New research is offering a fresh way to think about one of history’s most enduring enigmas: the Voynich Manuscript. Long described as the world’s most mysterious book, the early 15th-century manuscript is filled with an unknown script and strange illustrations of plants, astrological diagrams, castles, and human figures. Despite more than a century of analysis, no one has convincingly explained who wrote it, what it says, or even whether its text carries meaning at all.
Two pages from the mysterious Voynich Manuscript, which remains undeciphered to this day. Public domain
A recent peer-reviewed study published in Cryptologia does not claim to solve the mystery, but it shows that the manuscript could plausibly have been produced using a cipher that was within medieval technological capabilities. The work was conducted by science journalist Michael Greshko, who wanted to test whether a historically realistic encryption system could generate text with the same unusual statistical features seen in the Voynich Manuscript.
The outcome is a proposed method, termed the “Naibbe cipher,” the name derived from a medieval Italian card game. Instead of decoding the manuscript, the cipher goes the opposite way: It takes ordinary Latin or Italian text and turns it into glyph sequences that closely resemble Voynichese. It does this by breaking continuous text up into short groupings of letters and then substituting them using structured tables, with elements of randomness introduced through tools such as dice and playing cards—objects that were widely available in 15th-century Europe.
The Extensible Voynich Alphabet (EVA), used throughout this paper to transliterate Voynichese into the Latin alphabet. Credit: Michael A. Greshko, Cryptologia (2025)
Applied to a variety of sample texts, the Naibbe cipher produces outputs that match many of the key properties of the Voynich Manuscript: the frequency of symbols, the typical length of “words,” and certain positional patterns that have long driven academics to perplexity. What is more, the method preserves fragments of the original linguistic structure in the form of short letter sequences, even though no single glyph consistently corresponds to a single plaintext letter.
The results indicate that the long-debated “cipher hypothesis” remains viable. The study simultaneously constrains what such a cipher would have looked like: any real system underlying the manuscript was highly complex and quite unlike conventional substitution ciphers. The research also allows for competing interpretations, saying that the text shows a constructed system, an unknown language, or even an elaborate forgery.
Experts not involved in the study have welcomed the work as a useful benchmark, not a definitive answer. The research serves to sharpen the questions future studies must address by showing that a hand-executable cipher can replicate many of the manuscript’s quirks. For now, the Voynich Manuscript remains undeciphered, but the new approach offers a clearer framework for understanding how such a baffling text might have been created and why it continues to resist simple explanations.
More information: Greshko, M. A. (2025). The Naibbe cipher: a substitution cipher that encrypts Latin and Italian as Voynich Manuscript-like ciphertext. Cryptologia, 1–37. doi:10.1080/01611194.2025.2566408
