The strait runs one way

Claim (verbatim)

Hagiography crossed the Mediterranean like a trade wind: one way. The Greek classics came west early and massively - Athanasius's Antony in Evagrius's Latin within a generation, the desert fathers wholesale, then the ninth- and tenth-century translation industries of papal Rome (Anastasius Bibliothecarius) and Naples (John the Deacon, Guarimpotus) - so the BHL is studded with items marked as versions from the Greek. The eastward channel barely existed: the Greek church, liturgically self-sufficient and serenely incurious about western sanctity, admitted few Latin saints to its calendar, and the famous exception - Pope Zacharias's eighth-century Greek version of Gregory's Dialogues, which carried Benedict east - is remembered precisely because it is an exception. Both repertoria are indexed for this: the BHL cross-references its Greek originals to BHG numbers, and the BHG marks its rare Latin-derived items. The asymmetry of surviving traffic is therefore countable from the instruments alone, and the claim is that it is not a lean but a regime: an order of magnitude or more, the signature of a prestige gradient that decided, century by century, whose saints needed whose language - with the BHO standing by to show the same wind blowing through Syriac and Coptic.

Prediction clause (verbatim)

Prediction: counting cross-references between the repertoria - BHL items marked as translations from the Greek (carrying or resolvable to a BHG number) against BHG items marked as translations from the Latin (Halkin's third edition with the 1984 Novum Auctarium) - the westward count will exceed the eastward count by at least 10:1, and the eastward absolute count will stay under 25 items (primary clause: both figures jointly; the verdict follows them); disambiguation: genuinely bilingual early dossiers count for neither side unless the repertoria themselves mark translation direction; the test voids if the BHL's ex-graeco marks cannot be enumerated across at least half the alphabet.

Kill-dataset (verbatim)

Kill: the BHL's ex graeco cross-references tallied against the BHG's ex latino items (Halkin's 1957 edition and the 1984 Novum Auctarium), with the IRHT's Pinakes database as the check on the Greek side - a two-column census read off the repertoria's own indices.

Novelty / leakage triage

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

The one-way gradient is printed doctrine with enumerations on both sides - Siegmund inventories the Latin translations from the Greek, Dekkers's Sacris Erudiri census shows how small the Greek-from-Latin corpus is, and Berschin narrates the regime - so the direction is beyond doubt. But the pinned two-repertoria count (BHL ex-graeco marks against the BHG and Novum Auctarium ex-latino items, with the joint 10:1 floor and sub-25 eastward ceiling) has not been run as such, and the eastward hagiographic subset is not separately tabulated in Dekkers; adjacent, with Siegmund and Dekkers as the prior censuses any resolution must report deltas against.

• A. Siegmund, Die Uberlieferung der griechischen christlichen Literatur in der lateinischen Kirche bis zum zwolften Jahrhundert (Munich, 1949)
• E. Dekkers, 'Les traductions grecques des ecrits patristiques latins', Sacris Erudiri 5 (1953)
• W. Berschin, Griechisch-lateinisches Mittelalter (Bern, 1980; Engl. Greek Letters and the Latin Middle Ages, Washington, 1988)

Predictions

No prediction registered yet.