The graves outlasted the stories

Claim (verbatim)

Among the oldest Welsh poetry, in the Black Book of Carmarthen (Aberystwyth, NLW, Peniarth MS 1, c. 1250), stands the Englynion y Beddau, the Stanzas of the Graves - a sequence of englynion each naming a hero and locating his grave in the landscape ('the grave of so-and-so is at such a place'), edited by Thomas Jones ('The Black Book of Carmarthen Stanzas of the Graves', Proceedings of the British Academy, 1967). The stanzas name on the order of seventy warriors, and the poem's whole rhetorical force depends on the hearer recognising each name - the grave is a mnemonic hook for a heroic tale the audience is presumed to carry. Almost all of those tales are gone. The loss mechanism is again the divide between an epitaph and an epic: a single grave-stanza is short, formulaic, and survives; the saga that made the buried man worth an englyn was long, oral, and did not. So the Stanzas of the Graves are a cemetery in the double sense - a list of burials that is also the burial-list of a lost narrative literature. Restated: of the distinct heroes named in Englynion y Beddau, fewer than one in four appear in any surviving Welsh narrative or in the Triads; the rest are names on graves whose stories were never dug up. The landscape kept its dead; the parchment lost their deeds.

Prediction clause (verbatim)

Prediction: of the distinct heroes named in the Englynion y Beddau (Thomas Jones's edition, 1967), fewer than 25 percent will appear in any surviving Welsh narrative or in Bromwich's Trioedd Ynys Prydein (primary clause: a cross-attestation rate below 0.25; the verdict follows it). A hero counts as attested only on an independent narrative or triadic occurrence identifying the same figure, not on a shared common name or the grave-stanza itself; ambiguous single-element names are retained in the denominator and flagged. The test voids for coverage if fewer than fifty distinct hero-names can be extracted from the stanzas.

Kill-dataset (verbatim)

Kill: the Englynion y Beddau as edited by Thomas Jones (Proceedings of the British Academy, 1967), each distinct hero-name checked for occurrence in the extant Welsh narrative corpus (Rhyddiaith Gymraeg) and in Bromwich's Trioedd Ynys Prydein index.

Novelty / leakage triage

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

Thomas Jones's edition is the printed census - his notes identify each grave's hero where identification is possible, and his lecture frames the poem as an index to a mostly lost heroic literature - and the standard surveys restate that reading. But the specific cross-attestation arithmetic (distinct hero-names against the narrative corpus plus the Triads, under a 0.25 threshold with ambiguous names retained in the denominator) has not been run, and it is not safely decided: alongside the wholly obscure dead the stanzas name well-connected figures (Pryderi, Owain ab Urien, Bedwyr, March, the famous Arthur stanza), and the Triads' index catches many heroes whose tales are lost, so the rate could land either side of one in four. Un-run counting over a published apparatus: adjacent.

• T. Jones, 'The Black Book of Carmarthen "Stanzas of the Graves"', Proceedings of the British Academy 53 (1967)
• R. Bromwich (ed.), Trioedd Ynys Prydein, 3rd ed. (Cardiff, 2006), index of personal names
• P. Sims-Williams, 'The Early Welsh Arthurian Poems', in R. Bromwich, A.O.H. Jarman and B.F. Roberts (eds.), The Arthur of the Welsh (Cardiff, 1991)

Predictions

No prediction registered yet.