# The Voynich manuscript is an undeciphered book hiding secret knowledge

**No verdict.** No proposed decipherment has survived expert scrutiny, but the text's statistical structure resembles real language closely enough that a meaningless hoax is not a clean fit either. A century on, no one actually knows what it is.

Category: Lost History & Ancient Mysteries · Era: 15th century · First circulated: 1912 (rediscovery); manuscript itself dates to c. 1404–1438 · Believed by: Cryptography's most famous unsolved document
URL: https://theconspiratory.com/theory/voynich-manuscript

## Summary
A 15th-century illustrated codex written in a script no one has ever read, full of botanical, astronomical and biological drawings that match nothing known: studied and defeated by professional cryptographers, WWII codebreakers and linguists for over a century, with every claimed 'solution' collapsing under scrutiny.

## The claim
That the Voynich manuscript, an illustrated book in an unidentified script and language, encodes real and meaningful content (whether a secret cipher, a forgotten or invented language, or genuine esoteric or scientific knowledge) that would be legible if only the code were broken.

## Origin and timeline
- c. 1404–1438: Radiocarbon dating of the vellum (2009, published 2011) places the manuscript's creation in the early 15th century, roughly a century earlier than many earlier guesses.
- 1666: Prague scholar Johannes Marcus Marci sends the book to Jesuit polymath Athanasius Kircher, in a cover letter (preserved with the manuscript) noting it had reputedly once belonged to Holy Roman Emperor Rudolf II.
- 1912: Polish-American antiquarian book dealer Wilfrid M. Voynich acquires the manuscript from the Jesuit Collegio Romano's holdings near Frascati, Italy, and brings it to public attention, giving it his name.
- 1940s–1950s: William F. Friedman, the era's foremost American cryptanalyst, organizes informal teams (including NSA colleagues) to attack the text; decades of work leave him convinced it is likely an invented or artificial language rather than a straightforward cipher.
- 1969: Rare book dealer Hans P. Kraus, unable to find a buyer, donates the manuscript to Yale University's Beinecke Rare Book & Manuscript Library, where it is catalogued as MS 408.
- 2011: A University of Arizona radiocarbon dating team publishes results confirming the vellum dates to 1404–1438; separate microscopy by McCrone Associates finds the inks and pigments consistent with the same period, arguing against a modern forgery.
- 2013–2014: Peer-reviewed statistical studies (Montemurro & Zanette in PLOS ONE; Stephen Bax's proposed partial reading) reopen debate over whether the text encodes real linguistic structure; Bax's specific word identifications are widely rejected by specialists.
- 2017–2020: Widely publicized 'solved' claims (a Times Literary Supplement piece proposing a Latin medical-abbreviation code, and a University of Bristol researcher's 'proto-Romance' reading) are both rejected by medievalists and linguists within weeks.

## The evidence, claim by claim
- Claim: The text isn't random; its statistics behave like a real language, so it must be meaningful.
  Evidence: This is true and important (word-frequency and entropy studies do show language-like structure), but it is not proof of meaning. Some engineered systems (verbose ciphers, certain constructed languages, or specific generative table-and-grille methods) can also produce language-like statistics without encoding a message a reader today could recover. Structure is consistent with meaning; it does not confirm it.
- Claim: Multiple qualified researchers have announced they 'solved' the manuscript.
  Evidence: Every widely publicized claim to date (including a 2014 'proto-Romance' partial reading and a 2017 claim that it was a women's-health manual using Latin abbreviations) produced translations that specialists in medieval Latin, linguistics or codicology found ungrammatical, cherry-picked, or unfalsifiable, and none has been independently reproduced or accepted by the field.
- Claim: The book must encode valuable knowledge, or no one would have gone to such trouble to write and illustrate it.
  Evidence: The effort invested is real but is not evidence of content: elaborate, labor-intensive hoaxes, private notation systems, and constructed languages with no practical secret to protect have all been documented elsewhere in history, and the manuscript's drawings of unidentifiable plants and impossible astronomical arrangements are as consistent with invented or symbolic imagery as with a real, lost body of knowledge.
- Claim: Professional codebreakers, including WWII and NSA cryptanalysts, failed to crack it, so it must be an extremely sophisticated cipher.
  Evidence: Their failure is well documented, but the cryptanalysts who spent the most time on it (most notably William Friedman) did not conclude it was an unbroken sophisticated cipher; his own tentative conclusion was that it was more likely an invented language, precisely because it resisted every classical cryptographic attack rather than yielding under one.

## Why people believe it
- The manuscript genuinely is unsolved by any rigorous, replicated standard, after more than a hundred years of attention from serious professionals: an authentic mystery, not a manufactured one, which makes it fertile ground for confident claims of all kinds.
- Beautiful, alien-looking illustrations of unidentifiable plants, star charts and nude figures in tubs invite pattern-matching to esoteric traditions (alchemy, herbal medicine, lost civilizations) that feel more satisfying than 'unknown.'
- A real, peer-reviewed finding (that the text's statistics resemble natural language) gets popularly flattened into 'so it definitely says something,' skipping the harder and less satisfying truth that structure alone cannot prove meaning.
- Each failed decipherment still generates a wave of credulous press coverage before the correction catches up, so the base rate of 'I heard it was solved' in public memory is far higher than the true rate of zero.
- As one of the few remaining undeciphered historical documents freely viewable online in full, it functions as an open invitation and a status prize: the amateur or academic who cracks it would join Champollion and the WWII Enigma team in the history books.

## Open questions
- Whether Voynichese encodes a real natural language, a cipher laid over one, a wholly constructed or invented language, or elaborate meaningless glossolalia engineered to imitate language statistically: word-frequency and entropy patterns argue against pure randomness, but at least one documented hoax-generation method has reproduced those same patterns, and no analysis has yet shown which explanation is correct.
- Why not one of the many well-publicized decipherment claims made across more than a century (including sustained mid-20th-century attempts by NSA-caliber cryptanalysts, a widely covered 2014 partial reading, and a 2017 'solved it' announcement that made international headlines) has gained lasting acceptance among specialists in medieval languages, cryptography, or manuscript studies.
- What the manuscript's roughly 113 unidentified plant drawings and its section of nude women bathing in interconnected pools and pipework actually depict (whether stylized or composite real botanical and medical imagery, invented or symbolic content, or something else entirely) and why, unlike the star charts, no proposed identification for either has held up under scrutiny.
- Who actually created the manuscript, and why. No author has ever been credibly identified; radiocarbon dating alone eliminated the once-popular guess that medieval friar and scientist Roger Bacon was responsible, and without a known author, workshop, or stated purpose, even a confirmed translation would still leave open why this specific book was made.

## Sources
- Voynich Manuscript (MS 408): collection record and full digital facsimile, Beinecke Rare Book & Manuscript Library, Yale University: https://beinecke.library.yale.edu/collections/highlights/voynich-manuscript
- Radiocarbon dating of the Voynich manuscript's vellum (4-sample AMS analysis, 95% confidence interval 1404–1438), University of Arizona Accelerator Mass Spectrometry Laboratory (G. Hodgins), presented 2009–2012 (2011): https://www.voynich.nu/extra/carbon.html
- Keywords and Co-Occurrence Patterns in the Voynich Manuscript: An Information-Theoretic Analysis, Montemurro, M.A. & Zanette, D.H., PLOS ONE 8(6): e66344 (2013): https://journals.plos.org/plosone/article?id=10.1371%2Fjournal.pone.0066344
- The Linguistics of the Voynich Manuscript, Bowern, C. & Lindemann, L., Annual Review of Linguistics, Vol. 7 (2021): https://www.annualreviews.org/content/journals/10.1146/annurev-linguistics-011619-030613
- What We Know About The Voynich Manuscript, Reddy, S. & Knight, K., Proceedings of the 5th ACL-HLT Workshop on Language Technology for Cultural Heritage, Social Sciences, and Humanities, pp. 78-86 (2011): https://aclanthology.org/W11-1511/
- Hoaxing statistical features of the Voynich Manuscript, Rugg, G. & Taylor, G., Cryptologia 41(3), pp. 247-268 (2016): https://www.tandfonline.com/doi/full/10.1080/01611194.2016.1206753
- William F. Friedman's Transcription of the Voynich Manuscript, Jim Reeds (documenting Friedman/NSA study group records): https://www.apprendre-en-ligne.net/crypto/mystere/wff.pdf
- Yet again the Voynich manuscript (Lisa Fagin Davis's on-the-record rejection of the 2017 Gibbs decipherment claim), Language Log, University of Pennsylvania (2024): https://languagelog.ldc.upenn.edu/nll/?p=65930
- Voynich manuscript, Wikipedia: https://en.wikipedia.org/wiki/Voynich_manuscript

Rated by The Conspiratory, a neutral, sourced encyclopedia of conspiracy theories. Full page: https://theconspiratory.com/theory/voynich-manuscript