Ptolemy's Fingerprint on Plainchant

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)

Historians of astronomy identify the ancestry of a medieval table not by its words but by its parameters: the buried numerical constants betray the source tradition even when everything else is rewritten, and DISHAS was built to fingerprint them. Gregorian chant is also a parameterized system — every antiphon carries a mode and a differentia (the formulaic ending that links it to its psalm tone) — and the Cantus Database records these assignments for hundreds of thousands of chant entries. I conjecture that rare differentia-to-antiphon assignments behave exactly like rare astronomical parameters: two manuscripts sharing several rare assignments are transmission-related, and clustering manuscripts on this signature reconstructs the genealogy of liturgical use without reading a single melody. The mechanism is that a differentia assignment is an arbitrary convention with many workable alternatives, so agreement in rare choices is evidence of copying, not convergence — precisely the logic of parameter fingerprinting. If this holds, the astronomers' instrument hands chant scholars a stemmatics of liturgical practice that operates at database scale.

Prediction clause (verbatim)

Clustering Cantus manuscripts by shared rare differentia-to-antiphon assignments (assignments occurring in fewer than 5% of sources) will recover the known liturgical families — Cistercian, Franciscan, monastic versus secular cursus — with an adjusted Rand index of at least 0.6 against expert classification. Primary clause (the verdict follows it): adjusted Rand index >= 0.6. Secondary clause: the clustering outperforms a null model built on common assignments alone (permutation test, p < 0.01).

Kill-dataset (verbatim)

The Cantus Database of chant manuscripts with mode and differentia fields; kill is a statistical test (adjusted Rand index against known liturgical families with a permutation null).

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, emitted as a single JSON text message per the fresh-lane blindness protocol.

Novelty / leakage triage

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

The Differentiae Database (170+ manuscripts, 1,400+ unique differentiae) was built precisely for cross-manuscript differentia analysis and chant-transmission questions, and Cantus offers manuscript-clustering dendrogram tools, so the target is squarely anticipated; the specific rare-assignment clustering scored by adjusted Rand index against liturgical families 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.