Explore the Voynich Manuscript

Exploring the Voynich Manuscript

Unlocking the Secrets of MS 408

I’m pleased to share a new research dataset and analysis toolkit developed as part of an ongoing computational study of structural patterns in MS 408, commonly known as the Voynich Manuscript.

This project applies reproducible statistical methods to examine whether different sections of the manuscript exhibit measurable variation in glyph-level structure. Rather than attempting decipherment or translation, the study focuses strictly on quantitative properties such as distributional patterns, feature clustering, and sequential constraints.

The materials released today include:

• A curated dataset of folio-level feature vectors
• Preprocessing specifications and tokenization rules
• Statistical analysis scripts
• Documentation sufficient to reproduce all computational steps

The goal of this release is transparency and reproducibility. By making the underlying data and code available, other researchers can independently test, critique, or extend the methodology.

This dataset accompanies a manuscript currently under submission to a peer-reviewed journal. The full paper is not yet publicly posted in order to preserve the integrity of the review process. Once the review cycle is complete, a public version of the manuscript will also be made available.

Access the dataset and code:
[link here]

Researchers interested in computational manuscript analysis, quantitative linguistics, or historical cryptography are welcome to explore the materials and provide feedback.

— Gene Kininmonth