Five shelves, three of them ash

Claim (verbatim)

Motolinía's prologue letter enumerates, from direct observation in the 1530s and 1540s, the five kinds of books the Nahuas kept: of the years and times; of the days and feasts; of dreams and auguries; of baptism and the names given to children; and of the rites and ceremonies of marriage. An eyewitness shelf-list is a sampling frame nobody planted for us, and running the surviving corpus through it exposes a clean genre-level extinction. Genre one, the year-count annals, survives robustly in colonial copies — the Boturini, Aubin, and Mexicanus tradition and dozens of alphabetized annals. Genre two, the day-count and festival books, survives in the Borbonicus, the Tonalamatl Aubin, and the divinatory screenfolds. Genres three, four, and five — the dream-books, the name-giving books, the marriage-rite books, the working papers of the day-keeper's client practice — survive in zero exemplars of any date; their content reaches us only as prose digests inside the missionary encyclopedias, Sahagún's Books 4 and 5 and Durán's calendar treatise. The mechanism is a double filter pointing one way: the friars copied and studied the genres they needed to police time and history, while the client-facing divinatory genres were both maximally incriminating to own and physically ephemeral — small documents consumed in use, held by the practitioners most exposed to prosecution. Three ordinary genres at zero, on an eyewitness taxonomy of five, is the sharpest genre-resolved measurement of loss the hemisphere offers. Prediction: classifying every entry of the Glass-Robertson Census of Native Middle American Pictorial Manuscripts against Motolinía's five genres, the annals genre and the day-count genre will each count ten or more entries, while the dream-book, name-giving, and marriage-rite genres will count zero surviving exemplars each, at any date (primary clause: the zero-of-three genre census; a single surviving exemplar of any of the three kills it). Kill: John B. Glass with Donald Robertson, A Census of Native Middle American Pictorial Manuscripts (Handbook of Middle American Indians, vol. 14, 1975), entry classifications, with Motolinía's prologue letter (Memoriales) and the digitized Florentine Codex, Books 4-5, as the digest control.

Prediction clause (verbatim)

Prediction: classifying every entry of the Glass-Robertson Census of Native Middle American Pictorial Manuscripts against Motolinía's five genres, the annals genre and the day-count genre will each count ten or more entries, while the dream-book, name-giving, and marriage-rite genres will count zero surviving exemplars each, at any date (primary clause: the zero-of-three genre census; a single surviving exemplar of any of the three kills it).

Kill-dataset (verbatim)

Kill: John B. Glass with Donald Robertson, A Census of Native Middle American Pictorial Manuscripts (Handbook of Middle American Indians, vol. 14, 1975), entry classifications, with Motolinía's prologue letter (Memoriales) and the digitized Florentine Codex, Books 4-5, as the digest control.

Novelty / leakage triage

already answered in the literature

The stated primary clause is the zero-of-three genre census, and that answer is already secure in the published record: no standalone pre-Hispanic or colonial pictorial 'dream-book,' 'name-giving book,' or 'marriage-rite book' survives as a manuscript, their content reaching us only as prose digests in Sahagun (Florentine Codex, Books 4-5) and Duran, while the year-count annals and the day-count tonalamatl survive in numbers. Boone's synthesis of the surviving divinatory and historical corpora against the Glass-Robertson census makes the genre-level survival map explicit, so a competent reader already knows the three client-divinatory genres stand at zero exemplars of any date. Only the exact per-genre entry tallies (the ten-or-more counts) are un-run arithmetic; the verdict-bearing zero-of-three is guaranteed by the known corpus.

Predictions

No prediction registered yet.