Gregory read passions we cannot

Claim (verbatim)

Gregory of Tours is the sixth century's best witness to a hagiographic library we no longer have. His Gloria martyrum and its sibling books of miracles constantly cite passiones as physical texts: he read them, disbelieved some, had copies made, and once - for the Seven Sleepers of Ephesus - produced a Latin version himself, as he says, with a Syrian interpreting. He reports (Gloria martyrum 63) that the clergy of Troyes long kept the martyr Patroclus at arm's length precisely because they lacked a written passio, until a copy arrived from Italy - cult waiting on text, in a working bishop's own account. Each such citation is a dated attestation: this passio existed in Gaul in the 580s. Run the census forward and the claim is grim: a substantial share of the passiones Gregory handled either do not survive at all or survive only as Carolingian-and-later rewritings, the Merovingian text he actually read having perished inside the very rewriting economy that kept the saint alive. Gregory is thus a terminus for measuring twelve centuries of subtraction: the difference between his shelf and ours, computed title by title, is one of the few places where the loss rate of early hagiography can be observed rather than assumed.

Prediction clause (verbatim)

Prediction: enumerating the passiones that the Gloria martyrum explicitly cites as written texts Gregory read or possessed (the census taken from Krusch's index in MGH SRM I.2 and the notes to Van Dam's translation), at least one third of the cited passiones will prove to have either no surviving text under the saint's BHL entry or no version dated before 700 in the standard editions and the Bollandist literature (primary clause: the >=1/3 lost-or-later rate; the verdict follows it); disambiguation: a citation counts only where Gregory names a written passio or historia, not mere knowledge of a martyr; the test voids if fewer than 15 such explicit citations are enumerable.

Kill-dataset (verbatim)

Kill (partly not yet built): the citation census runs now from Krusch's MGH SRM I.2 apparatus and Van Dam's Glory of the Martyrs; the decisive join - an earliest-version dating per cited saint's BHL chain - exists only as scattered judgements in Analecta Bollandiana and the editions, and must be assembled into a dated-version table before the one-third clause can be scored.

Novelty / leakage triage

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

Gregory's handling of written passiones is documented piecemeal - Krusch's apparatus and Van Dam's notes identify the cited texts case by case, and Lanery's Hagiographies chapter now supplies dated version-chains for the Italian passions that dominate his shelf - but nobody has assembled the explicit-citation census and scored earliest-surviving-version dates across it; the >=1/3 lost-or-later rate exists nowhere in print. The kill note itself concedes the dated-version table must be built: register-before-compute.

• B. Krusch (ed.), Gregorii Turonensis Opera: Miracula et opera minora (MGH SRM I.2, Hanover, 1885)
• R. Van Dam (trans.), Gregory of Tours: Glory of the Martyrs (Translated Texts for Historians, Liverpool, 1988)
• C. Lanery, 'Hagiographie d'Italie (300-550). I. Les Passions latines composees en Italie', in G. Philippart (ed.), Hagiographies V (Turnhout, 2010)

Predictions

No prediction registered yet.