Dated, shelved, and unsingable

Claim (verbatim)

For Byzantine chant before about 1100 we possess the manuscripts and have lost the ability to read them - a decipherment loss, distinct from physical loss and exactly countable. The early notations are of two kinds. Ekphonetic signs, ruling the solemn reading of scripture in lectionaries from the tenth century onward (studied in Høeg's MMB monograph), form a closed system whose precise melodic realization died with its cantors: the late sign-lists that survive teach the signs' names, not their tunes. The palaeo-Byzantine melodic notations - the Chartres and Coislin families, gathered in Strunk's Specimina Notationum Antiquiorum and the Monumenta Musicae Byzantinae facsimile volumes - are adiastematic: they record gesture and grouping but not interval size, so a melody can be recovered only where the same chant recurs in a later manuscript in interval-precise Middle Byzantine round notation, and every such recovery is a retro-projection of the twelfth century onto the tenth. The mechanism: notations invented as memory-aids for a living oral tradition encode exactly as much as a trained singer needed and no more; when the chain of singers broke, what remained on parchment fell below the decipherability threshold - the books survived at parchment's rate while their readability survived at the oral tradition's rate. The measurable consequence is a shelf of precisely dated, physically intact musical manuscripts, not one of which can be sung from its own page.

Prediction clause (verbatim)

Prediction: censusing the Greek notated sources datable before c. 1100 assembled in Strunk's Specimina and the MMB facsimiles and catalogues, at least 95 percent will carry ekphonetic or palaeo-Byzantine (Chartres/Coislin-family) notation that is melodically unreadable from the page alone, and the number of pre-1100 melodies recoverable without recourse to a post-1100 round-notation concordance will be zero (primary clause: the zero-independent-recoveries count over a >=20-source census; the verdict follows it); a melody counts as recoverable only if a published transcription derives its intervals from the pre-1100 source itself; the test voids if the census cannot assemble 20 pre-1100 notated sources.

Kill-dataset (verbatim)

Kill (not yet built): a dated-source census compiled from Strunk's Specimina Notationum Antiquiorum (MMB, 1966), Høeg's La notation ekphonétique (MMB, 1935), and the Monumenta Musicae Byzantinae facsimile series, each source scored for notation family and for whether any published transcription reads it without a round-notation concordance.

Novelty / leakage triage

already answered in the literature

The undecipherability is the stated foundation of Byzantine musical palaeography: Hoeg's monograph states that the ekphonetic system's melodic realization is lost, Strunk's Specimina and the MMB corpus present the Chartres and Coislin notations as adiastematic - readable only through Middle Byzantine round-notation concordances - and Wellesz's handbook says both in as many words. The primary zero (no pre-1100 melody recoverable from the page alone) is precisely this stated fact; the 95-percent and twenty-source dressing changes nothing about an answer the handbooks guarantee.

• E. Wellesz, A History of Byzantine Music and Hymnography, 2nd ed. (Oxford, 1961), on ekphonetic and early Byzantine notation
• O. Strunk, Specimina Notationum Antiquiorum (Monumenta Musicae Byzantinae VII, Copenhagen, 1966)
• C. Hoeg, La notation ekphonetique (Monumenta Musicae Byzantinae, Subsidia I.2, Copenhagen, 1935)

Predictions

No prediction registered yet.