Run · Voynich Manuscript

T41 generation 1: lattice EM, planted control and structure-preserving null (pw 0.25, alpha 0.05)

Control plaintext Pliny NH 16 enciphered by period.generate seed 77 (the same control as M-T40; its true-plaintext coverage 0.697 reproduces T40's). Nulls: pseudo-Latin from a letter-trigram chain of the reference Latin, chain seeds 501/502, generator seeds 78/79. Restart seeds 1000-1029.

Fails the control gate (0.324 < 0.488), so the manuscript was not decoded. Key accuracy with learned boundaries rises from T40's 11-23 % to 36 %, but the decoded control is Latin-like letter soup, no better on coverage than a permuted key. The pseudo-Latin null decodes to the same coverage as the control: with a bigram element model, coverage cannot tell a real key from structure.control 0.324 / 0.364 / 0.083 (bar 0.488; true plaintext 0.697); permuted-key 0.282 (p 0.19). Pseudo-Latin null: 0.309 and 0.351 decoded coverage (their own plaintexts 0.359, 0.337)0

Result

Run	Measured	Result	Against	Decision	Date
Run	Held-out dictionary coverage (bar 0.7 x true plaintext's) / key accuracy by use / character similarity	control 0.324 / 0.364 / 0.083 (bar 0.488; true plaintext 0.697); permuted-key 0.282 (p 0.19). Pseudo-Latin null: 0.309 and 0.351 decoded coverage (their own plaintexts 0.359, 0.337)	M-T40 rung B (learned boundaries, element solver): key 0.114-0.228, coverage 0.232-0.271 · rung A (true boundaries): key 0.706, coverage 0.336	Fails the control gate (0.324 < 0.488), so the manuscript was not decoded. Key accuracy with learned boundaries rises from T40's 11-23 % to 36 %, but the decoded control is Latin-like letter soup, no better on coverage than a permuted key. The pseudo-Latin null decodes to the same coverage as the control: with a bigram element model, coverage cannot tell a real key from structure.

What it used and found

Lattice EM decipherment: sign boundaries and key learned together, with a structure-preserving null

Method

MTH

Herbal A and Herbal B paragraph text: training and held-out pages

Dataset

DAT

Latin genre texts

Dataset

DAT

Naibbe cipher tables and example plaintexts (greshko/naibbe-cipher, commit 719f2f4c)

Dataset

DAT

ZL3b transcription

Dataset

DAT

Latin as a re-spaced homophonic stream (period practices)

Answer it tests

ANS

Parameters

pw = 0.25, alpha = 0.05, signs = chunks of 1-3 glyph units seen 3+ times on training pages, plus every single glyph (1,429), restarts = 30, settings = {'K': 50, 'h_c': 4, 'h_v': 1, 'lam': 0.5, 'Ssyl': 0, 'abbrev': 0, 'p_copy': 0.3, 'p_stay': 0.9, 'window': 5, 'novowel': 0, 'p_break': 0.5, 'spacing': 'stream', 'simplify': 1, 'positional': 0}, permutations = 20, em_iterations = 60, restart_choice = best training likelihood

Cite it

Mysteries Department. (2026). T41 generation 1: lattice EM, planted control and structure-preserving null (pw 0.25, alpha 0.05) [Run RUN0DCF745R93NZ]. https://mysteriesdept.com/r/RUN0DCF745R93NZ

@misc{mysteriesdept_run0dcf745r93nz,
  title     = {T41 generation 1: lattice EM, planted control and structure-preserving null (pw 0.25, alpha 0.05)},
  author    = {{Mysteries Department}},
  year      = {2026},
  note      = {Run, catalog key RUN0DCF745R93NZ},
  url       = {https://mysteriesdept.com/r/RUN0DCF745R93NZ}
}