The wastepaper clock

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)

Offices in Greco-Roman Egypt did not keep files forever: they weeded, and weeded sheets were reused — flipped over for letters and school exercises or sold off in bulk. Every papyrus carrying a dated document on the front and a dated reuse on the back is therefore the fossil of one weeding decision, and this conjecture claims those decisions followed policy, not chance: reuse lags should cluster at a modal retention horizon on the order of a working generation or an audit cycle (roughly 20-40 years), with different modes for different document classes, because a register expires when its audit does while a contract must outlive its obligations. If this holds, the retention schedules of ancient bureaucracy — its records-management rules — can be recovered from scrap paper, giving the archive-as-selector a written-in-the-object chronology.

Prediction clause (verbatim)

Primary clause: in papyri.info recto-verso pairs where both texts are independently datable, the reuse-lag distribution for administrative registers and accounts is unimodal with a mode between 15 and 45 years and an interquartile range under 40 years. Secondary clause: contracts show a significantly longer modal lag than registers, and cartonnage-derived documents show lags consistent with bulk disposal events rather than continuous discard.

Kill-dataset (verbatim)

papyri.info: doubly dated recto/verso texts and cartonnage provenance records.

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 with zero tool use and no information ingress of any kind; the packet was emitted as a single JSON text message for the orchestrator to persist.

Novelty / leakage triage

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

Papyrus reuse of obsolete documents is well documented and papyrologists use rough recto-verso lag rules for dating, anticipating the direction, but the reuse-lag distribution computed at corpus scale as a recoverable retention schedule by document class is un-run.

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.