The antiphoner you can read but cannot hear

Claim (verbatim)

The Leon Antiphoner (Leon, Archivo Capitular, MS 8) is the fullest surviving witness to the Old Hispanic (Mozarabic) office, thousands of chant items copied in the tenth or eleventh century from a much older exemplar - and almost none of it can be sung, because the melodies are written in Visigothic neumes that float free above the text with no line, clef, or heighting to fix their pitches. The rite was suppressed at the Roman changeover of the 1080s before any scriptorium recopied its chant into a heightened, diastematic notation that a later reader could decode, so the signs record contour and ornament but not intervals. A handful of Old Hispanic melodies - on the order of twenty - can be recovered, and only because those particular chants happen to survive a second time in Aquitanian or Gregorian heightened notation that pins the pitches the Leon scribe left ambiguous. Recoverability is therefore an accident of double transmission, not a property of the chant: a melody lives today if and only if some other book, in a decipherable notation, preserved it too. This is loss of the strangest kind - a repertory fully present as writing and fully absent as sound, the maximal case of what a chant database can inventory as a text-and-genre record but can never index as a melody. Prediction restated: of everything the Leon Antiphoner notates, under one chant in twenty has a transcribable melody, an order of magnitude below the Gregorian norm where essentially every office chant has at least one pitch-readable witness.

Prediction clause (verbatim)

Prediction: taking the full chant inventory of the Leon Antiphoner (Leon, Archivo Capitular MS 8) as edited by Brou and Vives (Antifonario visigotico mozarabe de la catedral de Leon), the fraction of its notated chants for which a pitch-readable melody can be reconstructed - via a heightened, diastematic concordance in any surviving source - is under 5 percent, against a Gregorian comparison in which over 95 percent of office chants have at least one diastematic witness (primary clause: the under-5-percent transcribable fraction of the Old Hispanic inventory; the verdict follows it). A chant counts as transcribable only on an actual heighted concordance, not a conjectural restitution; coverage guard: the count is of chants the manuscript notates, and the un-transcribable remainder is precisely the melodic body the databases cannot represent, so the instrument undercounts the loss rather than the survival.

Kill-dataset (verbatim)

Kill (partly built): count the Leon Antiphoner's notated chant inventory (Brou-Vives edition) as denominator against the register of Old Hispanic melodies recoverable through heighted concordances (the roughly twenty transcribed in Randel and the Hispania Vetus literature) as numerator; the Cantus Database supplies the Gregorian diastematic-coverage baseline. Operation: a transcribable-to-total ratio.

Novelty / leakage triage

already answered in the literature

The two numbers the primary clause divides are both already in print: the Leon antiphoner's notated inventory runs to some 3,000 chants (the Brou-Vives edition; Randel's rite-wide index), and the register of Old Hispanic melodies recoverable through heighted concordances is the famous 'about twenty' pieces surviving in Aquitanian pitch-readable notation, stated as such throughout the survey literature. Twenty of three thousand is under one percent, so the under-5-percent transcribable fraction is guaranteed arithmetic on stated figures, and the near-total pitch-readability of the Gregorian office repertory is likewise textbook. This is the flagged Old Hispanic untranscribability leak: the conjecture restates the field's best-known fact with a threshold the fact guarantees.

• L. Brou & J. Vives (eds.), Antifonario visigotico mozarabe de la catedral de Leon (Monumenta Hispaniae Sacra; Barcelona-Madrid, 1953-1959)
• D.M. Randel, An Index to the Chant of the Mozarabic Rite (Princeton, 1973)
• D. Hiley, Western Plainchant: A Handbook (Oxford, 1993), Old Hispanic chapter on the c. 20 decipherable melodies
• S. Zapke (ed.), Hispania Vetus: Musical-Liturgical Manuscripts from Visigothic Origins to the Franco-Roman Transition (Bilbao, 2007)

Predictions

No prediction registered yet.