AI-generated conjecture · below the evidence/publication boundary

The fiscal-year funeral

Status: No prior located

Status is derived only from the shepherd-authored triage/prediction data above -- community submissions and claims are a separate overlay and can never change it (see the participation panel below).

This is a proposed connection between two domains, generated by a language model. It is not an article and not evidence: it sits below the evidence/publication boundary. A quantitative prediction and a named kill-dataset are attached (when registered) so the claim stays falsifiable rather than merely evocative.

Claim (verbatim)

Archaeologists distinguish archives that were deliberately closed from archives killed by catastrophe using burn layers and collapsed walls — evidence that looting and old excavations have often destroyed. But bureaux lived by the fiscal calendar: accounts were balanced and files culled at the year-end audit, so an office wound down in an orderly way should leave its last dated tablets bunched at the end of the administrative year, while a sacked office should die in a random month. The conjecture is that the terminal-month distribution of an archive is by itself a destruction detector, recovering the manner of death for archives whose stratigraphy is lost. If it holds, hundreds of unprovenanced or badly excavated archives can be sorted into murdered and retired, and sack events can be dated to the month from the tablets alone.

Prediction clause (verbatim)

Across Ur III and Old Babylonian archives with at least 50 dated tablets, archives from independently documented destruction horizons will show terminal-month distributions consistent with uniformity, while other archives will concentrate at least 35 percent of their final-year tablets in the last two months of the local calendar. Primary clause, which decides the verdict: the two archive classes differ significantly in terminal-month distribution (chi-square or permutation test, p<0.01). Secondary clause: within the peaceful class, the terminal spike falls in the locally attested audit months.

Kill-dataset (verbatim)

BDTNS (Ur III archives with month-resolved dates) and Archibab (Old Babylonian archives, including those ending in the Samsu-iluna disruptions).

Nobody has run this test. The kill-data is named above. If you can run it — or you know the paper that already settles it — claim the kill or submit the prior. Kills and priors are credited here, by name, as they come in.

In the atlas

This conjecture is bridged, as an L1 lead, onto these Inferpedia subject pages.

Provenance

Run: Fresh agent generation · model: claude-fable-5

Composed blind by claude-fable-5 from internal knowledge only, with zero tool calls, and emitted directly as a single JSON text message.

Novelty / leakage triage

no prior formulation located (search dated 2026-07-10)

Searched archive-termination diagnostics from dated tablets (terminal-month distribution, audit-cycle bunching). Found archive-death and culling discussions but no prior use of final-year month distributions as a destruction-vs-retirement detector; no prior formulation located (search dated 2026-07-10).

Predictions

No prediction registered yet.

Weigh in

No community feedback yet.

Add your take

Posted immediately (spam is removed). Community feedback is never an adjudicated verdict and never changes this conjecture's triage label or status above.