The corpus is an evacuation list

Claim (verbatim)

In January 1864 Joseph-Eugène Eyraud, the first missionary to live on Rapa Nui, reported wooden tablets and staves covered in hieroglyphic characters ‘in all the houses’; he had arrived a year after the Peruvian labor raids of 1862–63 carried off some 1,400–1,500 islanders, the chiefly and priestly families among them, and the epidemics that came home with the few repatriated took the island to 111 people by 1877. The surviving corpus of rongorongo is roughly twenty-five inscribed objects, and their collection histories are not a sample of a stable population — they are the departure log of a perishing one. The Échancrée reached Bishop Jaussen in Tahiti in 1869 as the spool for a gift coil of hair, the event that told Europe the script existed; his standing request thereafter netted him five tablets in all; officers of the Chilean corvette O'Higgins obtained the three Santiago pieces in 1870; Miklouho-Maclay carried the St Petersburg pair away in 1871; and Paymaster Thomson of the USS Mohican bought the last tablets ever collected on the island in 1886, over initial refusals. If Eyraud's ‘all the houses’ means even one inscribed object per household of the post-raid population, the 1864 stock stood in the hundreds; the corpus is under a tenth of that, and the window in which collecting was possible at all was one generation wide — the survivorship curve of a literature, with dates on it. Prediction: tabulating collection histories across the full corpus from Fischer's per-object catalogue, every object whose departure from Rapa Nui is documented will prove to have left the island within the single generation 1868–1888, and none of the three major twentieth-century expeditions that searched — Routledge in 1914–15, Métraux and Lavachery in 1934–35, Heyerdahl in 1955–56 — will have recovered one authenticated new inscription on the island (primary clause: the 1868–1888 window covering 100% of documented departures; the verdict follows it). Kill: the object-by-object provenance histories in Steven Roger Fischer, Rongorongo: The Easter Island Script (Oxford, 1997), read against Routledge, The Mystery of Easter Island (1919), Métraux, Ethnology of Easter Island (Bishop Museum Bulletin 160, 1940), and the Norwegian Archaeological Expedition reports.

Prediction clause (verbatim)

Prediction: tabulating collection histories across the full corpus from Fischer's per-object catalogue, every object whose departure from Rapa Nui is documented will prove to have left the island within the single generation 1868–1888, and none of the three major twentieth-century expeditions that searched — Routledge in 1914–15, Métraux and Lavachery in 1934–35, Heyerdahl in 1955–56 — will have recovered one authenticated new inscription on the island (primary clause: the 1868–1888 window covering 100% of documented departures; the verdict follows it).

Kill-dataset (verbatim)

Kill: the object-by-object provenance histories in Steven Roger Fischer, Rongorongo: The Easter Island Script (Oxford, 1997), read against Routledge, The Mystery of Easter Island (1919), Métraux, Ethnology of Easter Island (Bishop Museum Bulletin 160, 1940), and the Norwegian Archaeological Expedition reports.

Novelty / leakage triage

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

Fischer's Rongorongo (1997) gives the object-by-object provenance for the whole corpus, and the closed-corpus fact — everything authentic collected in the second half of the nineteenth century, nothing new recovered since — is foundational: the Echancree-to-Jaussen (1869), O'Higgins (1870), Miklouho-Maclay (1871) and Thomson (1886) departures are all in print, and Routledge, Metraux and Heyerdahl each reported finding no new authenticated inscription. That second, no-new-finds half of the prediction is therefore essentially leaked. But the primary clause — that 100% of documented departures fall inside a single 1868–1888 generation, framed as a survivorship curve — has not been tabulated as such from Fischer's catalogue; the materials guarantee it while the statistic itself is un-run, exactly as Horley's corpus-wide lunar survey left the maximal-run count un-tabulated in wave 1. Adjacent.

• Steven Roger Fischer, Rongorongo: The Easter Island Script — History, Traditions, Texts (Oxford University Press, 1997), the per-object provenance catalogue
• Alfred Metraux, Ethnology of Easter Island (Bishop Museum Bulletin 160, 1940)
• Katherine Routledge, The Mystery of Easter Island (London, 1919)

Predictions

No prediction registered yet.