Half the canon was imported

Claim (verbatim)

A large stratum of classical Geʿez literature is not Ethiopian composition but translation from Arabic, made during the Solomonic centuries (thirteenth to fifteenth): the law code Fǝtḥa Nägäśt from al-Ṣafī Ibn al-ʿAssāl's Arabic nomocanon, the Zena Ayhud from the Hebrew-via-Arabic Josippon, the Ethiopian Synaxarion from the Copto-Arabic, and Barlaam and Josaphat (Baralam wä-Yǝwasǝf), the Physiologus, and much hagiography from Arabic Vorlagen. The mechanism is a translation pipeline running through the Christian-Arabic Christianity of Egypt, and it has an evidential consequence that outlives the pipeline: the existence of the Geʿez versions proves a substantial Christian-Arabic source library once accessible to Ethiopian translators, much of which is itself now poorly attested or lost on the Arabic side — so for a whole class of works the Geʿez corpus is the best surviving witness to texts translated into it, and the missing literature to be inferred lies partly on the source side, in the vanished Arabic exemplars. Prediction: within the classical Geʿez literary corpus (excluding biblical and purely local compositions), works demonstrably translated from Arabic will form a large minority — at least a quarter of titles — and for a meaningful subset of these the Geʿez version will be a better or the sole substantial witness relative to the surviving Arabic Vorlage (primary clause: at least a quarter of non-biblical classical Geʿez titles are Arabic-derived translations; the verdict follows it). Kill: the Beta maṣāḥǝft research environment's work-level records (which tag translation origin) and the EMML catalogues (Getatchew Haile), with the standard editions of the Fǝtḥa Nägäśt and the Zena Ayhud; the translation-share is a work-origin tabulation over the catalogued corpus.

Prediction clause (verbatim)

Prediction: within the classical Geʿez literary corpus (excluding biblical and purely local compositions), works demonstrably translated from Arabic will form a large minority — at least a quarter of titles — and for a meaningful subset of these the Geʿez version will be a better or the sole substantial witness relative to the surviving Arabic Vorlage (primary clause: at least a quarter of non-biblical classical Geʿez titles are Arabic-derived translations; the verdict follows it).

Kill-dataset (verbatim)

Kill: the Beta maṣāḥǝft research environment's work-level records (which tag translation origin) and the EMML catalogues (Getatchew Haile), with the standard editions of the Fǝtḥa Nägäśt and the Zena Ayhud; the translation-share is a work-origin tabulation over the catalogued corpus.

Novelty / leakage triage

anticipated in the literature — this exact test has never been run

That a large stratum of Solomonic-era (thirteenth- to fifteenth-century) Geʿez literature is translation from Christian Arabic — the Fǝtḥa Nägäśt, the Zena Ayhud, the Synaxarion, Barlaam and Josaphat, the Physiologus — is a textbook periodization of Ethiopian literary history (Cerulli). But the primary clause pins a proportion: at least a quarter of non-biblical classical Geʿez TITLES demonstrably Arabic-derived. Whether translations reach a quarter of titles depends on a work-origin tabulation over the catalogued corpus (which also holds a very large body of original Ethiopian hagiography and homiletics), and the conjecture's kill concedes this tabulation must be built from Beta maṣāḥəft's work-level records. The stratum is described; the quarter-share is uncomputed.

• Enrico Cerulli, La letteratura etiopica (3rd ed., Sansoni / Accademia, 1968)
• Paulos Tzadua (tr.) & Peter L. Strauss (ed.), The Fetha Nagast: The Law of the Kings (Addis Ababa, 1968)
• Beta maṣāḥəft work-level records; EMML catalogues (Getatchew Haile)

Predictions

No prediction registered yet.