Colophons don't heap

Claim (verbatim)

Colophons don't heap. This connects a demographer's instrument with scribal diplomatics. Dates recalled or reconstructed from memory heap on round numbers; dates written down on the day itself do not. A colophon is written at the moment of completion, often with weekday and month attached for the closing blessing, while a biographical dictionary's death date is often fixed decades later from hearsay and inference. The two great date corpora of the Islamic book world should therefore sit at opposite ends of the heaping spectrum, and the size of the gap measures how much of the civilization's chronology was contemporaneous record versus retrospective reconstruction.

Prediction clause (verbatim)

Among dated colophons, the share of AH years ending in 0 or 5 stays at or below 23% (against a 20% uniform null), while death dates in the major biographical dictionaries show a 0/5-ending share of at least 28%, and heaping in the dictionaries increases with the interval between the subject's death and the compiler's lifetime. Primary clause: the conjunction of the colophon share at or below 23% and the dictionary share at or above 28%; the verdict follows it.

Kill-dataset (verbatim)

Kill: FIHRIST dated-manuscript records (thousands of colophon years) against death dates extracted from al-Dhahabi and al-Safadi in the OpenITI corpus.

Novelty / leakage triage

no prior located yet (provisional)

A provisional first pass authored by the model (Opus), not yet confirmed by the shepherd. It carries the same dated-search requirements as an authoritative verdict but is excluded from every headline figure and cannot underwrite a prediction until a shepherd confirms it.

[Independent blind re-audit 2026-07-08 (2nd pass, generator-independent), confidence medium. Double-confirmed: both the provisional pass and this re-audit located no prior formulation of the specific operationalized claim.] The instrument (digit-preference / age-heaping, Whipple's index) is textbook, and it has been applied to historical populations (A'Hearn et al.; Poland-Lithuania and Ireland studies) and, in cognate form, to date digit preference (Spennemann) and to Roman epitaph age-rounding (Duncan-Jones). The two Islamic date corpora are separately well studied (colophon diplomatics; Romanov/Bulliet on biographical death dates). But I located no work that runs the specific comparison proposed: measuring the share of AH years ending 0/5 in dated colophons (predicted <=23%, contemporaneous record) versus death dates in biographical dictionaries (predicted >=28%, retrospective reconstruction), and testing that dictionary heaping grows with the death-to-compiler interval. Method is standard; this specific colophon-vs-dictionary heaping contrast as a measure of contemporaneous-vs-reconstructed chronology is unlocated. Romanov analyzed death-date distributions but is not reported to have measured round-number heaping against colophons.

• Whipple's index (digit preference / age heaping measure)
• A'Hearn et al. and related age-heaping-as-numeracy literature (e.g. 'Rethinking age heaping', Economic History Review 2022)
• Maxim Romanov, computational reading of al-Dhahabi's death-date data (no reported heaping analysis)

Predictions

No prediction registered yet.