The Voynich Manuscript
The Voynich Manuscript is an illustrated book written in a script no one can read. It is held by the Beinecke Rare Book and Manuscript Library at Yale University as MS 408.[1] Its parchment has been radiocarbon-dated to 1404–1438.[2] The book is named after Wilfrid Voynich, the book dealer who bought it in 1912.
Its pages show plants that have never been identified for certain, star charts, and small figures bathing in green pools. Since 1921 several people have announced a reading of the text; none has held up when tested on pages it was not built from.[3] Of the 23 questions on this site, 3 are answered, all about the book as an object. The most likely answer to “What kind of text is it?” is a notation or memory system, which is not yet confirmed.
Key questionsset by the editors
Found here
Work on this site has ruled out the simplest explanations. Letter-for-letter ciphers of Latin, Italian, German, French and Czech all fail, even though the same tests recover keys we plant ourselves, and so does a mixed cipher key of the kind used in fifteenth-century Italy.
What's left looks like a writing system with rules of its own: words spelled differently depending on where they fall in a line, and pages from the same sheet of parchment sharing vocabulary.
How we got here: 9 research sessionsDescription
The manuscript has 102 leaves of calfskin parchment, gathered in 18 quires. Some leaves are missing and several fold out.[1] The text runs to about 38,000 words, written left to right in a script of roughly two dozen distinct glyphs. Scholars divide the book by its pictures into six sections: herbal, astronomical, bathing, cosmological, pharmaceutical and recipes.
Lisa Fagin Davis identified five scribes from their letterforms.[4] In the 1970s Prescott Currier found two statistically distinct "languages", now called Currier A and B, which partly line up with the scribes.[5] In 2026 Layfield and Davis argued that the book was made from single leaves and is no longer in its maker's order.[6]
History
The earliest trace is a signature on the first page, erased and later recovered under ultraviolet light, of Jacobus Horcicky de Tepenec, a chemist at the court of Rudolf II in Prague.[1] In 1639 the Prague alchemist Georg Baresch wrote to the Jesuit scholar Athanasius Kircher in Rome, asking for help reading it. In 1665 Johannes Marcus Marci sent the book to Kircher with a letter, still kept with it, which repeats a story that Rudolf II had paid 600 ducats for it.
The book stayed with the Jesuits for two centuries. Voynich bought it in 1912 from the Jesuit collection at the Villa Mondragone near Rome. The dealer H. P. Kraus gave it to Yale in 1969.[1]
Timeline
Attempts to read it
William Newbold announced in 1921 that the text was a cipher by Roger Bacon, read from tiny strokes inside the letters. John Manly showed in 1931 that the strokes were cracks in the ink.[7] The cryptologist William Friedman came to suspect an early artificial language.[1]
Later work split between meaning and no meaning. Gordon Rugg showed in 2004 that text with similar features can be made with a table and a card grille.[8] Torsten Timm and Andreas Schinner proposed in 2020 that the scribes made words by copying and changing nearby words.[9] Against that, keywords cluster by section much as they do in real texts,[10] and a 2021 review of the linguistic evidence found word-level patterns consistent with meaningful text, even though the spelling is unlike any known language.[3] In 2025 Michael Greshko showed that a hand cipher of the period could produce text with similar statistics.[11]
Published claims
| Year | Who | Claimed | Status | Reason |
|---|---|---|---|---|
| 1921 | Newbold | Roger Bacon's cipher, read in microscopic strokes | Ruled outBased on something false | Refuted by Manly (1931): the strokes are cracks in the ink |
| 1944 | Friedman | An early artificial, a priori language | Still possible | Close to the second-ranked answer, an invented language |
| 1976 | Currier | Two 'languages', A and B, and several hands | Accepted | Holds; matches the scribe and spelling evidence |
| 2004 | Rugg | A hoax made with tables and a grille | Ruled outFailed its test | Simulated here; it reproduces too little of the text's structure |
| 2013 | Montemurro & Zanette | Keywords cluster by section, as in language | Accepted | Holds; the same clustering shows up here |
| 2014 | Bax | Readings of ten words, mostly plant names | Still possible | Too few words proposed to test yet |
| 2016 | Hauer & Kondrak | Anagrammed text, perhaps Hebrew | Ruled outFailed its test | Anagram tests with controls found nothing |
| 2020 | Timm & Schinner | Generated by copying and changing nearby words | Disputed | Tests here could not tell either way |
| 2020 | Davis | Five scribes | Accepted | Holds; the basis of the scribe tests |
| 2025 | Greshko | Each word encodes one letter (Naibbe cipher) | Ruled outFailed its test | Its output does not match the text's word patterns |
| 2026 | Layfield & Davis | Single leaves; the book is not in its maker's order | Accepted | Holds; the leaf evidence here agrees |
Questions
3 of 23 questions are answered: when it was made, and by how many hands. Nothing about what the text says is settled yet.
Artifacts
Every artifact is a copy of the book, or of something about it, with its own key and its distance from the object.
References
- 1.Clemens, R. (ed.) 2016. The Voynich Manuscript. Yale University Press.cited 5 times hereSRC08JHN1NG6AMW
- 2.University of Arizona 2011. UA experts determine age of book 'nobody can read'. UA News.cited 3 times hereSRC08JJE7AR0ZE1
- 3.Bowern, C., Lindemann, L. 2021. The linguistics of the Voynich manuscript. Annual Review of Linguistics.doicited 9 times hereSRC08JHTHEG6MSY
- 4.Davis, L. 2020. How many glyphs and how many scribes? Digital paleography and the Voynich manuscript. Manuscript Studies: A Journal of the Schoenberg Institute for Manuscript Studies.doicited 10 times hereSRC08JHRPVG4Z28
- 5.Currier, P. 1976. Some important new statistical findings. New Research on the Voynich Manuscript: Proceedings of a Seminar, 30 November 1976.cited 4 times hereSRC08JHQAX84PKW
- 6.Layfield, C., Davis, L. 2026. Singulion structure and the Voynich manuscript. Digital Medievalist.doicited 4 times hereSRC08JHT2SR33NX
- 7.Manly, J. 1931. Roger Bacon and the Voynich MS. Speculum.cited 2 times hereSRC08JJENZG4AS3
- 8.Rugg, G. 2004. An elegant hoax? A possible solution to the Voynich manuscript. Cryptologia.doicited 6 times hereSRC08JHY6MG0DDD
- 9.Timm, T., Schinner, A. 2020. A possible generating algorithm of the Voynich manuscript. Cryptologia.doicited 9 times hereSRC08JHZ3Y07R9B
- 10.Montemurro, M., Zanette, D. 2013. Keywords and co-occurrence patterns in the Voynich manuscript: an information-theoretic analysis. PLoS ONE.doicited 8 times hereSRC08JHX9B03EPJ
- 11.Greshko, M. 2025. The Naibbe cipher: a substitution cipher that encrypts Latin and Italian as Voynich manuscript-like ciphertext. Cryptologia.doicited 9 times hereSRC08JHZJJR0Z8T