Japan wrote its music down in the twelfth century and then lost most of it too

Claim (verbatim)

The survival of Tang-derived music in Japan is real but partial even inside Japan, and there is a twelfth-century yardstick to prove it. Fujiwara no Moronaga (1138-1192), the great late-Heian musician, compiled two enormous tablature encyclopedias: the Jinchi yoroku (仁智要録) for the so (zither) and the Sango yoroku (三五要録) for the biwa (lute), each collecting scores for a repertoire far larger than what any modern gagaku ensemble plays. Between Moronaga's day and the Meiji standardization the living Japanese repertoire itself contracted sharply - pieces notated in full in the twelfth century that no ensemble has performed for centuries. So even the survival-abroad tradition is a shrinking remnant of its own medieval self, and the shrinkage is countable because Moronaga wrote the fuller repertoire down. The mechanism is ritual-calendar attrition operating a second time, now inside Japan: as court patronage narrowed, the performed set collapsed toward a core, leaving the rest as silent tablature. Prediction restated: the number of pieces notated in the Jinchi yoroku and Sango yoroku greatly exceeds the modern performed togaku canon - the twelfth-century books hold at least several times as many pieces as are played today - so the medieval Japanese score-corpus is itself a record of subsequent Japanese loss.

Prediction clause (verbatim)

Prediction: counting the distinct pieces notated in Fujiwara no Moronaga's Jinchi yoroku (仁智要録) and Sango yoroku (三五要録) and comparing with the modern performed togaku canon, the two twelfth-century books will together contain at least three times as many distinct pieces as are performed in the modern repertoire, and a majority of the pieces they notate will have no continuous modern performance. Primary clause: the twelfth-century piece-count is at least three times the modern performed count. Disambiguation: pieces are counted by distinct title within each source and then unioned across the two books, with modal preludes and tuning-pieces excluded. Coverage guard: if the piece-inventory of both books cannot be established to at least 100 titles combined, the test voids.

Kill-dataset (verbatim)

Kill: the facsimile and critical editions of the Jinchi yoroku (仁智要録) and Sango yoroku (三五要録) - in the Nihon koten zenshu series and the scholarship of Steven G. Nelson and Allan Marett - for the medieval piece-inventory, against the modern togaku performance canon (Imperial Household Agency / Meiji senteifu) - count distinct pieces on each side.

Novelty / leakage triage

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

That Moronaga's two encyclopedias notate a Heian repertoire far exceeding the modern canon is the standard account - Nelson's history describes their scope, the Picken-Marett volumes are built on precisely these sources and repeatedly note pieces preserved there without modern performance, and Endo's study works the old scores directly - but the specific unioned piece-count against the Meiji-senteifu canon with a three-to-one threshold is not stated as a run result in anything located. The qualitative contraction is published; the ratio is un-run.

• S.G. Nelson, 'Court and religious music (1): history of gagaku and shomyo', in A. McQueen Tokita and D.W. Hughes (eds.), The Ashgate Research Companion to Japanese Music (Aldershot: Ashgate, 2008)
• L. Picken et al., Music from the Tang Court, 7 vols (1981-2000), which edit from the Jinchi yoroku and Sango yoroku
• Endo Toru, Heiancho no gagaku: kogakufu ni yoru togakukyoku no gakuriteki kenkyu (Tokyo: Tokyodo Shuppan, 2005)

Predictions

No prediction registered yet.