Mysteries Department
Sign inRequest access
Run · Voynich Manuscript

T38 grid: word-to-word glyph dependency and near-twins, every variant, as written and without the first glyph

Shuffling the manuscript's own words removes most of its edge dependency, so it lives in word order (how one word ends predicts how the next begins), not in the vocabulary, and removing first glyphs leaves most of it. In Naibbe each word is an independent cipher unit, so nothing links the end of one to the start of the next. Near-twins, one-glyph variants close together, are partly vocabulary and partly local.

Not supported: no variant in any section comes within the spread on either statisticedge MI: manuscript 0.19-0.32 by section, the 72 variants 0.01-0.10; on bodies 0.11-0.27 against 0.01-0.12. twin10: manuscript 0.24-0.33, variants 0.13-0.190

Result

RunMeasuredResultAgainstDecisionDate
RunMI between a word's last glyph and the next word's first (edge MI, bits); share of words with a one-glyph variant within the next 10 words (twin10); edge MI on word bodiesedge MI: manuscript 0.19-0.32 by section, the 72 variants 0.01-0.10; on bodies 0.11-0.27 against 0.01-0.12. twin10: manuscript 0.24-0.33, variants 0.13-0.19The section's own words shuffled across its pages: edge MI 0.02-0.08, twin10 0.18-0.27Not supported: no variant in any section comes within the spread on either statistic

What it used and found

Parameters

corpora = 10, sections = 6, variants = 72

Cite it

Mysteries Department. (2026). T38 grid: word-to-word glyph dependency and near-twins, every variant, as written and without the first glyph [Run RUN0DC8YFNJRFFC]. https://mysteriesdept.com/r/RUN0DC8YFNJRFFC
Private beta Downloads are for beta members

We're in a private beta while we get things right. Everything here is free to read; downloading datasets, sources and code is open to members for now.

It takes a couple of minutes. We'll reply by email.