For release 24 October 2026

Not a simple cipher: open tests narrow what the Voynich Manuscript could be

Mysteries Department publishes the first release of its open research on the fifteenth-century book no one can read.

Letter-for-letter ciphers of Latin, Italian, German, French and Czech can be ruled out as the source of the Voynich Manuscript's text, according to tests published today by Mysteries Department, an open research platform for unsolved mysteries. The same tests recover keys planted in control texts 98 to 100 per cent of the time.

The work also found that the text follows rules of its own. Words that can end in -m or -in are written with -m far more often at the end of a line, as scribes shortened words to fit, and the first word of each herbal page appears on no other page, as a plant's name would.

None of this reads the manuscript. No proposed reading, including the project's own first attempt in March 2026, has produced a dictionary word on pages it was not built from. Every test, dataset and result is public at mysteriesdept.com, where researchers can rerun them.