Rank buys copies

Claim (verbatim)

How widely a chant was copied was set less by its age or beauty than by the rank of the feast it served. A duplex or totum duplex feast was mandated across a whole province and demanded a full proper, so its chants were copied into every choirbook in the region; a three-lessons commemoration carried proper chant only where the local cult was strong, and a bare ferial entry demanded almost nothing new. Rank was a copying instruction. The mechanism is administrative: the grade a chapter assigned a feast determined how many books had to contain its music, so a chant's surviving concordance breadth - the number of sources that carry it - should be a monotone function of its feast's liturgical rank, independent of the chant's date or quality. This turns feast rank into a survival lottery run in advance: high-rank chants were insured by mandated multiplication, low-rank proper chants were left as near-unica the moment they were composed. Distinguish this from the age of a feast: a young high-rank feast still spreads wide, an old low-rank commemoration stays narrow, so it is rank, not antiquity, that the breadth tracks. Prediction restated: the median concordance breadth of top-rank feast chants exceeds that of the lowest-ranked proper commemorations by at least a factor of ten, with a monotone gradient across the intermediate grades.

Prediction clause (verbatim)

Prediction: in Cantus Index, the median cross-source concordance count of chants proper to top-rank feasts (duplex and above) exceeds the median for chants of the lowest-ranked proper commemorations by at least a factor of 10, and the median breadth rises monotonically across the intervening rank grades (primary clause: the at-least-10x top-to-bottom median-breadth ratio with a monotone gradient; the verdict follows it). Rank is read from the source rubrics and the test is run within comparable feast ages to hold antiquity fixed; coverage guard: chants shared with the common core are excluded so the gradient measures proper chant, and the test voids if rank can be assigned for fewer than 200 chants.

Kill-dataset (verbatim)

Kill (partly built): Cantus Index concordance counts joined to feast-rank as read from the Cantus Database source rubrics, computing median concordance breadth per rank grade; the operation is a rank-to-breadth gradient over the concordance-count distribution, with feast age held fixed.

Novelty / leakage triage

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

That feast grade governed provision - duplex feasts mandating full propers across a province, lesser commemorations only local provision - is standard liturgiology, and CAO and LMLO publish the raw inventories; but no study computes median cross-source concordance breadth per rank grade over Cantus Index, and the monotone rank-to-breadth gradient with a 10x top-to-bottom ratio is genuinely un-run arithmetic - rank is not a database field and must be read from rubrics, as the prediction itself concedes.

• J. Harper, The Forms and Orders of Western Liturgy from the Tenth to the Eighteenth Century (Oxford, 1991)
• R.-J. Hesbert, Corpus Antiphonalium Officii, 6 vols (Rome, 1963-1979)
• A. Hughes, Late Medieval Liturgical Offices: Resources for Electronic Research, 2 vols (Toronto, 1994-1996)

Predictions

No prediction registered yet.