AI-generated conjecture · below the evidence/publication boundary
← All conjectures · Philosophy, theology & the schools
The purvapaksa is a fossil
Status is derived only from the shepherd-authored triage/prediction data above -- community submissions and claims are a separate overlay and can never change it (see the participation panel below).
Claim (verbatim)
The purvapaksa is a fossil: Sanskrit philosophical debate preserved its enemies more faithfully than its authors. Once a rival school went extinct, its positions fossilized into stock quotable verses that every refuter reproduced nearly verbatim — fairness conventions required quoting the opponent exactly, and after the opponents' own books vanished, the inherited quotation was the only carrier — while each school's own siddhanta prose was freely rewritten in every generation, because living doctrine is maintained by paraphrase. Textual stability should therefore be higher across school lines (shared opponent material) than within them (own-school doctrine).
Prediction clause (verbatim)
Across digitized doxographies and polemical works — Sarvadarsanasamgraha, Haribhadra's Saddarsanasamuccaya with Gunaratna's commentary, Jayanta's Nyayamanjari, Santaraksita's Tattvasamgraha with Kamalasila's Panjika — align the shared Carvaka/Lokayata citations. Primary clause: mean normalized verbatim overlap of Carvaka material between texts of different schools exceeds the overlap of parallel doctrinal prose between texts of the same school by at least 20 points of alignment score; the verdict follows this clause. Secondary clause: at least 70 percent of the Carvaka material across these texts reduces to a shared stock of fewer than 30 distinct verses and sutra-fragments.
Kill-dataset (verbatim)
Kill: the GRETIL and SARIT open Sanskrit e-text corpora, which contain all the named texts in machine-readable form; standard sequence alignment closes the question.
Provenance
Run: Fresh agent generation · model: claude-fable-5
Generated by a fresh Fable-tier instance (claude-fable-5) at maximum effort with generation-first blindness (no repo reads, no web searches, no DB queries; single packet Write was the only tool use); 188-title list supplied at launch, titles only, no verdicts/dossiers seen; prompt pre-committed in docs/generated/conjectures_1001_wave_ledger.md and docs/generated/conjecture_fresh_fable_w02_prompt_20260708.txt. Wave W02 of docs/GOAL_CONJECTURES_1001_CAMPAIGN_20260708.md. Novelty unverified by construction. DEVIATION DISCLOSURE: after an output-limit resume, one accidental no-op placeholder Write to the session scratchpad preceded this packet Write; nothing was read and no information entered the generation; blindness intact, but the run used two Writes, not one.
Novelty / leakage triage
provisional — model-triaged, shepherd review pending
A provisional first pass authored by the model (Opus), not yet confirmed by the shepherd. It carries the same dated-search requirements as an authoritative verdict but is excluded from every headline figure and cannot underwrite a prediction until a shepherd confirms it. Provisional reading: Adjacent (closely related prior work exists).
That the Carvaka/Lokayata survives only as verbatim stock quotations preserved by its opponents -- fairness conventions obliged exact quotation, and after the school's own books vanished the inherited quotation was the sole carrier -- is Bhattacharya's established finding (the ~30 surviving commentary fragments, seventeen Jain preservers, the explicit Presocratic analogy), so the fossilization premise and even the 'fewer than 30 distinct verses' secondary clause are close to known. Not located: the item's comparative-stability test -- cross-school Carvaka overlap exceeding within-school own-doctrine prose overlap by >=20 alignment points -- which is the decidable operationalization on the GRETIL/SARIT e-texts.
- Ramkrishna Bhattacharya, Studies on the Carvaka/Lokayata (Anthem Press, 2011); 'On the Carvaka/Lokayata' — Carvaka reconstructed from opponents' verbatim quotations; ~30 fragments; a small shared stock
- GRETIL and SARIT open Sanskrit e-text corpora — The kill dataset containing the named doxographies and polemics for sequence alignment
Predictions
No prediction registered yet.
Weigh in
No community feedback yet.
Add your take
Posted immediately (spam is removed). Community feedback is never an adjudicated verdict and never changes this conjecture's triage label or status above.