The enemy kept the better book

Claim (verbatim)

Kumārila Bhaṭṭa, the seventh-century champion of Mīmāṃsā against the Buddhists, left two great verse recastings of the system's foundations: the Ślokavārttika, which his school adopted as its classroom text and wrapped in a commentarial ladder (Umbeka, Sucaritamiśra, Pārthasārathimiśra), and the vaster Bṛhaṭṭīkā — which the school let die. It survives nowhere in its own tradition; what remains of it was salvaged by the enemy. Erich Frauwallner showed (1962) that Śāntarakṣita's Tattvasaṃgraha, the eighth-century Buddhist summa, quotes the Bṛhaṭṭīkā at length in order to refute it — long runs of Kumārila's verses that appear in no Mīmāṃsā source, preserved because a polemicist needs his opponent verbatim while a school only recopies what it teaches. The asymmetry is a law of śāstra economics: curriculum texts ride the classroom treadmill, polemical siege-engines are abandoned where they stood once the battle moves. Both the summa and the classroom text are now online, so the lost book's enemy-held remains are countable. Prediction: matching every verse the Tattvasaṃgraha (with Kamalaśīla's Pañjikā) presents as Kumārila's position against the complete Ślokavārttika, at least fifty quoted verses will be absent from the Ślokavārttika — Bṛhaṭṭīkā material with no Mīmāṃsā-side transmission — while the Mīmāṃsā school's own commentarial chain yields not one verse of the lost work (primary clause: the fifty-verse floor of enemy-only survival; the verdict follows it). Exact computation: string-match the e-texts of the two works and check the residual verses against Frauwallner's fragment list. Kill: the GRETIL e-texts of Śāntarakṣita's Tattvasaṃgraha with Kamalaśīla's Pañjikā (Gaekwad's Oriental Series edition) and of Kumārila's Ślokavārttika, controlled against Erich Frauwallner, "Kumārila's Bṛhaṭṭīkā" (Wiener Zeitschrift für die Kunde Süd- und Ostasiens 6, 1962) and Kei Kataoka's Bṛhaṭṭīkā fragment studies.

Prediction clause (verbatim)

Prediction: matching every verse the Tattvasaṃgraha (with Kamalaśīla's Pañjikā) presents as Kumārila's position against the complete Ślokavārttika, at least fifty quoted verses will be absent from the Ślokavārttika — Bṛhaṭṭīkā material with no Mīmāṃsā-side transmission — while the Mīmāṃsā school's own commentarial chain yields not one verse of the lost work (primary clause: the fifty-verse floor of enemy-only survival; the verdict follows it). Exact computation: string-match the e-texts of the two works and check the residual verses against Frauwallner's fragment list.

Kill-dataset (verbatim)

Kill: the GRETIL e-texts of Śāntarakṣita's Tattvasaṃgraha with Kamalaśīla's Pañjikā (Gaekwad's Oriental Series edition) and of Kumārila's Ślokavārttika, controlled against Erich Frauwallner, "Kumārila's Bṛhaṭṭīkā" (Wiener Zeitschrift für die Kunde Süd- und Ostasiens 6, 1962) and Kei Kataoka's Bṛhaṭṭīkā fragment studies.

Novelty / leakage triage

already answered in the literature

Frauwallner 1962 established exactly this result: the Tattvasamgraha's Kumarila quotations absent from the Slokavarttika are Brhattika material surviving only in the opponent's summa, and his fragment identifications (extended by Kataoka's Brhattika studies) already cover far more than fifty verses with no Mimamsa-side transmission. The conjecture's own named control is the leak.

• E. Frauwallner, 'Kumarila's Brhattika', Wiener Zeitschrift fur die Kunde Sud- und Ostasiens 6 (1962)
• K. Kataoka, Kumarila on Truth, Omniscience, and Killing (Vienna, 2011), on the Brhattika fragments in the Tattvasamgraha

Predictions

No prediction registered yet.