The quire chooses the scene

Claim (verbatim)

The quire chooses the scene. Which Shahnama episodes got painted is normally explained by narrative salience and patron taste. In commercial production the causality should often run the other way: painting spaces were reserved at ruling time, spread across the planned quires so a buyer flipping the book meets pictures at steady intervals, and whatever episode happened to fall on a reserved opening got illustrated. The picture grid was codicological before it was literary.

Prediction clause (verbatim)

Primary clause (the verdict follows it): within fifteenth- and sixteenth-century commercial Shahnamas holding at least 20 paintings, the coefficient of variation of inter-painting folio gaps is at most 0.6 times the coefficient of variation produced by placing the same number of paintings at that text's corpus-wide most-illustrated episodes — spacing is markedly more regular than salience-driven choice can produce. Secondary clauses: the probability that a painting depicts a corpus-rare episode rises with folio distance from the nearest corpus-popular episode (the grid demanded a picture and took what was there); and royal manuscripts fail the regularity test, since their programmes really were chosen scene-first.

Kill-dataset (verbatim)

Kill: the Cambridge Shahnama Project corpus, which records folio positions and episode identifications for thousands of paintings across hundreds of manuscripts.

Novelty / leakage triage

provisional — model-triaged, shepherd review pending

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. Provisional reading: Adjacent (closely related prior work exists).

Encyclopaedia Iranica's 'Sah-nama ii. Illustrations' states outright that scene choice in Ilkhanid Shahnamas 'depended on codicological considerations, particularly the layout of the folios and the sequence of their transcription and illustration', with the scribe leaving pre-planned blank spaces -- so the item's surprising connection (placement is codicological before literary) is documented. Not located: the item's decidable prediction -- that the coefficient of variation of inter-painting folio gaps in 15th-16th-century commercial Shahnamas is <= 0.6x the CV a salience-driven programme produces (a browsing buyer's regular grid), with corpus-rare episodes rising with distance from popular ones and royal manuscripts failing the regularity test -- measured over the Cambridge Shahnama Project. Codicological priority documented; the regular-spacing measurement and the commercial/royal contrast are unrun.

• Encyclopaedia Iranica, 'Sah-nama ii. Illustrations' — Scene choice depended on codicological considerations and the sequence of transcription/illustration; the item's connection, published.
• Cambridge Shahnama Project corpus — Folio positions and episode identifications for thousands of paintings; the kill dataset for the spacing-variance test.

Predictions

No prediction registered yet.