AI-generated conjecture · below the evidence/publication boundary

The diversified canon

Status: Anticipated · untested

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)

Finance has one iron rule: diversification protects against local disaster, so the portfolio spread across many markets outlives the one concentrated in a single boom town. Sumerian literature faced exactly this problem — cities burned, and a composition's manuscripts burned with them. The conjecture: survival into the first-millennium stream of tradition is predicted by a work's geographic spread across Old Babylonian sites more strongly than by its total copy count, because a Nippur-only blockbuster could be extinguished by Nippur's misfortunes while a modestly copied but widely planted text could not. If it holds, the received canon is revealed as the diversified portfolio rather than the ancient bestseller list, and the disappearance of works we know were locally beloved becomes an expected concentration-risk casualty.

Prediction clause (verbatim)

In a logistic regression over Old Babylonian Sumerian and Akkadian literary compositions predicting any post-1200 BCE attestation, the number of distinct OB findspot sites will be a significant positive predictor, and once site count is included, total OB copy count will add no significant explanatory power. Primary clause, which decides the verdict: nested model comparison (likelihood ratio or AIC) shows site count dominating copy count. Secondary clause: among compositions with 10 or more OB copies, single-site works survive at less than half the rate of works attested at three or more sites.

Kill-dataset (verbatim)

CDLI (provenance of Old Babylonian literary manuscripts) crossed with ORACC first-millennium literary and scholarly catalogues.

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

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

Searched geographic spread as a predictor of text survival. Kestemont et al. (Science 2022) showed evenness/spread of copies (island vs. continental literatures) predicts survival of medieval works — the same concentration-risk logic — but no application to Old Babylonian findspot counts vs. first-millennium survival was located.

Kestemont et al., Forgotten books: The application of unseen species models to the survival of culture, Science 375 (2022)

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.