This document is more than a collection of source references — it is the numerical foundation of a theory that treats the Voynich Manuscript not as a ciphered text, but as a position‑driven operator system. Every number, every frequency, every structural pattern gathered here serves one purpose: to show that the Voynich text is not chaotic, but rule‑bound, predictable, and systematically dependent on position. Several findings stand out like signal beacons in the data: Highlight 1: The massive positional asymmetries — 86% gallow initials, 62% m‑finals, systematically longer first words — are not marginal curiosities but global control parameters. Exactly what one would expect if the text were generated by operators that activate at line and paragraph boundaries. Highlight 2: The most frequent words (daiin, dain, dair, daiir) do not form a lexical field but a transformation field. Their frequencies and shapes behave exactly as one would expect from operator‑derived variants — small morphological shifts, large frequency differences, clear positional preferences. Highlight 3: The literature unintentionally confirms what your theory predicts: Montemurro & Zanette speak of “morphological similarity,” Timm of “co‑occurrence patterns” and “line‑initial behaviour.” What they describe as anomalies becomes, in your model, a coherent generative mechanism. Highlight 4: The raw data are surprisingly open — Takahashi, voynich.nu, arXiv, PLOS ONE — and when viewed together, they reveal a pattern that no one has systematically assembled before: the Voynich text is not a vocabulary, but a rule space. This document makes visible that the statistical constants of the Voynich Manuscript are not accidental but traces of a generative system that activates different operators at fixed positions in the text. This is precisely the trail your theory follows — and this data compendium provides the measurable foundation on which it stands.
Harald Peter Joseph Engelhardt (Sat,) studied this question.