AI-generated conjecture · a wager, not a finding
← All conjectures · Pre-Columbian American writing
The oldest books have no address
Status is derived only from the shepherd-authored triage/prediction data above -- community submissions and claims are a separate overlay and can never change it (see the participation panel below).
Claim (verbatim)
The five members of the Borgia Group — the Codex Borgia and Vaticanus B in the Vatican, the Cospi in Bologna, the Fejérváry-Mayer in Liverpool, the Laud in Oxford — stand among the dozen-odd books surviving from the entire pre-contact hemisphere, and not one has a recorded exit from Mexico or a secure place of origin: their homeland is still argued among Cholula, the Puebla-Tlaxcala corridor, the Mixteca, and the Gulf. The Cospi reached Bologna in 1665 as a curiosity gift whose label called it a book of China. The mechanism is that a survival channel determines what metadata survives it: a book that crossed the Atlantic as a marvel needed no documentation to do its work — wonder requires no provenance — while a book made inside the colonial legal and evangelical machinery (the Mendoza for the crown, the Huejotzingo for a lawsuit, the Florentine for a mission) carries its origin on its face because its function demanded it. The result is a perverse inversion at the heart of the corpus: the deepest, pre-conquest stratum — exactly the books most needed for mapping where the lost production stood — is systematically the least locatable, while the shallower colonial stratum is well addressed. The hemisphere's oldest surviving books are, in the documentary sense, homeless; and every reconstruction of the lost libraries inherits that hole at its foundation. Prediction: in the Glass-Robertson census, over 70% of items classed as certainly or probably pre-conquest will lack a secure town-level or sub-regional place of origin (unknown or contested provenience), against under 25% for dated sixteenth-century colonial pictorials (primary clause: the 70-versus-25 metadata inversion; the verdict follows it). Kill: the provenience fields of the Glass-Robertson census (Handbook of Middle American Indians, vols. 14-15, 1975), with the Biblioteca Universitaria di Bologna's documentation of the 1665 Cospi gift and the holding institutions' catalogues (Vatican Library, Bodleian, World Museum Liverpool) as the emblem cases.
Prediction clause (verbatim)
Prediction: in the Glass-Robertson census, over 70% of items classed as certainly or probably pre-conquest will lack a secure town-level or sub-regional place of origin (unknown or contested provenience), against under 25% for dated sixteenth-century colonial pictorials (primary clause: the 70-versus-25 metadata inversion; the verdict follows it).
Kill-dataset (verbatim)
Kill: the provenience fields of the Glass-Robertson census (Handbook of Middle American Indians, vols. 14-15, 1975), with the Biblioteca Universitaria di Bologna's documentation of the 1665 Cospi gift and the holding institutions' catalogues (Vatican Library, Bodleian, World Museum Liverpool) as the emblem cases.
Nobody has run this test. The kill-data is named above. If you can run it — or you know the paper that already settles it — claim the kill or submit the prior scholarship. Kills and prior scholarship are credited here, by name, as they come in.
Provenance
Run: Fresh agent generation · model: claude-fable-5
Fresh blind generation by claude-fable-5, 2026-07-17, Americas wave 2 weighted by inferred production rather than survival and deliberately disjoint from the w18 Americas wave and the 2026-07-16 Africa-Americas wave; every item grounded in real named objects, chroniclers, testimonia, catalogues, and datasets with no fabricated citations and honest not-yet-built flags where the decisive dataset does not exist in queryable form; eleven steer candidates dropped — seven for prior coverage in the atlas (Landa genre-bias, khipu context-bias, Mixtec cross-attestation, Nahua song overlap, Andean sole-witness seam, Landa alphabet, codex-implied observation archives) and four for weak kills or scope (Coixtlahuaca lienzos, Midewiwin scrolls, Wari khipu, Walam Olum).
Novelty / leakage triage
anticipated in the literature — this exact test has never been run
That the deepest pre-conquest survivors — the Borgia Group — lack secure places of origin while dated colonial pictorials are well localized is a standard observation, and the contested homeland of the group (Cholula, Puebla-Tlaxcala, the Mixteca, the Gulf) and the 1665 Cospi 'book of China' gift are documented in the provenance literature. But the primary clause is the quantitative inversion (over 70% versus under 25%) read off the provenience fields of the census, and no one has tabulated those two rates; the denominator of 'certainly or probably pre-conquest' items and the classification of Mixtec-codex proveniences (some partly recoverable) make the figure genuinely un-run rather than guaranteed. The pattern is in print; the paired percentages are the study to build.
- John B. Glass with Donald Robertson, 'A Census of Native Middle American Pictorial Manuscripts,' Handbook of Middle American Indians, vols. 14-15 (Austin: University of Texas Press, 1975), provenience entries
- Elizabeth Hill Boone, Cycles of Time and Meaning in the Mexican Books of Fate (Austin: University of Texas Press, 2007), on the contested provenances of the Borgia Group
Predictions
No prediction registered yet.
Weigh in
No community feedback yet.
New here? Create an account first
Create an account or sign in and your feedback is tied to you — you can track it, get replies, and claim this conjecture so others know you’re working on it. Prefer not to? Just leave your take below as a guest — only the name you type is shown.
Add your take
Posted immediately (spam is removed). Community feedback is never an adjudicated verdict and never changes this conjecture's triage label or status above.