Four literatures hang from one wounded codex

Claim (verbatim)

The oldest stratum of the eastern Indo-Aryan vernaculars — the caryā songs of the tantric siddhas, claimed as the fountainhead by Bengali, Assamese, Odia, and Maithili literary history alike — survives in exactly one manuscript: the palm-leaf codex of the fifty-song collection with Munidatta's Sanskrit commentary that Haraprasād Śāstrī found in the Nepal Durbar Library in 1907. The codex is wounded — missing leaves have taken several songs outright — and its wounds are patched from abroad: the Tibetan Tanjur carries a translation of the collection with its commentary, from which the lost songs' content is read back, both sides printed together in Kvaerne's edition. One damaged export copy plus a foreign shadow: that is the entire home survival of four literatures' first centuries. And the Tanjur measures how much more there was, for it preserves a whole shelf of caryā and dohā works under the siddhas' names, translated from Indic originals that no longer exist anywhere. Prediction: counting the caryā- and dohā-genre works attributed to the siddhas in the Tanjur, under 15% will have any surviving witness in an Indic language, and the survivors will be precisely the pieces with continuing Nepalese transmission — Saraha's and Kāṇha's dohākoṣas, the caryā collection itself — with none surviving through mainland Indian collections (primary clause: the under-15% Indic survival share; the verdict follows it). Exact computation: harvest the genre entries from the Tanjur catalogues and resolve each against the published Indic-side editions and the Nepalese microfilm record. Kill: the Tōhoku Complete Catalogue of the Tibetan Buddhist Canons (Sendai, 1934) with the Vienna rKTs Kanjur-Tanjur database (rkts.org) for the count, Per Kvaerne's An Anthology of Buddhist Tantric Songs (Oslo, 1977), M. Shahidullah's Les chants mystiques de Kāṇha et de Saraha (Paris, 1928), and the NGMCP online catalogue for the Nepal side.

Prediction clause (verbatim)

Prediction: counting the caryā- and dohā-genre works attributed to the siddhas in the Tanjur, under 15% will have any surviving witness in an Indic language, and the survivors will be precisely the pieces with continuing Nepalese transmission — Saraha's and Kāṇha's dohākoṣas, the caryā collection itself — with none surviving through mainland Indian collections (primary clause: the under-15% Indic survival share; the verdict follows it). Exact computation: harvest the genre entries from the Tanjur catalogues and resolve each against the published Indic-side editions and the Nepalese microfilm record.

Kill-dataset (verbatim)

Kill: the Tōhoku Complete Catalogue of the Tibetan Buddhist Canons (Sendai, 1934) with the Vienna rKTs Kanjur-Tanjur database (rkts.org) for the count, Per Kvaerne's An Anthology of Buddhist Tantric Songs (Oslo, 1977), M. Shahidullah's Les chants mystiques de Kāṇha et de Saraha (Paris, 1928), and the NGMCP online catalogue for the Nepal side.

Novelty / leakage triage

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

The sole-codex survival of the carya collection and the Tanjur's role in restoring its lost songs are textbook (Kvaerne prints both sides together), and it is a commonplace that most siddha doha and carya works survive only in Tibetan translation; but the genre-wide Tanjur census with an Indic-survival share and the Nepalese-channel-exclusivity clause was not located as ever having been computed.

• P. Kvaerne, An Anthology of Buddhist Tantric Songs (Oslo, 1977)
• K.R. Schaeffer, Dreaming the Great Brahmin: Tibetan Traditions of the Buddhist Poet-Saint Saraha (Oxford, 2005)

Predictions

No prediction registered yet.