The models a master listed, and buried

Claim (verbatim)

Bana opens the Harsacarita (c. 640) with a roll-call of his poetic ancestors. The opening verses salute, among others, the Brhatkatha of Gunadhya, the Vasavadatta of Subandhu, the Prakrit anthology of the Satavahana, Pravarasena's Setubandha, Bhattara Hariscandra, the plays reckoned to Bhasa, and Kalidasa, a canon assembled by a working poet naming the summits he wrote beneath. Fourteen centuries later that list is a survival audit waiting to be run, because Bana vouches that every name on it was a living, admired body of work in the seventh century. The prediction is that his roll of honour is now a mixed graveyard: the securely surviving whole works are a minority, and the losses fall hardest on the story-literature and the older names, while the school-canonized poets, Kalidasa above all, come through. A poet's own list of masters, dated by his life, converts into a timed measurement of how much even the acknowledged best of a literature fails to reach us. The instrument is not a catalogue's guesswork but a great author's sworn testimony to what he had read.

Prediction clause (verbatim)

Prediction: resolving the distinct poets and works named in the Harsacarita's opening praise-verses, together with the parallel salutation in the Kadambari, against the New Catalogus Catalogorum, at least one-third will survive only as fragments, quotation, or a lost but attested title rather than as a complete independently transmitted work; the at-least-one-third non-whole-survival share is the deciding clause. Each name is scored once, disputed identifications such as the Adhyaraja are set aside rather than counted as lost, and the test voids for coverage if fewer than nine distinct predecessors can be securely read from the verses.

Kill-dataset (verbatim)

Kill: the GRETIL and SARIT e-text and the standard editions of Bana's Harsacarita (Ucchvasa 1 mangala) and Kadambari, enumerating the named predecessors and scoring each for whole-work survival against the New Catalogus Catalogorum.

Novelty / leakage triage

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

Bana's predecessor-roll is among the most-worked testimonia in Sanskrit literary history: Kane's Harsacarita edition annotates the introductory verses name by name and the standard histories score the survivals informally, so every per-name verdict (Brhatkatha lost, Hariscandra lost, Subandhu, Pravarasena and Kalidasa extant, Bhasa disputed) is in print. But the aggregated fraction has never been stated, and its outcome hinges on scoring conventions over a small denominator: with only the Brhatkatha and Hariscandra securely lost among roughly eight scoreable names, the one-third threshold turns on how the disputed Bhasa plays and the heavily reworked Sattasai are counted. That contestable arithmetic keeps the test live rather than leaked.

• P. V. Kane (ed.), The Harshacarita of Banabhatta (Bombay, 1918), notes on the introductory verses of Ucchvasa 1
• A. B. Keith, A History of Sanskrit Literature (Oxford, 1928)

Predictions

No prediction registered yet.