The Voynich Manuscript: The Book Nobody Can Read

Six Centuries of Silence

Housed in Yale University's Beinecke Rare Book and Manuscript Library under catalog number MS 408, the Voynich Manuscript is a 240-page illustrated codex written entirely in an unknown script. Radiocarbon dating places its vellum (calfskin parchment) between 1404 and 1438. Despite more than a century of study by professional cryptographers, linguists, historians, and amateur sleuths, not a single word of its text has been reliably deciphered. The manuscript remains the most studied undeciphered document in the world.

Physical Description and Layout

The manuscript measures approximately 23.5 by 16.2 centimeters and contains 240 pages, though some are missing from the original binding. Several pages fold out to reveal larger illustrations. The text is written left to right in a flowing script that shows no corrections, erasures, or hesitation marks — suggesting the scribe was either copying from another source or writing fluently in whatever system this represents.

Section Breakdown

The illustrations are as puzzling as the text. Roughly 75 percent of the plants depicted in the herbal section do not correspond to any known species. Some appear to be composites — roots from one plant grafted onto leaves from another. Whether this reflects artistic license, botanical ignorance, or deliberate obfuscation is unknown.

The Script: Familiar Yet Unreadable

The Voynich script contains between 20 and 30 distinct character types, depending on how analysts group variant forms. The characters resemble Latin letters in some cases but combine in ways that no known European language follows. Statistical analysis reveals consistent patterns.

• Word length averages 4 to 5 characters, shorter than most European languages
• Certain characters appear almost exclusively at the beginning, middle, or end of words — a feature seen in natural languages but unusual in random or encrypted text
• The text follows Zipf's law, meaning a small number of words appear very frequently while most appear rarely — a hallmark of natural language
• There is virtually no repetition of adjacent words, which is uncommon in natural prose but could indicate a constructed or encoded system
• Different sections of the manuscript show slightly different statistical profiles, as if written in different registers or dialects

These patterns have led most researchers to conclude the manuscript is not random gibberish. Something structured is happening. The question is what.

A History of Ownership and Rediscovery

The earliest confirmed owner was Georg Baresch, a 17th-century Prague alchemist who described it as a mysterious book "taking up space uselessly" in his library. After his death, it passed to Jan Marek Marci, rector of Prague University, who sent it to Athanasius Kircher, a Jesuit polymath in Rome, in 1666. The accompanying letter mentioned that the manuscript had previously belonged to Emperor Rudolf II, who reportedly paid 600 gold ducats for it — equivalent to roughly $90,000 today.

The manuscript vanished for over two centuries. In 1912, Wilfrid Voynich, a Polish-Lithuanian book dealer, purchased it from the Jesuit college at Villa Mondragone near Rome. He spent the rest of his life trying to decode it. After passing through several hands, it was donated to Yale in 1969.

Failed Decipherment Attempts

The list of people who have tried — and failed — to crack the Voynich code reads like a roster of 20th-century cryptographic talent.

If World War II codebreakers who cracked Enigma and PURPLE could not decipher it, the Voynich Manuscript is either encrypted with extraordinary sophistication or operates on principles that cryptanalysis is not equipped to handle.

Leading Theories

No single theory commands consensus. The major hypotheses fall into three categories.

Natural Language Hypothesis

Some researchers argue the text encodes a real language — possibly an obscure European dialect, a Central Asian tongue, or even an early attempt at a universal language. In 2014, botanist Arthur Tucker proposed that some plants matched species native to Mexico, suggesting a Mesoamerican origin. Linguist Stephen Bax claimed to have identified a handful of words by matching plant illustrations to their names in various languages. Neither claim has been widely accepted.

Cipher or Code

The text could be a natural language encrypted with a substitution or transposition cipher. The difficulty is that no known cipher from the 15th century produces text with the Voynich Manuscript's specific statistical properties. The lack of common cipher artifacts — repeated patterns, tell-tale letter frequencies — makes this hypothesis difficult to sustain.

Elaborate Hoax

Gordon Rugg demonstrated in 2004 that a Cardan grille (a card with windows placed over a table of syllables) could generate text with Voynich-like statistical properties. This proved a hoax was technically possible with period technology. It did not prove the manuscript is a hoax. The 240 pages of detailed illustrations represent thousands of hours of work — an extraordinary investment for a forgery.

Why It Still Matters

The Voynich Manuscript endures as a challenge because it sits at the intersection of multiple disciplines: linguistics, cryptography, art history, botany, and medieval studies. It resists the tools of each. Natural language processing algorithms trained on dozens of known languages have found no match. Machine learning approaches have produced contradictory results. Every few years, someone announces a breakthrough that collapses under peer review within weeks.

The manuscript is freely available online through the Beinecke Library's digital collection. Anyone can examine it. No one has read it. After six centuries, MS 408 remains exactly what Georg Baresch called it — a book that takes up space, beautifully and maddeningly, without revealing a single secret.