaded guilty to unlawful possession\\nof a firearm in violation of 18 U.S.C. §§ 922(g)(1) and\\n924(a)(2). At sentencing, the government requested an\\nenhancement under § 2K2.1(a)(4)(A) of the 2016\\nSentencing Guidelines, which provides for an increase to a\\nbase offense level of 20 if the defendant has a prior “felony\\nconviction of . . . a crime of violence.” The government\\nbased its request on Fitzgerald’s prior Nevada conviction for\\nattempted battery with substantial bodily harm in violation\\nof Nev. Rev. Stat. §§ 200.481(2)(b) and 193.330. The\\ndistrict court declined to apply the enhancement, concluding\\nthat Fitzgerald’s Nevada conviction qualified neither as a\\n\\n4 UNITED STATES V. FITZGERALD\\n\\n“felony conviction” nor a “crime of violence.” The\\ngovernment has appealed from the sentence imposed. We\\ndisagree with the district court on both counts, and therefore\\nvacate Fitzgerald’s sentence.\\n\\n I\\n\\n Fitzgerald first argues that his Nevada conviction is not\\na “felony conviction” because it is a “wobbler.” That is,\\nunder state law, it may be treated as either a felony or a\\nmisdemeanor. See Nev. Rev. Stat. § 193.330(1)(a)(4);\\nUnited States v. Bridgeforth, 441 F.3d 864, 870 (9th Cir.\\n2006). The Sentencing Guidelines define a “felony\\nconviction” as “a prior adult federal or state conviction for\\nan offense punishable by death or imprisonment for a term\\nexceeding one year, regardless of whether such offense is\\nspecifically designated as a felony and regardless of the\\nactual sentence imposed.” U.S.S.G. § 2K2.1 cmt. n.1.\\n“Despite this clear admonition, our binding circuit precedent\\nrequires us, where wobblers are concerned, to ignore the\\nmaximum sentence allowed by statute and instead adopt the\\ndesignation that [the State] gives to the offense.” United\\nStates v. Johnson, 920 F.3d 628, 634 n.3 (9th Cir. 2019).\\nBecause “a state court’s subsequent treatment of a wobbler\\nis controlling,” Bridgeforth, 441 F.3d at 872, we must\\nexamine how Fitzgerald was actually punished. Here, it is\\nclear that the state court treated his conviction as a felony.\\n\\n Fitzgerald argues that our precedents on this point did\\nnot survive Moncrieffe v. Holder, 569 U.S. 184 (2013), but\\nwe recently rejected that very argument. See Johnson,\\n920 F.3d at 637–38. Fitzgerald’s Nevada conviction\\ntherefore qualifies as a “felony conviction” for purposes of\\nU.S.S.G. § 2K2.1.\\n\\n UNITED STATES V. FITZGERALD 5\\n\\n II\\n\\n Fitzgerald next contends that his Nevada conviction does\\nnot qualify as a “crime of violence.” The commentary to\\n§ 2K2.1 defines “crime of violence” by cross-reference to\\n§ 4B1.2, which reads:\\n\\n (a) The term “crime of violence” means any\\n offense under federal or state law, punishable\\n by imprisonment for a term exceeding one\\n year, that—\\n\\n (1) has as an element the use, attempted use,\\n or threatened use of physical force against the\\n person of another, or\\n\\n (2) is murder, voluntary manslaughter,\\n kidnapping, aggravated assault, a forcible sex\\n offense, robbery, arson, extortion, or the use\\n or unlawful possession of a firearm described\\n in 26 U.S.C. § 5845(a) or explosive material\\n as defined in 18 U.S.C. § 841(c).\\n\\nU.S.S.G. § 4B1.2(a). The first of the numbered clauses is\\nknown as the “elements” clause, the second as the\\n“enumerated offenses” clause. “We use the categorical\\napproach to determine whether a state crime qualifies as a\\ncrime of violence for Guidelines purposes.” United States v.\\nMolinar, 881 F.3d 1064, 1067 (9th Cir. 2017). Because we\\nconclude that Fitzgerald’s conviction qualifies as a crime of\\nviolence under the elements clause, we do not address the\\nenumerated offenses clause.\\n\\n Under the elements clause, we ask whether the Nevada\\ncrime of attempted battery with substantial bodily harm “has\\nas an element the use, attempted use, or threatened use of\\n\\n6 UNITED STATES V. FITZGERALD\\n\\nphysical force against the person of another.” U.S.S.G.\\n§ 4B1.2(a)(1). Because this language is identical to that\\ninterpreted by the Supreme Court in Johnson v. United\\nStates, 559 U.S. 133, 136 (2010), we apply Johnson’s\\ndefinition of “physical force”: “violent force—that is, force\\ncapable of causing physical pain or injury to another\\nperson.” Id. at 140; see Molinar, 881 F.3d at 1068 & n.3.\\nThat threshold requires, at the very least, more than “a mere\\nunwanted touching.” Johnson, 559 U.S. at 142.\\n\\n In Nevada, a person can commit simple battery with\\nnothing more than an offensive touching. Nev. Rev. Stat.\\n§ 200.481(1)(a); Hobbs v. State, 251 P.3d 177, 179 (Nev.\\n2011). But battery with substantial bodily harm, as the name\\nsuggests, requires that the battery result in “substantial\\nbodily harm to the victim.” Nev. Rev. Stat. § 200.481(2)(b).\\nAnd attempted battery with substantial bodily harm—\\nFitzgerald’s crime of conviction—requires that the\\ndefendant act with the specific intent both to commit battery\\nand to bring about substantial bodily harm. See Nev. Rev.\\nStat. § 193.330(1); Tanksley v. State, 944 P.2d 240, 243\\n(Nev. 1997). 1\\n\\n Fitzgerald’s offense, therefore, is defined by the bodily\\ninjury the defendant intends to produce, not by the actual\\nlevel of force used. In United States v. Castleman, 572 U.S.\\n157 (2014), the Supreme Court explicitly declined to decide\\n“[w]hether or not the causation of bodily injury necessarily\\nentails violent force.” Id. at 167. But our court has held that,\\n\\n 1\\n Because “Nevada’s definition of attempt is coextensive with the\\nfederal definition,” United States v. Sarbia, 367 F.3d 1079, 1086 (9th\\nCir. 2004), there is no possibility that Nevada attempt sweeps more\\nbroadly than § 4B1.2’s reference to “attempted use . . . of physical\\nforce.”\\n\\n UNITED STATES V. FITZGERALD 7\\n\\nin general, “in the context of assault statutes, bodily injury\\nentails the use of violent, physical force.” United States v.\\nCalvillo-Palacios, 860 F.3d 1285, 1291 (9th Cir. 2017).\\n\\n Our rule is premised on a straightforward inference that\\nit takes Johnson-level force to produce bodily injury. See id.\\nat 1290. However, as we acknowledged in Calvillo-\\nPalacios, the validity of that inference depends on how a\\nstate defines “bodily injury” or, in this case, “substantial\\nbodily harm.” See id. at 1291–92. To take an extreme\\nexample, a state statute that defined bodily injury as merely\\n“an offensive touching” would not require violent force\\nunder Johnson. 559 U.S. at 142.\\n\\n Fitzgerald argues that we are dealing with such a statute\\nin this case. Nevada defines “substantial bodily harm” as\\neither “(1) [b]odily injury which creates a substantial risk of\\ndeath or which causes serious, permanent disfigurement or\\nprotracted loss or impairment of the function of any bodily\\nmember or organ; or (2) [p]rolonged physical pain.” Nev.\\nRev. Stat. § 0.060. The government does not argue that the\\nstatute is divisible, and Fitzgerald does not contest that a\\nconviction under the first definition would necessarily\\nrequire the use of violent force. So only the second\\ndefinition—“prolonged physical pain”—matters for our\\npurposes.\\n\\n In Collins v. State, 203 P.3d 90 (Nev. 2009), the Supreme\\nCourt of Nevada considered the meaning of the phrase\\n“prolonged physical pain.” In sustaining the statute against\\na vagueness challenge, the court held that “[t]he term ‘pain’\\nhas multiple meanings, ranging from mild discomfort or dull\\ndistress to acute often unbearable agony . . . and cannot be\\ndefined further.” Id. at 92 (internal quotation marks\\nomitted). For pain to be “prolonged,” the court stated, there\\nmust be “at least some physical suffering that lasts longer\\n\\n8 UNITED STATES V. FITZGERALD\\n\\nthan the pain immediately resulting from the wrongful act.”\\nId. at 93. “In a battery,” the court further explained, “the\\nwrongdoer would not be liable for ‘prolonged physical pain’\\nfor the touching itself,” but “would be liable for any lasting\\nphysical pain resulting from the touching.” Id. at 93 n.3.\\n\\n Fitzgerald argues that, because “substantial bodily harm”\\ncan mean only “mild discomfort” lasting “longer than the\\npain immediately resulting from the wrongful act,” the\\ninference that violent force is required to inflict such harm is\\nnot warranted. He points out that the Collins court used\\n“touching the skin of a person who has suffered third degree\\nburns” as an example of an act that would cause “exquisite\\npain.” Id. at 92 (citation omitted). Therefore, he argues,\\nNevada battery with substantial bodily harm can be\\ncommitted with only a mere touch (and attempted with only\\nan attempted touch), which Johnson teaches cannot count as\\nviolent force. 559 U.S. at 142.\\n\\n Supreme Court precedent, however, “requires more than\\nthe application of legal imagination to a state statute’s\\nlanguage.” Gonzales v. Duenas-Alvarez, 549 U.S. 183, 193\\n(2007). Fitzgerald “must demonstrate a realistic probability,\\nnot a theoretical possibility, that Nevada would apply its\\nstatute to conduct that falls outside the definition of a crime\\nof violence.” United States v. Guizar-Rodriguez, 900 F.3d\\n1044, 1052 (9th Cir. 2018) (internal quotation marks\\nomitted). Ordinarily, a defendant “must at least point to his\\nown case or other cases in which the state courts in fact did\\napply the statute in the overbroad manner for which he\\nargues.” Id. at 1048 (internal quotation marks omitted).\\nOtherwise, the overbreadth of the state statute must be\\n“evident from its text” or “evident from state court\\nprecedents interpreting that text.” Id. at 1052 (internal\\nquotation marks omitted).\\n\\n UNITED STATES V. FITZGERALD 9\\n\\n We do not think that Collins makes it “evident” that the\\nstatute sweeps as broadly as Fitzgerald contends. The\\ndecision neither holds nor states that substantial bodily harm\\ncan be caused by a mere touch (or by anything less than\\n“violent force”). Although Collins uses a touch as an\\nexample of an act that may cause pain, 203 P.3d at 92, it\\nnever says that the pain would count as “prolonged.” Indeed,\\nit clarifies that a batterer is not “liable for ‘prolonged\\nphysical pain’ for the touching itself,” but only “for any\\nlasting physical pain resulting from the touching.” Id. at 93\\nn.3. These statements make the answer to the question we\\nconfront here—whether a defendant could realistically be\\nconvicted of attempted battery with substantial bodily harm\\nwithout the attempted use of violent force—far from evident.\\n\\n For Fitzgerald to prevail, he would have to show that a\\ndefendant could realistically be convicted of attempted\\nbattery with substantial bodily harm for trying, with the\\nintent to cause lasting discomfort, merely to touch his victim\\n(or use other nonviolent force). That odd hypothetical\\nstrikes us as an exercise of “legal imagination.” Duenas-\\nAlvarez, 549 U.S. at 193. When someone intends to inflict\\nprolonged pain, even relatively minor pain, it is highly\\nimprobable that they would choose to do so through the use\\nof nonviolent force, which could easily fail to accomplish\\ntheir goal. And for the same reason, it is equally improbable\\nthat Nevada prosecutors would be able to secure convictions\\nfor attempted battery with substantial bodily harm in cases\\nin which the defendant tried to use only a touch or other\\nnonviolent force. Even if the possibility of such a conviction\\nis not theoretically foreclosed by Collins, that decision does\\nnot make evident that such a conviction is a “realistic\\nprobability.” Fitzgerald’s argument therefore falls short.\\n\\n * * *\\n\\n10 UNITED STATES V. FITZGERALD\\n\\n We conclude that Fitzgerald’s Nevada conviction for\\nattempted battery with substantial bodily harm qualifies as a\\nfelony conviction for a crime of violence under U.S.S.G.\\n§ 2K2.1. We therefore vacate Fitzgerald’s sentence and\\nremand for resentencing consistent with this opinion.\\n\\n VACATED AND REMANDED.\\n[END_CASE]\",\n \"role\": \"user\"\n }\n]","isConversational":true}}}],"truncated":false,"partial":false},"paginationData":{"pageIndex":40,"numItemsPerPage":100,"numTotalItems":31175,"offset":4000,"length":100}},"jwt":"eyJhbGciOiJFZERTQSIsImtpZCI6IjVHZDBvd0g5MTM2eDZjc1FvbE1zcktNWUZoRVFoUm5PVVVybEpjOUhaUEEifQ.eyJyZWFkIjp0cnVlLCJwZXJtaXNzaW9ucyI6eyJyZXBvLmNvbnRlbnQucmVhZCI6dHJ1ZX0sImlhdCI6MTc4NjUwNDUwNSwianRpIjoiMzI4ZmQ0NTAtNDhjZC00Zjg4LTgwYTUtYmY5ZjYzOTE4NGZiIiwic3ViIjoiL2RhdGFzZXRzL2VpdGFucy1wdS9jbC1jb2RpbmctaW5wdXQiLCJleHAiOjE3ODY1MDgxMDUsImlzcyI6Imh0dHBzOi8vaHVnZ2luZ2ZhY2UuY28ifQ.QyIunzINzDIGzSo0BZrfGYq27PatF82THGV0Q90OF5IW6VVj4OJ5gjc-hSJJVFMZDy5AvnTFxag9UiFES67sBQ","displayUrls":true,"splitSizeSummaries":[{"config":"default","split":"train","numRows":31175,"numBytesParquet":1629387251}]},"discussionsStats":{"closed":0,"open":1,"total":1},"fullWidth":true,"hasGatedAccess":true,"hasFullAccess":true,"isEmbedded":false,"savedQueries":{"community":[],"user":[]}}">
id int64 | name string | name_abbreviation string | decision_date string | docket_number string | court string | casebody string | judge1 string | judge2 string | judge3 string | exactly_three_caps int64 | extra_judges string | judge_designations string | opinions_count int64 | majority_count int64 | dissent_count int64 | concurrence_count int64 | other_count int64 | primary_citation string | reporter string | volume null | resource_url string | precedential_status string | 010combined_count int64 | 020lead_count int64 | 030concurrence_count int64 | 035concurrenceinpart_count int64 | 040dissent_count int64 | year int64 | month int64 | messages list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
4,649,395 | UNITED STATES of America, Plaintiff - Appellee v. Sacorey L. CLARK, Defendant - Appellant | United States v. Sacorey Clark | 2019-08-19 | 18-2119 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Arnold, Kelly", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2119\n ... | SMITH | ARNOLD | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649395/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,397 | UNITED STATES of America, Plaintiff - Appellee, v. Andrew SPALLEK, Defendant - Appellant. United States of America, Plaintiff - Appellee, v. Andrew Spallek, Defendant - Appellant. | United States v. Andrew Spallek | 2019-08-19 | 18-3113; 18-3114 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Gruender, Erickson", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-3113\n ___________________________\n\... | COLLOTON | GRUENDER | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649397/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,398 | Mary R. MEIER Plaintiff - Appellant v. ST. LOUIS, MISSOURI, CITY OF ; Doc's Towing, Inc. Defendants - Appellees St. Louis Board of Police Commissioners; Richard H. Gray, Member, St. Louis Board of Police Commissioners, in His Official Capacity; Thomas Irwin, Member, St. Louis Board of Police Commissioners, in His Offic... | Mary Meier v. St. Louis, Missouri, City of | 2019-08-19 | 18-1597 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Arnold, Kelly", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1597\n ... | SMITH | ARNOLD | KELLY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649398/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,399 | Marilyn Louise Beavers v. Arkansas Housing Authorities | 2019-08-19 | 18-1826 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before LOKEN, GRASZ, and STRAS, Circuit Judges.", "parties": "", "opinions": [{"author": "GRASZ, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | LOKEN | GRASZ | STRAS | 1 | {"LOKEN": ", Circuit", "GRASZ": ", Circuit", "STRAS": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649399/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,400 | KEARNEY REGIONAL MEDICAL CENTER, LLC, Plaintiff - Appellant, v. UNITED STATES DEPARTMENT OF HEALTH AND HUMAN SERVICES; Alex M. Azar, II, in His Official Capacity as Secretary of the United States Department of Health and Human Services, Defendants - Appellees. | Kearney Reg'l Med. Ctr., LLC v. U.S. Dep't of Health & Human Servs. | 2019-08-19 | 17-1207 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Beam, Colloton", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-1207\n ... | SMITH | BEAM | COLLOTON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649400/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,401 | Ibrahim Kariuki MWANGI Petitioner v. William P. BARR, Attorney General of the United States Respondent | Ibrahim Mwangi v. William P. Barr | 2019-08-19 | 18-1970 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Loken, Wollman, Stras", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1970\n ... | LOKEN | WOLLMAN | STRAS | 1 | {} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649401/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
8,444,063 | Furlandare SINGLETON, individually and as administrator of the estates of Dequan Singleton, Syndi Singleton, and Haylee Singleton, decedents Clyde Hatchett, individually and as administrator of the estate of Emily Beaver, decedent and Marilyn Louise Beavers, individually as an administrator of the estate of Marilyn Bea... | Singleton v. Ark. Hous. Auths Prop. & Cas. Self-Insured Fund, Inc. | 2019-08-19 | Nos. 18-1826; 18-1874; 18-1932 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Grasz, Loken, Stras", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/8415501/", "author_raw": ""}]} | GRASZ | LOKEN | STRAS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/8444063/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,712 | William ANDERSON, AS TRUSTEE FOR the NEXT-OF-KIN OF Jacob William ANDERSON (Deceased) Plaintiff - Appellant v. CITY OF MINNEAPOLIS; County of Hennepin; Hennepin Healthcare System, Inc.; Dr. Brian Mahoney, M.D., as Then-Medical Director of the HCMC Ambulance Service; Shana D. York; Anthony J. Buda; Raul A. Ramos, and Jo... | William Anderson v. City of Minneapolis | 2019-08-20 | 18-1941 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Arnold, Kobes, Shepherd", "parties": "", "opinions": [{"author": "KOBES, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-1941\... | ARNOLD | KOBES | SHEPHERD | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649712/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,713 | UNITED STATES of America, Plaintiff - Appellee, v. Rodolfo ANGUIANO, Jr., Defendant - Appellant. | United States v. Rodolfo Anguiano, Jr. | 2019-08-20 | 18-2154 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Kelly, Loken", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-215... | COLLOTON | KELLY | LOKEN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649713/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,714 | UNITED STATES of America, Plaintiff-Appellee, v. Rico v. GILLIAM, Defendant-Appellant. | United States v. Rico Gilliam | 2019-08-20 | 18-2017 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Kelly, Kobes, Smith", "parties": "", "opinions": [{"author": "SMITH, Chief Judge", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2017\n ... | KELLY | KOBES | SMITH | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649714/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,719 | Joshua RAWA, on Behalf of Himself and All Others Similarly Situated; Elisabeth Martin; Robert Ravencamp; Amy Ward; Cynthia Davies ; Christopher Abbott; Owen Olson; Jeannie A. Gilchrist; Zachary Sholar; Matthew Myers; John W. Beard, Jr.; Michael Overstreet, Plaintiffs-Appellees v. MONSANTO COMPANY, Defendant-Appellee v.... | Joshua Rawa v. James Migliaccio | 2019-08-20 | 18-2346 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Arnold, Kelly, Smith", "parties": "", "opinions": [{"author": "SMITH, Chief Judge", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2346\n ... | ARNOLD | KELLY | SMITH | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649719/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,649,720 | Jerry JANVRIN, Doing Business as J&J Trucking, Plaintiff-Appellee v. CONTINENTAL RESOURCES, INC., an Oklahoma Corporation, Defendant-Appellant | Jerry Janvrin v. Continental Resources, Inc. | 2019-08-20 | 17-2661 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Arnold, Kelly, Wollman", "parties": "", "opinions": [{"author": "WOLLMAN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 17-2661\n ... | ARNOLD | KELLY | WOLLMAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649720/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,654,453 | United States v. Orlando Dones-Vargas | 2019-08-21 | 18-2600 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before COLLOTON, KELLY, and ERICKSON, Circuit Judges.", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | COLLOTON | KELLY | ERICKSON | 1 | {"COLLOTON": ", Circuit", "KELLY": ", Circuit", "ERICKSON": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654453/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,654,455 | Geronimo DeLuna v. Mower County | 2019-08-21 | 18-1933 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before GRUENDER, BENTON, and GRASZ, Circuit Judges.", "parties": "", "opinions": [{"author": "GRASZ, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | GRUENDER | BENTON | GRASZ | 1 | {"GRUENDER": ", Circuit", "BENTON": ", Circuit", "GRASZ": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654455/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,471 | United States v. Alfredo Valquier | 2019-08-23 | 18-1636 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before COLLOTON, MELLOY, and SHEPHERD, Circuit Judges.", "parties": "", "opinions": [{"author": "MELLOY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | COLLOTON | MELLOY | SHEPHERD | 1 | {"COLLOTON": ", Circuit", "MELLOY": ", Circuit", "SHEPHERD": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655471/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,472 | Telescope Media Group v. Rebecca Lucero | 2019-08-23 | 17-3352 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before SHEPHERD, KELLY, and STRAS, Circuit Judges.", "parties": "", "opinions": [{"author": "STRAS, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | SHEPHERD | KELLY | STRAS | 1 | {"SHEPHERD": ", Circuit", "KELLY": ", Circuit", "STRAS": ", Circuit"} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655472/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,473 | Silvia Sepulveda-Rodriguez v. Metropolitan Life Insurance Co | 2019-08-23 | 18-1760 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before COLLOTON, BEAM, and SHEPHERD, Circuit Judges.", "parties": "", "opinions": [{"author": "BEAM, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No... | COLLOTON | BEAM | SHEPHERD | 1 | {"COLLOTON": ", Circuit", "BEAM": ", Circuit", "SHEPHERD": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655473/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,474 | Edward Blackorby v. BNSF Railway Company | 2019-08-23 | 18-2372 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before SHEPHERD, MELLOY, and GRASZ, Circuit Judges.", "parties": "", "opinions": [{"author": "MELLOY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | SHEPHERD | MELLOY | GRASZ | 1 | {"SHEPHERD": ", Circuit", "MELLOY": ", Circuit", "GRASZ": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655474/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,984 | Yuri CHACHANKO, Petitioner-Appellant v. UNITED STATES of America, Respondent-Appellee | Yuri Chachanko v. United States | 2019-08-26 | 18-2498 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Gruender, Benton, Grasz", "parties": "", "opinions": [{"author": "BENTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2498... | GRUENDER | BENTON | GRASZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655984/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,655,985 | United States v. Ira Alan Arias | 2019-08-26 | 18-2604 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before COLLOTON, KELLY, and ERICKSON, Circuit Judges.", "parties": "", "opinions": [{"author": "ERICKSON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | COLLOTON | KELLY | ERICKSON | 1 | {"COLLOTON": ", Circuit", "KELLY": ", Circuit", "ERICKSON": ", Circuit"} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655985/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,987 | Tasha McNeil v. Union Pacific Railroad Co | 2019-08-26 | 18-2333 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before COLLOTON, MELLOY, and SHEPHERD, Circuit Judges.", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | COLLOTON | MELLOY | SHEPHERD | 1 | {"COLLOTON": ", Circuit", "MELLOY": ", Circuit", "SHEPHERD": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655987/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,656,325 | United States v. Andre Dewberry | 2019-08-27 | 17-1649 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before GRUENDER, KELLY, and GRASZ, Circuit Judges.", "parties": "", "opinions": [{"author": "GRASZ, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | GRUENDER | KELLY | GRASZ | 1 | {"GRUENDER": ", Circuit", "KELLY": ", Circuit", "GRASZ": ", Circuit"} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4656325/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,656,752 | Katelyn Webb v. Chelsea Smith | 2019-08-28 | 18-2541 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before GRUENDER, ARNOLD, and STRAS, Circuit Judges.", "parties": "", "opinions": [{"author": "ARNOLD, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | GRUENDER | ARNOLD | STRAS | 1 | {"GRUENDER": ", Circuit", "ARNOLD": ", Circuit", "STRAS": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4656752/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,657,288 | United States v. Anthony Dunlap | 2019-08-29 | 18-2696 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before COLLOTON, BEAM, and SHEPHERD, Circuit Judges.", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | COLLOTON | BEAM | SHEPHERD | 1 | {"COLLOTON": ", Circuit", "BEAM": ", Circuit", "SHEPHERD": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657288/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,657,289 | Norman Unverzagt v. United States | 2019-08-29 | 17-2325 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before COLLOTON, GRUENDER, and ERICKSON, Circuit Judges.", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | COLLOTON | GRUENDER | ERICKSON | 1 | {"COLLOTON": ", Circuit", "GRUENDER": ", Circuit", "ERICKSON": ", Circuit"} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657289/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,657,290 | Juan Jose Inzunza REYNA, Petitioner, v. William P. BARR, Attorney General of the United States, Respondent. | Juan Inzunza Reyna v. William P. Barr | 2019-08-29 | 18-1614 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Melloy, Shepherd", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-16... | COLLOTON | MELLOY | SHEPHERD | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657290/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,657,291 | Jeffrey Klingenberg v. Vulcan Ladder USA, LLC | 2019-08-29 | 18-1742 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before BENTON, WOLLMAN, and GRASZ, Circuit Judges.", "parties": "", "opinions": [{"author": "BENTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | BENTON | WOLLMAN | GRASZ | 1 | {"BENTON": ", Circuit", "WOLLMAN": ", Circuit", "GRASZ": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657291/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,657,293 | Albert RIDGELL, Plaintiff - Appellee, v. CITY OF PINE BLUFF, a Public Body Corporate and Politic, Defendant - Appellant, Debe Hollingsworth, in Her Individual and Official Capacity as Mayor for the City of Pine Bluff, Arkansas, Defendant. | Albert Ridgell v. City of Pine Bluff | 2019-08-29 | 18-2663 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Gruender, Erickson", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-2663\n... | COLLOTON | GRUENDER | ERICKSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657293/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,657,740 | UNITED STATES of America, Plaintiff-Appellee, v. Michael B. LOWRY, Defendant-Appellant. | United States v. Michael Lowry | 2019-08-30 | 18-3109 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Smith, Kelly, Kobes", "parties": "", "opinions": [{"author": "KOBES, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 18-3109\n ... | SMITH | KELLY | KOBES | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657740/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,657,741 | Linda A. Miller v. Huron Regional Medical Center | 2019-08-30 | 18-1522 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before LOKEN, KELLY, and ERICKSON, Circuit Judges.", "parties": "", "opinions": [{"author": "KELLY, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | LOKEN | KELLY | ERICKSON | 1 | {"LOKEN": ", Circuit", "KELLY": ", Circuit", "ERICKSON": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657741/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,657,742 | Jessica Kampschroer v. Anoka County | 2019-08-30 | 18-1770 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before LOKEN, COLLOTON, and KELLY, Circuit Judges.", "parties": "", "opinions": [{"author": "COLLOTON, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | LOKEN | COLLOTON | KELLY | 1 | {"LOKEN": ", Circuit", "COLLOTON": ", Circuit", "KELLY": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657742/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,657,744 | Douglas Atwood v. Stephen Peterson | 2019-08-30 | 16-1152 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before LOKEN and WOLLMAN, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n No. 16-1152\n ... | LOKEN | WOLLMAN | LOKEN | 1 | {"LOKEN": ", Circuit", "WOLLMAN": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657744/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,657,745 | Arnold Fleck v. Joe Wetch | 2019-08-30 | 16-1564 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Before LOKEN, COLLOTON, and KELLY, Circuit Judges.", "parties": "", "opinions": [{"author": "LOKEN, Circuit Judge.", "type": "010combined", "text": "United States Court of Appeals\n For the Eighth Circuit\n ___________________________\n\n ... | LOKEN | COLLOTON | KELLY | 1 | {"LOKEN": ", Circuit", "COLLOTON": ", Circuit", "KELLY": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4657745/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
8,444,071 | Jessica Leah KAMPSCHROER Cory Patrick Kampschroer v. ANOKA COUNTY City of Anoka City of Apple Valley City of Arlington City of Baxter City of Belle Plaine Beltrami County Benton County City of Blaine City of Bloomington Blue Earth County City of Braham City of Brooklyn Center City of Brooklyn Park City of Brownton City... | Kampschroer v. Anoka Cnty. | 2019-08-30 | No. 18-1770 | U.S. Court of Appeals for the Eighth Circuit | {"judges": "Colloton, Kelly, Loken", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/8415510/", "author_raw": ""}]} | COLLOTON | KELLY | LOKEN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/8444071/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | |||
4,644,928 | James Cole v. Crst Van Expedited, Inc. | 2019-08-01 | 17-55606 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Johnnie B. Rawlinson and Andrew D. Hurwitz, Circuit Judges, and Stephen R. Bough,* District Judge.", "parties": "", "opinions": [{"author": "The Honorable Stephen R. Bough, United States District Judge for", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n ... | JOHNNIE B RAWLINSON | ANDREW D HURWITZ | STEPHEN R BOUGH | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4644928/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,645,328 | Sammy Page v. Audrey King | 2019-08-02 | 17-16364 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard A. Paez and Marsha S. Berzon, Circuit Judges, and Gary Feinerman, * District Judge Opinion by Judge Feinerman * The Honorable Gary Feinerman, United States District Judge for the Northern District of Illinois, sitting by designation.", "parties": "", "opinions": [{"author": "FEINERMAN, Distr... | RICHARD A PAEZ | MARSHA S BERZON | GARY FEINERMAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4645328/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,645,329 | David Flores-Vega v. William Barr | 2019-08-02 | 14-70690 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Michael Daly Hawkins and William A. Fletcher, Circuit Judges, and Richard Seeborg,* District Judge.", "parties": "", "opinions": [{"author": "W. FLETCHER, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n DAVI... | MICHAEL DALY HAWKINS | WILLIAM A FLETCHER | RICHARD SEEBORG | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4645329/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,645,330 | Ahmet Dogan v. Ehud Barak | 2019-08-02 | 16-56704 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Carlos T. Bea and Mary H. Murguia, Circuit Judges, and Stanley Allen Bastian, * District Judge.", "parties": "", "opinions": [{"author": "BEA, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n AHMET DOĞAN, ind... | CARLOS T BEA | MARY H MURGUIA | STANLEY ALLEN BASTIAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4645330/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,645,617 | State Farm Mutual Automobile v. Michael Mizuno | 2019-08-05 | 17-15947 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Kim McLane Wardlaw, Marsha S. Berzon, and Johnnie B. Rawlinson, Circuit Judges.", "parties": "", "opinions": [{"author": "United States Circuit Judge, Presiding", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nSTATE FARM MUTUAL ... | KIM MCLANE WARDLAW | MARSHA S BERZON | JOHNNIE B RAWLINSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4645617/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,646,696 | Agua Caliente Tribe of Cupeno v. Tara Katuk Mac Lean Sweeney | 2019-08-07 | 17-16838 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Danny J. Boggs, ** Richard A. Paez, and John B. Owens, Circuit Judges.", "parties": "", "opinions": [{"author": "PAEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n AGUA CALIENTE TRIBE OF CUPEÑO ... | DANNY J BOGGS | RICHARD A PAEZ | JOHN B OWENS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4646696/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,646,691 | Nimesh Patel v. Facebook, Inc. | 2019-08-08 | 18-15982 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould and Sandra S. Ikuta, Circuit Judges, and Benita Y. Pearson,* District Judge.", "parties": "", "opinions": [{"author": "IKUTA, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n NIMESH PATEL, Ind... | RONALD M GOULD | SANDRA S IKUTA | BENITA Y PEARSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4646691/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,646,692 | James Andrews v. Sirius Xm Radio, Inc. | 2019-08-08 | 18-55169 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: MILAN D. SMITH, JR. and MICHELLE T. FRIEDLAND, Circuit Judges, and STANLEY A. BASTIAN, * District Judge.", "parties": "", "opinions": [{"author": "M. SMITH, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n JA... | MILAN D SMITH JR | MICHELLE T FRIEDLAND | STANLEY A BASTIAN | 1 | {"MILAN D SMITH JR": ", Circuit", "MICHELLE T FRIEDLAND": ", Circuit", "STANLEY A BASTIAN": ", District"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4646692/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,646,693 | David Ehrman v. Cox Communications, Inc. | 2019-08-08 | 19-55658 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: MILAN D. SMITH, JR. and MICHELLE T. FRIEDLAND, Circuit Judges, and MICHAEL H. SIMON, * District Judge.", "parties": "", "opinions": [{"author": "M. SMITH, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nDAVID EHR... | MILAN D SMITH JR | MICHELLE T FRIEDLAND | MICHAEL H SIMON | 1 | {"MILAN D SMITH JR": ", Circuit", "MICHELLE T FRIEDLAND": ", Circuit", "MICHAEL H SIMON": ", District"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4646693/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,646,694 | Carlos Avena v. Kevin Chappell | 2019-08-08 | 14-99004 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sidney R. Thomas, Chief Judge, and Susan P. Graber and Milan D. Smith, Jr., Circuit Judges.", "parties": "", "opinions": [{"author": "Opinion by Chief Judge Thomas", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCARLOS J. AVENA... | SIDNEY R THOMAS | SUSAN P GRABER | MILAN D SMITH JR | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4646694/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,646,695 | Alex Bakalian v. Central Bank Rep. of Turkey | 2019-08-08 | 13-55664 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Kim McLane Wardlaw, Marsha S. Berzon, and Andrew D. Hurwitz, Circuit Judges.", "parties": "", "opinions": [{"author": "HURWITZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\nALEX BAKALIAN; ANAIS ... | KIM MCLANE WARDLAW | MARSHA S BERZON | ANDREW D HURWITZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4646695/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,141 | United States v. Royce Jobe | 2019-08-09 | 18-50204 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Kim McLane Wardlaw and Andrew D. Hurwitz, Circuit Judges, and Edward R. Korman, * District Judge.", "parties": "", "opinions": [{"author": "KORMAN, District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITED STATE... | KIM MCLANE WARDLAW | ANDREW D HURWITZ | EDWARD R KORMAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647141/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,142 | United States v. Lost Creek Trust | 2019-08-09 | 18-30122 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard A. Paez and Carlos T. Bea, Circuit Judges, and C. Ashley Royal, * District Judge.", "parties": "", "opinions": [{"author": "The Honorable C. Ashley Royal, United States District Judge for", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE ... | RICHARD A PAEZ | CARLOS T BEA | C ASHLEY ROYAL | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647142/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,143 | Steven Livaditis v. Ron Davis | 2019-08-09 | 14-99011 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould, Richard R. Clifton, and Carlos T. Bea, Circuit Judges.", "parties": "", "opinions": [{"author": "CLIFTON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nSTEVEN LIVADITIS, ... | RONALD M GOULD | RICHARD R CLIFTON | CARLOS T BEA | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647143/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,144 | Khalil Janjua v. Donald Neufeld | 2019-08-09 | 17-16558 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: J. Clifford Wallace, A. Wallace Tashima, and M. Margaret McKeown, Circuit Judges.", "parties": "", "opinions": [{"author": "TASHIMA, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nKHALIL JANJUA, ... | J CLIFFORD WALLACE | A WALLACE TASHIMA | M MARGARET MCKEOWN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647144/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,145 | Coeur D'Alene Tribe v. Steve Hawks | 2019-08-09 | 17-35755 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard R. Clifton and Consuelo M. Callahan, Circuit Judges, and Roger T. Benitez,* District Judge.", "parties": "", "opinions": [{"author": "CLIFTON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n COEUR D’... | RICHARD R CLIFTON | CONSUELO M CALLAHAN | ROGER T BENITEZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647145/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,146 | Billy Riley v. Timothy Filson | 2019-08-09 | 17-15335 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: M. Margaret McKeown, Milan D. Smith, Jr., and Andrew D. Hurwitz, Circuit Judges.", "parties": "", "opinions": [{"author": "McKEOWN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nBILLY RAY RILEY, ... | M MARGARET MCKEOWN | MILAN D SMITH JR | ANDREW D HURWITZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647146/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,497 | United States v. David Sainz | 2019-08-12 | 17-10310 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: A. Wallace Tashima and Milan D. Smith, Jr., Circuit Judges, and Lawrence L. Piersol, * District Judge.", "parties": "", "opinions": [{"author": "PIERSOL, District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITED... | A WALLACE TASHIMA | MILAN D SMITH JR | LAWRENCE L PIERSOL | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647497/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,498 | Paul G. v. Monterey Peninsula U.S.D. | 2019-08-12 | 18-16536 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder and Milan D. Smith, Jr., Circuit Judges, and Jed S. Rakoff,* District Judge.", "parties": "", "opinions": [{"author": "SCHROEDER, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n PAUL G., a ... | MARY M SCHROEDER | MILAN D SMITH JR | JED S RAKOFF | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647498/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,499 | Nagr v. Jeff Mangan | 2019-08-12 | 18-35010 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Susan P. Graber and Marsha S. Berzon, Circuit Judges, and John R. Tunheim, * District Judge.", "parties": "", "opinions": [{"author": "BERZON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nNATIONAL ASSOCIATION ... | SUSAN P GRABER | MARSHA S BERZON | JOHN R TUNHEIM | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647499/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,500 | Lamar Dawson v. Ncaa | 2019-08-12 | 17-15973 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sidney R. Thomas, Chief Judge, Andrew J. Kleinfeld, Circuit Judge, and George H. Wu,* District Judge.", "parties": "", "opinions": [{"author": "Opinion by Chief Judge Thomas", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n ... | SIDNEY R THOMAS | ANDREW J KLEINFELD | GEORGE H WU | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647500/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,501 | Howard Young v. Christian Pfeiffer | 2019-08-12 | 19-70286 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: M. Margaret McKeown, Jay S. Bybee, and John B. Owens, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n HOWARD ALLEN YOUNG, No... | M MARGARET MCKEOWN | JAY S BYBEE | JOHN B OWENS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647501/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,502 | Christopher Batterton v. Dutra Group | 2019-08-12 | 15-56775 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sidney R. Thomas, Chief Judge, and Andrew J. Kleinfeld and Jacqueline H. Nguyen, Circuit Judges.", "parties": "", "opinions": [{"author": "", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n CHRISTOPHER BATTERTON, ... | SIDNEY R THOMAS | ANDREW J KLEINFELD | JACQUELINE H NGUYEN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647502/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,503 | Aldf v. Usda | 2019-08-12 | 18-16327 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: A. Wallace Tashima, William A. Fletcher, and Marsha S. Berzon, Circuit Judges.", "parties": "", "opinions": [{"author": "W. FLETCHER, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nANIMAL LEGAL DEFENSE FUND, ... | A WALLACE TASHIMA | WILLIAM A FLETCHER | MARSHA S BERZON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647503/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
9,024,892 | ANIMAL LEGAL DEFENSE FUND v. UNITED STATES DEPARTMENT OF AGRICULTURE Animal and Plant Health Inspection Service | Animal Legal Defense Fund v. U.S. Dep't of Agric. | 2019-08-12 | No. 18-16327 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Fletcher", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9018145/", "author_raw": ""}]} | FLETCHER | FLETCHER | 0 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/9024892/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,845 | United States v. Parthava Nejad | 2019-08-13 | 18-30082 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: N. Randy Smith, Paul J. Watford, and Ryan D. Nelson, Circuit Judges.", "parties": "", "opinions": [{"author": "WATFORD, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 1... | N RANDY SMITH | PAUL J WATFORD | RYAN D NELSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647845/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,846 | United States v. Jose Hernandez-Martinez | 2019-08-13 | 15-30309 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Susan P. Graber and Marsha S. Berzon, Circuit Judges, and Eduardo C. Robreno, * District Judge.", "parties": "", "opinions": [{"author": "BERZON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF A... | SUSAN P GRABER | MARSHA S BERZON | EDUARDO C ROBRENO | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647846/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,647,847 | Ninth Inning, Inc. v. Directv, LLC | 2019-08-13 | 17-56119 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sandra S. Ikuta and N. Randy Smith, Circuit Judges, and George Caram Steeh III,* District Judge.", "parties": "", "opinions": [{"author": "IKUTA, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nIN RE NATIONAL FOO... | SANDRA S IKUTA | N RANDY SMITH | GEORGE CARAM STEEH III | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4647847/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
9,024,944 | United States v. Jose Luis HERNANDEZ-MARTINEZ, aka Efigenio Aispuro-Aispuro, United States of America v. Efigenio Aispuro-Aispuro, aka Jose Luis Hernandez-Martinez, United States of America v. Alejandro Renteria-Santana, United States of America v. Bartolo Favela Gonzales, aka Bartolo, aka Gordo, aka Jose Everando Sanc... | United States v. Hernandez-Martinez | 2019-08-13 | No. 15-30309; No. 15-30310; No. 15-30315; No. 15-30347; No. 15-30351; No. 15-30352; No. 15-30353; No. 16-30000; No. 16-30170; No. 16-30199; No. 16-30294; No. 17-30013; No. 15-30354; No. 15-30377; No. 15-30385; No. 16-30004; No. 15-30383; No. 15-30391; No. 16-30040; No. 16-30041; No. 16-30090; No. 16-30089; No. 16-30162 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Berzon", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9018198/", "author_raw": ""}]} | BERZON | BERZON | 0 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/9024944/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,648,240 | Timothy Barnes v. Chase Home Finance, LLC | 2019-08-14 | 18-35616 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Paul J. Watford and Eric D. Miller, Circuit Judges, and Barbara Jacobs Rothstein, * District Judge.", "parties": "", "opinions": [{"author": "Opinion by Judge Rothstein", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nTIMOTHY BA... | PAUL J WATFORD | ERIC D MILLER | BARBARA JACOBS ROTHSTEIN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4648240/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,648,241 | Louis Taylor v. County of Pima | 2019-08-14 | 17-16980 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder, Eugene E. Siler,* and Susan P. Graber, Circuit Judges.", "parties": "", "opinions": [{"author": "The Honorable Eugene E. Siler, United States Circuit Judge for the", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCU... | MARY M SCHROEDER | EUGENE E SILER | SUSAN P GRABER | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4648241/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,648,242 | Gerald Pizzuto, Jr. v. Randy Blades | 2019-08-14 | 16-36082 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Raymond C. Fisher, Ronald M. Gould and Johnnie B. Rawlinson, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nGERALD ROSS PIZZUTO, JR., No. 16-36... | RAYMOND C FISHER | RONALD M GOULD | JOHNNIE B RAWLINSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4648242/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,648,243 | Eva Gonzalez Romo v. William Barr | 2019-08-14 | 16-71559 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ferdinand F. Fernandez, Susan P. Graber, and John B. Owens, Circuit Judges.", "parties": "", "opinions": [{"author": "FERNANDEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nEVA ISABEL GONZALEZ ROMO, Eva ... | FERDINAND F FERNANDEZ | SUSAN P GRABER | JOHN B OWENS | 1 | {} | 3 | 0 | 1 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4648243/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,648,244 | Dennis Claiborne v. Blauser | 2019-08-14 | 16-16077 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Sidney R. Thomas, Chief Judge, Richard A. Paez, Circuit Judge, and Gary Feinerman, * District Judge.", "parties": "", "opinions": [{"author": "PAEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n DENNIS GERAL... | SIDNEY R THOMAS | RICHARD A PAEZ | GARY FEINERMAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4648244/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
9,024,987 | Louis TAYLOR, a single man v. COUNTY OF PIMA, a body politic City of Tucson, a body politic | Taylor v. Cnty. of Pima | 2019-08-14 | No. 17-16980 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Deliberations, Hurwitz, Vote", "parties": "", "opinions": [{"author": "", "type": "020lead", "text": "", "resource_uri": "https://www.courtlistener.com/api/rest/v4/opinions/9018241/", "author_raw": ""}]} | HURWITZ | HURWITZ | 0 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/9024987/ | Published | 0 | 1 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,648,689 | Jenny Flores v. William Barr | 2019-08-15 | 17-56297 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: A. Wallace Tashima, William A. Fletcher, and Marsha S. Berzon, Circuit Judges.", "parties": "", "opinions": [{"author": "BERZON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nJENNY LISETTE FLORES, et al., ... | A WALLACE TASHIMA | WILLIAM A FLETCHER | MARSHA S BERZON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4648689/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,648,690 | Apollo Education Group, Inc. v. National Union Fire Insurance | 2019-08-15 | 17-17293 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard A. Paez and Richard R. Clifton, Circuit Judges, and Gary S. Katzmann, * Judge.", "parties": "", "opinions": [{"author": "The Honorable Gary S. Katzmann, Judge for the United States", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH C... | RICHARD A PAEZ | RICHARD R CLIFTON | GARY S KATZMANN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4648690/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,090 | amazon.com Inc. & Subsidiaries v. Cir | 2019-08-16 | 17-72922 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: William A. Fletcher, Consuelo M. Callahan, and Morgan Christen, Circuit Judges.", "parties": "", "opinions": [{"author": "CALLAHAN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nAMAZON.COM, INC. & SUBSIDIARIES,... | WILLIAM A FLETCHER | CONSUELO M CALLAHAN | MORGAN CHRISTEN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649090/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,091 | United States v. Miguel Cano | 2019-08-16 | 17-50151 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Susan P. Graber and Jay S. Bybee, Circuit Judges, and M. Douglas Harpool,* District Judge.", "parties": "", "opinions": [{"type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITED STATES OF AMERICA, No. 17-50... | SUSAN P GRABER | JAY S BYBEE | M DOUGLAS HARPOOL | 1 | {} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649091/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,092 | United States v. Marcus Crum | 2019-08-16 | 17-30261 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: William A. Fletcher, Jay S. Bybee, and Paul J. Watford, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 17-30... | WILLIAM A FLETCHER | JAY S BYBEE | PAUL J WATFORD | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649092/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,093 | Josephine Tijerino v. Stetson Desert Project, LLC | 2019-08-16 | 18-16013 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ferdinand F. Fernandez, A. Wallace Tashima, and John B. Owens, Circuit Judges.", "parties": "", "opinions": [{"author": "TASHIMA, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nJOSEPHINE TIJERINO; TAMICKA ... | FERDINAND F FERNANDEZ | A WALLACE TASHIMA | JOHN B OWENS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649093/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,094 | John O'Rourke v. No. California Electrical | 2019-08-16 | 17-17419 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: J. Clifford Wallace, Eugene E. Siler, * and M. Margaret McKeown, Circuit Judges.", "parties": "", "opinions": [{"author": "WALLACE, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nJOHN J. O’ROURKE, ... | J CLIFFORD WALLACE | EUGENE E SILER | M MARGARET MCKEOWN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649094/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,095 | East Bay Sanctuary Covenant v. William Barr | 2019-08-16 | 19-16487 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.", "parties": "", "opinions": [{"type": "010combined", "text": "FOR PUBLICATION FILED\n UNITED STATES COURT OF APPEALS AUG 16 2019\n ... | TASHIMA | M SMITH | BENNETT | 1 | {"TASHIMA": ", Circuit", "M SMITH": ", Circuit", "BENNETT": ", Circuit"} | 3 | 0 | 1 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649095/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,096 | Aaron Senne v. Kansas City Royals Baseball | 2019-08-16 | 17-16245 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Michael R. Murphy, * Richard A. Paez, and Sandra S. Ikuta, Circuit Judges.", "parties": "", "opinions": [{"author": "PAEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nAARON SENNE; MICHAEL LIBERTO; ... | MICHAEL R MURPHY | RICHARD A PAEZ | SANDRA S IKUTA | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649096/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,426 | United States v. Vicente Cuevas-Lopez | 2019-08-19 | 17-10438 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard R. Clifton, Sandra S. Ikuta, and Michelle T. Friedland, Circuit Judges.", "parties": "", "opinions": [{"author": "FRIEDLAND, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, ... | RICHARD R CLIFTON | SANDRA S IKUTA | MICHELLE T FRIEDLAND | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649426/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,427 | United States v. Randly Begay | 2019-08-19 | 14-10080 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Dorothy W. Nelson, Richard R. Clifton, and N. Randy Smith, Circuit Judges.", "parties": "", "opinions": [{"author": "D.W. NELSON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, ... | DOROTHY W NELSON | RICHARD R CLIFTON | N RANDY SMITH | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649427/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,428 | United States v. Joseph Shayota | 2019-08-19 | 17-10270 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder, Diarmuid F. O’Scannlain, and Johnnie B. Rawlinson, Circuit Judges.", "parties": "", "opinions": [{"author": "Opinion by Judge O'Scannlain", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERI... | MARY M SCHROEDER | DIARMUID F O'SCANNLAIN | JOHNNIE B RAWLINSON | 1 | {} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649428/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,429 | Chemehuevi Indian Tribe v. John McMahon | 2019-08-19 | 17-56791 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Kim McLane Wardlaw and Andrew D. Hurwitz, Circuit Judges, and Edward R. Korman, * District Judge.", "parties": "", "opinions": [{"author": "HURWITZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCHEMEHUEVI INDI... | KIM MCLANE WARDLAW | ANDREW D HURWITZ | EDWARD R KORMAN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649429/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,767 | Michael Murray v. Mayo Clinic | 2019-08-20 | 17-16803 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould and Sandra S. Ikuta, Circuit Judges, and Benita Y. Pearson, * District Judge.", "parties": "", "opinions": [{"author": "PEARSON, District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMICHAEL J. MURRAY,... | RONALD M GOULD | SANDRA S IKUTA | BENITA Y PEARSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649767/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,768 | Michael Dorman v. the Charles Schwab Corporation | 2019-08-20 | 18-15281 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould and Sandra S. Ikuta, Circuit Judges, and Benita Y. Pearson, * District Judge.", "parties": "", "opinions": [{"author": "PEARSON, District Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nMICHAEL F. DORMAN,... | RONALD M GOULD | SANDRA S IKUTA | BENITA Y PEARSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649768/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,769 | Kristanalea Dyroff v. the Ultimate Software Group | 2019-08-20 | 18-15175 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Dorothy W. Nelson, Johnnie B. Rawlinson, and Carlos T. Bea, Circuit Judges.", "parties": "", "opinions": [{"author": "D.W. NELSON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nKRISTANALEA DYROFF, individually ... | DOROTHY W NELSON | JOHNNIE B RAWLINSON | CARLOS T BEA | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649769/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,770 | Jose Valtierra v. Medtronic Inc. | 2019-08-20 | 17-15282 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder and Milan D. Smith, Jr., Circuit Judges, and Jed S. Rakoff,* District Judge.", "parties": "", "opinions": [{"author": "SCHROEDER, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n JOSE VALTIE... | MARY M SCHROEDER | MILAN D SMITH JR | JED S RAKOFF | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649770/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,649,771 | Baldemar Zuniga v. William Barr | 2019-08-20 | 16-72982 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Andrew J. Kleinfeld and Michelle T. Friedland, Circuit Judges, and William H. Pauley III, * District Judge.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n BALDEMAR ZUNIGA... | ANDREW J KLEINFELD | MICHELLE T FRIEDLAND | WILLIAM H PAULEY III | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4649771/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,654,476 | United States v. Jeffrey Green | 2019-08-21 | 17-30227 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: A. Wallace Tashima, William A. Fletcher, and Marsha S. Berzon, Circuit Judges.", "parties": "", "opinions": [{"author": "BERZON, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, ... | A WALLACE TASHIMA | WILLIAM A FLETCHER | MARSHA S BERZON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654476/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,654,477 | Tommie Harris v. K. Harris | 2019-08-21 | 16-55083 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Ronald M. Gould and Jacqueline H. Nguyen, Circuit Judges, and Algenon L. Marbley, * District Judge.", "parties": "", "opinions": [{"author": "NGUYEN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n TOMMIE LEE ... | RONALD M GOULD | JACQUELINE H NGUYEN | ALGENON L MARBLEY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654477/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,654,478 | Hooman Panah v. Kevin Chappell | 2019-08-21 | 13-99010 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Kim McLane Wardlaw, Jacqueline H. Nguyen, and John B. Owens, Circuit Judges.", "parties": "", "opinions": [{"author": "OWENS, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nHOOMAN ASHKAN PANAH, ... | KIM MCLANE WARDLAW | JACQUELINE H NGUYEN | JOHN B OWENS | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654478/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,654,479 | Geraldine Nicholson v. Miguel Gutierrez | 2019-08-21 | 17-56648 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Andrew J. Kleinfeld, R. Guy Cole, Jr., * and Jacqueline H. Nguyen, Circuit Judges.", "parties": "", "opinions": [{"author": "NGUYEN, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nGERALDINE NICHOLSON; JOSE ... | ANDREW J KLEINFELD | R GUY COLE JR | JACQUELINE H NGUYEN | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654479/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,654,959 | Trina Ray v. County of Los Angeles | 2019-08-22 | 17-56581 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Kim McLane Wardlaw and Mark J. Bennett, Circuit Judges, and Kathleen Cardone, * District Judge.", "parties": "", "opinions": [{"author": "BENNETT, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nTRINA RAY, indivi... | KIM MCLANE WARDLAW | MARK J BENNETT | KATHLEEN CARDONE | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654959/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,654,960 | Fidelity National Financial v. Colin Friedman | 2019-08-22 | 17-15913 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: A. Wallace Tashima and Mary H. Murguia, Circuit Judges, and Robert N. Chatigny,* District Judge.", "parties": "", "opinions": [{"author": "TASHIMA, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nFIDELITY NATIONA... | A WALLACE TASHIMA | MARY H MURGUIA | ROBERT N CHATIGNY | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4654960/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,515 | Ussec v. Hui Feng | 2019-08-23 | 17-56522 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Kermit V. Lipez, * Kim McLane Wardlaw, and Andrew D. Hurwitz, Circuit Judges.", "parties": "", "opinions": [{"author": "LIPEZ, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n U.S. SECURITIES & EXCHANGE ... | KERMIT V LIPEZ | KIM MCLANE WARDLAW | ANDREW D HURWITZ | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655515/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,516 | Luis Pleitez-Lopez v. William Barr | 2019-08-23 | 16-73656 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Mary M. Schroeder and Susan P. Graber, Circuit Judges, and Michael H. Watson,* District Judge.", "parties": "", "opinions": [{"author": "GRABER, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n LUIS URIEL PLE... | MARY M SCHROEDER | SUSAN P GRABER | MICHAEL H WATSON | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655516/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,517 | FTC v. Qualcomm Incorporated | 2019-08-23 | 19-16122 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: TASHIMA, M. SMITH, and BENNETT, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION FILED\n UNITED STATES COURT OF APPEALS AUG 23 2019\n ... | TASHIMA | M SMITH | BENNETT | 1 | {"TASHIMA": ", Circuit", "M SMITH": ", Circuit", "BENNETT": ", Circuit"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655517/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,518 | Courtney Bird v. State of Hawaii | 2019-08-23 | 17-16076 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: Richard C. Tallman, Jay S. Bybee, and N. Randy Smith, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nCOURTNEY BIRD, No. 17-16076\n ... | RICHARD C TALLMAN | JAY S BYBEE | N RANDY SMITH | 1 | {} | 2 | 0 | 0 | 1 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655518/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,655,519 | Adree Edmo v. Corizon, Inc. | 2019-08-23 | 19-35017 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: M. Margaret McKeown and Ronald M. Gould, Circuit Judges, and Robert S. Lasnik, ** District Judge.", "parties": "", "opinions": [{"author": "B. Lynn Winmill, Chief District Judge, Presiding", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH C... | M MARGARET MCKEOWN | RONALD M GOULD | ROBERT S LASNIK | 1 | {} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4655519/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,656,014 | United States v. Town of Colorado City | 2019-08-26 | 17-16472 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: MICHAEL DALY HAWKINS and MILAN D. SMITH, JR., Circuit Judges, and BARBARA M. G. LYNN, * District Judge.", "parties": "", "opinions": [{"author": "M. SMITH, Circuit Judge:", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\n UNITE... | MICHAEL DALY HAWKINS | MILAN D SMITH JR | BARBARA M G LYNN | 1 | {"MICHAEL DALY HAWKINS": "MILAN D SMITH, JR , Circuit", "MILAN D SMITH JR": "MILAN D SMITH, JR , Circuit", "BARBARA M G LYNN": ", District"} | 1 | 0 | 0 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4656014/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... | ||||
4,656,015 | United States v. Davion Fitzgerald | 2019-08-26 | 18-10116 | U.S. Court of Appeals for the Ninth Circuit | {"judges": "Before: William A. Fletcher, Paul J. Watford, and Andrew D. Hurwitz, Circuit Judges.", "parties": "", "opinions": [{"author": "PER CURIAM", "type": "010combined", "text": "FOR PUBLICATION\n\n UNITED STATES COURT OF APPEALS\n FOR THE NINTH CIRCUIT\n\n\nUNITED STATES OF AMERICA, No. 18-1... | WILLIAM A FLETCHER | PAUL J WATFORD | ANDREW D HURWITZ | 1 | {} | 2 | 0 | 1 | 0 | 1 | null | https://www.courtlistener.com/api/rest/v4/clusters/4656015/ | Published | 1 | 0 | 0 | 0 | 0 | 2,019 | 8 | [
{
"content": "You are an expert legal coding assistant trained to classify U.S. federal Courts of Appeals\ncases using an adaptation of the Supreme Court Database (SCDB_2023_01) codebook. You follow the coding procedure\nin the codebook step by step and use the precise definitions of terms presented in the code... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.