The moon is the only table

Claim (verbatim)

Rongorongo survives as roughly twenty-six inscribed wooden objects, and the only passage whose internal structure commands any consensus is the lunar sequence on side a of the Mamari tablet (Text C, lines Ca6–Ca9): Thomas Barthel's foundational corpus (Grundlagen zur Entzifferung der Osterinselschrift, 1958) flagged it as calendrical, and Jacques Guy (Journal de la Société des Océanistes, 1990) analysed its runs of crescent signs against the old Rapa Nui night-lists recorded by Thomson (1891) and Métraux (1940), reading it as a lunar month with an intercalation protocol. It is no accident that the one readable passage is the one that could be read without the readers: its structure is visible as structure, while everything else needed the trained rongorongo men, who did not survive the 1862–63 Peruvian slave raids and their aftermath. The claim is that the calendar is not the visible tip of a calendrical sub-genre but the corpus's only table: the remaining texts behave as one linear recitational genre — consistent with the dense web of parallel passages across tablets that Pozdniakov's alignments established — and the tradition committed exactly one instrument to wood, the one artifact that beats memory because it is consulted night by night rather than performed. Prediction: in the digitized Barthel-coded corpus, no passage outside Mamari lines Ca6–Ca9 contains a run of eight or more consecutive lunar-crescent signs, and the Mamari calendar remains the longest homogeneous run of any crescent-family sign corpus-wide (primary clause: zero non-Mamari runs of eight or more consecutive crescents; the verdict follows it). Exact computation: tabulate maximal same-sign runs of the crescent sign over every text in the digitized corpus and rank them. Kill: Barthel's 1958 transcriptions as digitized by the CEIPP and in Konstantin Pozdniakov's published corpus, with Guy's segmentation of the Mamari calendar (Journal de la Société des Océanistes, 1990) fixing the reference passage.

Prediction clause (verbatim)

Prediction: in the digitized Barthel-coded corpus, no passage outside Mamari lines Ca6–Ca9 contains a run of eight or more consecutive lunar-crescent signs, and the Mamari calendar remains the longest homogeneous run of any crescent-family sign corpus-wide (primary clause: zero non-Mamari runs of eight or more consecutive crescents; the verdict follows it). Exact computation: tabulate maximal same-sign runs of the crescent sign over every text in the digitized corpus and rank them.

Kill-dataset (verbatim)

Kill: Barthel's 1958 transcriptions as digitized by the CEIPP and in Konstantin Pozdniakov's published corpus, with Guy's segmentation of the Mamari calendar (Journal de la Société des Océanistes, 1990) fixing the reference passage.

Novelty / leakage triage

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

The consensus that Ca6-9 is the only understood rongorongo passage is published, and Horley's 2011 survey already asked the corpus-wide question — examining crescent groupings on Keiti and other tablets and concluding that Mamari holds 'the only complete lunar calendar in the survived rongorongo corpus'. The specific maximal-run statistic (no non-Mamari run of eight-plus crescents, Mamari the longest homogeneous run) has not been tabulated, so the exact test remains unrun.

• J.B.M. Guy, 'On the Lunar Calendar of Tablet Mamari', Journal de la Societe des Oceanistes 91 (1990), 135-149
• P. Horley, 'Lunar calendar in rongorongo texts and rock art of Easter Island', Journal de la Societe des Oceanistes 132 (2011)

Predictions

No prediction registered yet.