NuBerea/classical-greek

Classical Greek Corpus

Ancient and classical Greek (grc) text segments drawn from the open scholarly corpora of the Perseus Digital Library and the OpenGreekAndLatin project — the classical/secular comparand within the NuBerea corpus estate, alongside its biblical, Second-Temple, and patristic Greek collections. Coverage runs from the archaic canon (Homer, Hesiod, the tragedians, the historians, Plato, Aristotle) through Hellenistic and imperial prose (Plutarch, Lucian, Galen), including scholia and technical writers. Each segment retains a CTS URN provenance pointer back to the source TEI edition.

The corpus is fully rebuildable from the public OGL/Perseus TEI sources. A fourth config, feats_grammar_bridge, joins the corpus's Greek tokens' morphological features (Case/Mood/Tense/Voice) to Smyth's Greek Grammar section pointers — computed grammar-reference metadata only, no grammar prose text. The complete technical specification (schemas, join contracts, per-config detail, provenance) is available in the gated SPEC.md alongside this card.

Configs

config	rows	what
grc_segments	304,847	classical Greek text segments (verse/section grain), CTS URN provenance
classical_authors_floruit	334	author floruit/composition-date manifest (segs, resolved, t_compose, basis) — the dating spine grc_segments cites
feats_grammar_bridge	23,290,665	one row per token (excl. PUNCT/SYM) across all 16 grc_segments volumes, joining upos/feats (Case/Mood/Tense/Voice) to Smyth Greek Grammar section_id/section_part/section_heading pointers — no Smyth prose text shipped or reconstructible

License and Attribution

Upstream Source	URL	License
OpenGreekAndLatin — First1KGreek	https://github.com/OpenGreekAndLatin/First1KGreek	CC BY-SA 4.0
PerseusDL — canonical-greekLit (Perseus Digital Library)	https://github.com/PerseusDL/canonical-greekLit	CC BY-SA (3.0/4.0)

The constituent texts are redistributed from these open scholarly corpora and carry their own CC BY-SA terms and editor attribution (e.g. Burnet for Plato, Diels for the doxographers), preserved per segment via the CTS URN reference into the source TEI. This dataset is a derivative work and is distributed under CC BY-SA 4.0 (attribution and share-alike required). Downstream use must retain these attributions and the share-alike license. Packaged by the NuBerea project (2026).

The feats_grammar_bridge config additionally references Smyth's Greek Grammar (public domain) by section id/part/heading POINTER only — no grammar-author prose text is shipped or reconstructible from this dataset. Consumers needing the actual grammar prose call the smy_sections/mcp__nuberea__smy_sections tool with the shipped section_id.