Beyond Linguistic Decipherment: A Structural Analysis of the Voynich Manuscript as a Diagrammatic Information System (DIS)

The Voynich Manuscript (Beinecke MS 408) has resisted linguistic decipherment for over a century. This preprint proposes a structural reframing of the problem: rather than encoding a hidden language, the manuscript is analyzed as a Diagrammatic Information System (DIS), in which textual elements function as modular, position-bound parameters linked to visual structures rather than as components of linear prose. A full-manuscript quantitative analysis across botanical, balneological, cosmological, and pharmaceutical folios demonstrates that key statistical properties—such as low lexical continuity, repetition patterns, and positional constraints—are stable throughout the corpus. Quantitative results reveal a systematic suppression of lexical continuity between adjacent segments (J = 0. 0807), a structural signature that is statistically incompatible with medieval technical prose (Cohen’s d = 2. 45). These features instead support the interpretation of MS 408 as a non-linear technical artifact. The study introduces a falsifiable analytical framework for future computational and codicological research, providing empirical constraints for all future models of Voynich text generation without assuming semantic decipherment or language identification. Changes in Version 2. 0: Methodological Fortification: Explicitly defined the study's scope as a global structural analysis, establishing clear boundaries between global modularity and local generative mechanisms. Statistical Robustness: Introduced Cohen’s d effect size analysis to formally quantify the divergence between MS 408 and medieval Latin control corpora. Refined Comparative Baseline: Integrated a detailed comparison with medieval technical texts (De Materia Medica, Herbarius) to demonstrate that the observed discontinuity is unique to the Voynich Manuscript. Terminology Update: Formalized the "Diagrammatic Information System" (DIS) nomenclature to better reflect the manuscript’s non-linear information architecture.