He searched five provinces for a dying book

Claim (verbatim)

The most quoted entry in Hunayn's Risala is an obituary he did not know he was writing. Of Galen's On Demonstration - the fifteen-book logic and epistemology of science that the Baghdad philosophers wanted more than any other Galenic work - he reports: I searched for it earnestly, and travelled for it through Mesopotamia, all of Syria, Palestine and Egypt until I reached Alexandria, and found nothing except about half of it, in disorder and incomplete, in Damascus. That testimony is a data point, not a lament: an 856 census reading telling us the Greek stock of the work was already near zero a century into the translation movement. The work then died in Arabic too - what remains are quotations and testimonia gathered since Iwan von Müller's 1895 study - making it the exception that maps the rule: translation could only rescue what still had a live exemplar in the ninth century. The Risala is full of such exemplar testimony: works Hunayn found easily, works he found defective, works he could not find at all. Read as a scarcity gauge, his remarks should PREDICT the double deaths: works he already flagged as hard to find should be disproportionately the ones extinct today in Greek AND Arabic alike, while the works he handled without comment survived in at least one stream. The bibliographer's frustrations of 856 are a forecast of the next millennium's losses.

Prediction clause (verbatim)

Prediction: coding the Risala's entries by Hunayn's own exemplar testimony into a scarce class (explicit search trouble, single or defective exemplar) and an unremarked class, the scarce class's rate of complete survival in NEITHER Greek nor Arabic will be at least three times the unremarked class's rate (primary clause: the >=3x double-death ratio; the verdict follows it), with On Demonstration itself in the dead-in-both cell, and fewer than one in twenty unremarked works dead in both; scarcity is coded solely from the Risala's text (Lamoreaux's translation) blind to modern status, survival status is read from Fichtner's Corpus Galenicum and Sezgin GAS III, and the test voids for coverage if the scarce class numbers fewer than 8 works.

Kill-dataset (verbatim)

Kill: Lamoreaux's 2016 Risala for the exemplar-testimony coding, joined to Fichtner's Corpus Galenicum and Sezgin GAS III for per-work survival, with von Müller's 1895 fragment study for the On Demonstration dossier - a two-class double-death rate comparison.

Novelty / leakage triage

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

The On Demonstration dossier is textbook - Hunayn's five-province search report (Risala no. 115), von Muller's fragment study, death in both languages - and per-work survival is tabulated in Fichtner and GAS III; but the instrument is new: coding all 129 entries by Hunayn's own exemplar testimony, blind to modern status, and comparing double-death rates between a scarce and an unremarked class. No located study runs the Risala as a scarcity gauge with a quantified two-class ratio; the anecdotal form (translation saved only what still had a live exemplar) is a commonplace, the >=3x test is not.

• I. von Muller, Ueber Galens Werk vom wissenschaftlichen Beweis, Abhandlungen der Koeniglich Bayerischen Akademie der Wissenschaften 20 (1895)
• G. Bergstraesser, Hunain ibn Ishaq ueber die syrischen und arabischen Galen-Uebersetzungen (Leipzig, 1925), no. 115
• J.C. Lamoreaux, Hunayn ibn Ishaq on His Galen Translations (Provo, 2016)

Predictions

No prediction registered yet.