FAQ — Anticipated Objections

For the pedants — every sharp objection, answered as precisely as possible

"The verse could post-date the tables. Reverse causality."
Both the verse and the tables live in the same manuscript book, copied roughly 400 years before Reeds's 1998 analysis. The congruence is arithmetic fact regardless of which came first within the book's composition. The direction of dependence within the text is open — and we say so, at confidence 0.92 for the historical-origin reading versus 0.97 for the arithmetic identity itself.

Additionally: the same letter-values are load-bearing in five independent sections of the book (§2 PATER CREATOR = 140; §9 porta paradis = 132, which independently forces d = 29; the four humors in §5). A verse “added later” would have had to be fitted simultaneously to all of these constraints. That is not impossible but is increasingly implausible as the count of constraints grows.

"Kupin's 2014 transcription post-dates Reeds. The edition could be contaminated by Reeds's published f values."
Latin number-words (viginti novem, ter quinque, bis octo) are not graphically ambiguous the way Arabic numerals are. A scribe or editor wishing to introduce "29" into a Latin manuscript cannot do so as a digit — the word must be spelled out. An independent OCR pass of the Internet Archive source concurs with Kupin's readings of every number-word. And the same values are load-bearing in sections of the book far from Section 18 (see above). Fitting one verse is conceivable; fitting the book's whole arithmetic simultaneously is not a realistic contamination scenario.
"Didn't Reeds already know this?"
Reeds noted in passing that the book "assigns numerical values to letters" — and never connected those values to f. This is documented in his 1999 paper. The reason is structural: the full prose of the Book of Soyga was not transcribed until Kupin's 2014 edition. Reeds worked from the grids and from a partial summary of the text. The verse is in Section 18; the tables are roughly 150 pages away in the manuscript. The two halves of the answer sat in one codex for four centuries.

This finding’s debt to Reeds is total and stated everywhere in this work. The connection between Section 18 and f is new. If you know prior art for the Section 18 → f identity, write to jason@ultraculture.org and it will be credited here.

"What is the minus one in f(W) = V(W) − 1?"
An inclusive-versus-exclusive counting difference. The verse states the value V of each letter — a number the author treats as an inclusive ordinal: "the V-th letter, counting from and including the letter above." Reeds's recurrence uses f as an exclusive step count: "advance f steps past the cell above." The gap between "the Nth item counting from position P" and "N steps from P" is exactly 1.

This gloss is ours — it is not a quoted instruction from the manuscript. Under it, the tables regenerate exactly. The single uniform constant is the only one that works (every other offset gives 0/23).

"Is there a hidden message in the Soyga tables?"
No, within the scope of the traversals we enumerated. The test covered standard path types: rows, columns, main diagonals, broken diagonals, boustrophedon (snake-path), and inward spirals — both directions — across all 36 tables, against a 72,789-word dictionary and letter-preserving shuffled controls. Words appear at chance level (0.99× shuffled controls). Entropy sits at the random baseline.

We do not claim this covers every conceivable path through a 36×36 grid; we make no universal negative. What we claim is that the standard traversals yield nothing above chance, which is consistent with what the book itself says: the tables are operated (as a binding grid and scrying mirror), not decoded. See §26.1: “vicem speculi gerit… faciet videre quod libuerit” — it “serves as a mirror and makes the gazer see what they wish.”

"This work was AI-assisted. How is it credible?"
AI-assisted and human-directed, disclosed plainly. All arithmetic was run in code, never asserted from memory. Every quotation was verified at its stated line number in the Kupin text. The work went through seven review passes including:

The corrections are the credibility. A finding that has caught and corrected its own errors is more reliable than one that has not been tested. The verification code is public; every claim is reproducible.

"What do you NOT claim?"
"Are the 36 tables keyed to the 36 decans?"
This is a common misread. The book's "facies" catalogue is the 28 lunar mansions (per Kupin line 12439), not 36 decans. The book never explicitly states a 36-table count in the prose. Its living count of 36 emerges from the §19 thirty-six-star body. The tables are organised as 2×12 zodiac + 7 planets + 4 elements + 1 "Magistri" (Reeds's Table I), which is compositional, not a decan scheme. Describing them as "the 36 decans" imports a framework the text does not use.
"The book was probably Dee's copy — so could Dee have generated the tables himself?"
The manuscripts are probably Dee's copies — the attribution is probable, not certain. Given the whole book, Dee could in principle have read Section 18 (letter values) and Section 4 (the recurrence is described there in verse) and generated the tables. The "perplexity" was that the value-table and the table-generation rule sit roughly 150 pages apart in the manuscript. Whether Dee made that connection himself is an open historical question. What is new is that we can now state the connection precisely.
"The Viseua / UISEUR discrepancy — doesn't that undermine the derivation?"
No, for two reasons. First, the Moon row discrepancy is in the Section 28 ordinal recipe (which is partially corrupt — the book itself says "ista valde intorta sunt"), not in the main f(W) = V(W) − 1 identity. The arithmetic identity holds 23/23 regardless. Second, Reeds independently confirms the table keyword is UISEUA from his full manuscript examination; our generator produces UISEUA from the six-letter seed; the Moon margin + column 1 checks out 24/24. The printed recipe yields UISEUR; the unique one-ordinal repair (final xvi→vi) gives UISEUA. This is a single residual in the partially-recovered selection rule, stated openly and at appropriate confidence (0.45).