A Syriac Pharmaceutical Dispensatory Encoded in the Voynich Manuscript: Statistical and Philological Evidence from Consonant Skeleton Analysis

This deposit contains the Session 32 working draft of an ongoing computational decipherment study of the Voynich Manuscript (Beinecke MS 408), together with the supporting lexicon, methodology documentation, and selected reference materials. The project tests the hypothesis that the manuscript encodes Syriac medical and pharmacological vocabulary using an EVA-to-consonant mapping that omits vowels, consistent with the Syriac abjad tradition. Candidate decodings are matched against a curated Syriac medical lexicon drawn from Payne Smith, Löw, Gignoux/Stol, Müller-Kessler, and related sources. Statistical validation uses permutation testing against randomized baselines, with negative results reported alongside positive findings. Current pipeline state: 87.0% coverage across 35,307 EVA tokens on 202 folios; permutation z-score 3.83 (14.9 percentage points above baseline); 14 statistically validated text-image matches on plant folios (Fisher's exact p = 6.66×10⁻¹⁶); three-way Northwest Semitic discrimination favoring Syriac. Recent findings indicate that EVA vowels function as approximate notation rather than simple substitution, suggesting the consonant-skeleton approach is incomplete rather than incorrect. Included: draft manuscript, curated Syriac medical lexicon, session handoff documentation, recent finding briefs, and selected reference materials (Löw's Aramäische Pflanzennamen; Syriac Galen / Greek transliteration studies). This is an in-progress research artifact, not a completed proof. Author's calibrated confidence: ~40–50% that the underlying language is Syriac; ~10–15% on word-level decode accuracy; ~5% on the full pipeline surviving specialist review unchanged.