The first poet arrived four centuries late

Claim (verbatim)

Telugu literature begins with Nannaya's Mahābhārata (mid-eleventh century) — begins, that is, in the manuscript record. In stone, written Telugu is over four centuries older: the language appears in inscriptions from the sixth century (the Renāṭi Cōḷa records, the Kalamalla inscription of about 575 among the earliest), and by the ninth century the stones are carrying formal verse — the Addanki record of Pāṇḍaraṅga with its taruvoja stanza, the Bezawada inscription of Yuddhamalla with its madhyākkara verses — in native Telugu metres that then surface, fully disciplined, in Nannaya himself. Nobody engraves a metre that nobody composes on leaf: courtly verse cut in stone is the visible edge of a cultivated palm-leaf poetry, and that literature — four-plus centuries of it, including whatever the later tradition half-remembers of pre-Nannaya Jain letters — survives in zero manuscripts. Nannaya's title of ādikavi, "first poet," is a survivorship label, not a birth certificate, and the inscriptional corpus is the instrument that shows it. Prediction: a census of published pre-1050 Telugu inscriptions will yield at least ten records carrying metrical Telugu verse, in at least three distinct native metres, every one predating any extant Telugu literary manuscript's text, and the inscriptional metres will reappear in the earliest manuscript-transmitted poetry (primary clause: the ten-records-in-three-metres floor; the verdict follows it). Kill (not yet built, from real sources): a verse-bearing-inscription census assembled from South Indian Inscriptions and Epigraphia Indica (where the Addanki and Bezawada records are published), read with Korada Mahadeva Sastri's Historical Grammar of Telugu (1969) for the metrical identifications.

Prediction clause (verbatim)

Prediction: a census of published pre-1050 Telugu inscriptions will yield at least ten records carrying metrical Telugu verse, in at least three distinct native metres, every one predating any extant Telugu literary manuscript's text, and the inscriptional metres will reappear in the earliest manuscript-transmitted poetry (primary clause: the ten-records-in-three-metres floor; the verdict follows it).

Kill-dataset (verbatim)

Kill (not yet built, from real sources): a verse-bearing-inscription census assembled from South Indian Inscriptions and Epigraphia Indica (where the Addanki and Bezawada records are published), read with Korada Mahadeva Sastri's Historical Grammar of Telugu (1969) for the metrical identifications.

Novelty / leakage triage

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

Telugu literary history states the phenomenon outright - verse in native metres (Addanki's taruvoja, Bezawada's madhyakkara, Kandukuru and Dharmavaram sisa) predates Nannaya in the inscriptional record, and the adikavi title is recognized as a survivorship label; but the pinned census (at least ten verse-bearing records in at least three metres across the whole published pre-1050 corpus) was not located as compiled, so the ten-record floor is genuinely open arithmetic.

• V. Narayana Rao & D. Shulman, Classical Telugu Poetry: An Anthology (Berkeley, 2002), introduction
• K. Mahadeva Sastri, Historical Grammar of Telugu (1969)

Predictions

No prediction registered yet.