Six schoolfellows, one survivor and a half

Claim (verbatim)

The Carakasaṃhitā's frame narrative counts what Indian medicine's first generation wrote: Ātreya Punarvasu taught six pupils — Agniveśa, Bhela, Jatūkarṇa, Parāśara, Hārīta, Kṣārapāṇi — and each composed his own saṃhitā. Six parallel foundations. The ledger fifteen centuries on: Agniveśa's text lives, but only inside its redaction chain, recast by Caraka and patched by Dṛḍhabala; Bhela's survives as a single damaged palm-leaf codex in the Thanjavur palace library, a late southern copy first printed by the University of Calcutta in 1921; the other four are quotation-ghosts, cited by name in the medieval commentators and compilers, with no manuscript anywhere. And the ghost-names attracted squatters: the Hārītasaṃhitā in print is a much later text wearing the old pupil's name. Curriculum text, single-codex straggler, four ghosts, one impostor — that gradient is the survival physics of parallel foundations under winner-take-all pedagogy, and every step of it is checkable in the testimonia. Prediction: in the testimonia registers of Meulenbeld's history, at least three of the four ghost-pupils (Jatūkarṇa, Parāśara, Hārīta, Kṣārapāṇi) will each show ten or more attributed quotations in premodern medical literature against zero extant manuscript witnesses of the old works, and the extant Hārītasaṃhitā will show no substantial overlap with the old Hārīta quotations (primary clause: the quotations-without-manuscripts conjunction for at least three of the four; the verdict follows it). Kill: G. Jan Meulenbeld, A History of Indian Medical Literature (Groningen, 1999-2002), testimonia and fragment registers, with the New Catalogus Catalogorum for the manuscript null and the Bhela Saṃhitā editio princeps (University of Calcutta, 1921) from the unique Thanjavur codex.

Prediction clause (verbatim)

Prediction: in the testimonia registers of Meulenbeld's history, at least three of the four ghost-pupils (Jatūkarṇa, Parāśara, Hārīta, Kṣārapāṇi) will each show ten or more attributed quotations in premodern medical literature against zero extant manuscript witnesses of the old works, and the extant Hārītasaṃhitā will show no substantial overlap with the old Hārīta quotations (primary clause: the quotations-without-manuscripts conjunction for at least three of the four; the verdict follows it).

Kill-dataset (verbatim)

Kill: G. Jan Meulenbeld, A History of Indian Medical Literature (Groningen, 1999-2002), testimonia and fragment registers, with the New Catalogus Catalogorum for the manuscript null and the Bhela Saṃhitā editio princeps (University of Calcutta, 1921) from the unique Thanjavur codex.

Novelty / leakage triage

already answered in the literature

Meulenbeld's registers state the exact finding: Jatukarna, Parasara, Harita and Ksarapani survive only as named quotations in the medieval medical literature (Jatukarna profusely cited into Cakrapani's era, Ksarapani quoted into Vagbhata's) with zero manuscripts of the old works, and the extant Haritasamhita is a late text distinct from the 'old Harita' of the quotations. The kill dataset is the publication that already contains the verdict; only the ten-quotation threshold arithmetic is left.

Predictions

No prediction registered yet.