The Voynich Manuscript

Overview

Somewhere in the climate-controlled vaults of Yale University’s Beinecke Rare Book and Manuscript Library sits a book that has humiliated some of the greatest minds in the history of cryptography. Catalogued as MS 408, the Voynich Manuscript is roughly 240 pages of handwritten text in a script that no one — not medieval scholars, not WWII codebreakers, not modern AI — has been able to read. Its pages overflow with illustrations of plants that don’t exist on Earth (or at least don’t match any known species), astronomical diagrams of uncertain purpose, naked women bathing in pools of green liquid connected by elaborate plumbing, and what appear to be pharmaceutical recipes for medicines no one can identify.

It has been called the world’s most mysterious manuscript, and that designation isn’t hyperbole. The Voynich Manuscript isn’t some scrap of parchment with a few cryptic symbols scratched on it. It’s a full-length book — a codex, bound and illustrated with obvious care, written in a consistent hand using an alphabet of roughly 20 to 30 distinct characters that appear nowhere else in the historical record. The text flows left to right. Words are separated by spaces. Paragraphs have structure. The writing shows all the hallmarks of someone communicating meaningful information. And yet, for over a century, that information has remained locked behind a cipher that no key has fit.

What makes the Voynich Manuscript a conspiracy magnet isn’t just the mystery of the script. It’s the sheer improbability of the whole thing — a medieval book that appears to encode real knowledge in a system so sophisticated that modernity’s best tools can’t crack it. That combination has spawned theories ranging from the mundane (it’s a pharmacist’s notebook in an obscure shorthand) to the extraordinary (it’s the work of time travelers, or aliens, or an interdimensional being dictating forbidden knowledge to a medieval scribe). The truth, whatever it is, has so far proven stranger than any of the fiction.

Discovery and Provenance

The modern history of the Voynich Manuscript begins in 1912, when Wilfrid Voynich — a Polish-Lithuanian book dealer with a gift for sniffing out rare manuscripts — purchased a collection of texts from the Jesuit college at Villa Mondragone, a Renaissance palace near Frascati in the hills south of Rome. Among the acquisitions was a peculiar handwritten book with no title, no author attribution, and no text anyone could read.

Voynich was electrified. He recognized immediately that he had something extraordinary, though exactly what remained a question. Tucked inside the manuscript was a letter dated 1665 or 1666, written by Johannes Marcus Marci, the rector of Charles University in Prague, to the Jesuit polymath Athanasius Kircher in Rome. Marci was sending Kircher the manuscript in the hope that the famous scholar could decipher it. The letter mentioned that the book had previously been owned by Emperor Rudolf II of the Holy Roman Empire, who had reportedly purchased it for 600 gold ducats — a staggering sum, equivalent to a small fortune. Rudolf, Marci wrote, believed the book to be the work of the English friar Roger Bacon, the 13th-century pioneer of experimental science.

This provenance chain — Rudolf II to Marci to Kircher — is backed by solid documentary evidence and places the manuscript in Prague’s intellectual circles by the late 16th century at the latest. But what happened before Rudolf acquired it remains almost entirely blank. The book seems to materialize from nowhere into the court of one of Europe’s most eccentric monarchs, a man whose Prague castle was a cabinet of curiosities overflowing with alchemical apparatus, astronomical instruments, and esoteric objects of every description. Rudolf was exactly the kind of buyer who would drop 600 ducats on a book nobody could read.

After Kircher received the manuscript, it apparently gathered dust in the Jesuit archives for over two centuries until Voynich’s discovery. Voynich spent years promoting the manuscript and trying to get it deciphered, convinced it contained lost knowledge of genuine scientific importance. He died in 1930 without a solution. His widow, Ethel, held onto it until her death in 1960, at which point it passed to her friend Anne Nill, who sold it to the antiquarian book dealer Hans P. Kraus in 1961. Kraus tried to sell the manuscript for $160,000 — an astronomical price for the era — but couldn’t find a buyer. In 1969, he donated it to Yale’s Beinecke Library, where it has remained ever since, freely available for study, stubbornly refusing to give up its secrets.

What’s Actually in the Book

The Voynich Manuscript’s content falls into roughly six sections, though scholars disagree on the exact divisions. Each section has a distinctive character, and all of them are deeply strange.

The Botanical Section

The largest section of the manuscript — comprising roughly half its pages — features illustrations of plants accompanied by blocks of text. At first glance, this looks like a medieval herbal, the kind of botanical reference book that was common in the 15th century. But there’s a problem: most of the plants depicted don’t match any known species. Some have been tentatively identified — a few bear passing resemblance to sunflowers, water lilies, or ferns — but the majority are botanical fever dreams, with roots that look like organs, leaves with impossible geometries, and flowers that appear to be invented from scratch.

This is one of the manuscript’s deepest puzzles. Medieval herbals were functional documents. They existed to help physicians and apothecaries identify plants for medicinal use. Why would someone create a lavishly illustrated herbal of imaginary plants? The possible answers are all troubling: either the illustrator was depicting real plants so stylistically that they became unrecognizable, or they were drawing from a botanical tradition unknown to us, or the entire section is deliberate fantasy.

The Astronomical Section

Several pages contain circular diagrams featuring stars, suns, and moons arranged in patterns that suggest astronomical or astrological charts. Some diagrams show concentric rings with text running along them. Others depict what appear to be zodiac symbols, though they don’t correspond precisely to conventional Western or Eastern zodiacal systems. A few pages show sequences of stars that might represent constellations, though no definitive identification has been made.

One particularly arresting set of pages — sometimes called the “cosmological” section — features large, elaborate circular diagrams that could be maps, calendars, or schematic representations of something entirely outside modern categories. Some researchers have noted similarities to medieval diagrams of the cosmos, but the details resist confident interpretation.

The Biological Section

Here’s where things get truly weird. A series of pages depict small nude female figures — dozens of them — immersed in pools and channels of green liquid. The women are connected by an elaborate network of tubes and pipes, some of which appear to flow into or out of their bodies. The overall effect is somewhere between a medieval bath house, an alchemical diagram, and something out of a science fiction film.

No satisfactory explanation has been offered for these images. They don’t correspond to any known medieval artistic or scientific tradition. Some researchers have suggested connections to balneological (therapeutic bathing) practices, others to alchemical processes using the human body as metaphor. A few have speculated about fertility rituals or gynecological texts. None of these interpretations fully account for the elaborate green plumbing.

The Pharmaceutical Section

The final major section features drawings of containers — jars, vessels, and what appear to be pharmaceutical receptacles — alongside plant parts (roots, leaves, stems shown separately rather than as whole plants). This section has the strongest resemblance to known medieval genres, specifically the pharmacopoeia or recipe book. But without the ability to read the text, the recipes remain sealed.

The Codebreakers’ Graveyard

The Voynich Manuscript has attracted cryptanalysts the way a flame attracts moths, and with roughly the same outcome. The roster of brilliant minds who have thrown themselves at it and failed is almost as remarkable as the manuscript itself.

William Friedman and the First Systematic Assault

The most significant early attempt came from William Friedman, arguably the greatest cryptanalyst in American history. Friedman was the man who broke Japan’s PURPLE diplomatic cipher before World War II — one of the most consequential intelligence achievements of the 20th century. He founded what would become the National Security Agency’s codebreaking apparatus. If anyone alive could crack the Voynich Manuscript, it was Friedman.

He first encountered the manuscript in the 1920s and became obsessed. He assembled study groups — first informally during the 1940s, then an official “First Voynich Manuscript Study Group” at the NSA in the early 1950s. These were not amateurs. They were professional codebreakers with access to the most advanced analytical tools available.

Friedman’s working hypothesis was that the Voynich Manuscript was written in an early form of a constructed language — an artificial, logically structured language of the kind that philosophers would later attempt in the 17th century (like John Wilkins’ “Real Character”). This was a sophisticated guess, and it was wrong — or at least, Friedman couldn’t prove it right. He worked on the problem intermittently for over thirty years until his death in 1969, never arriving at a solution. His papers, now declassified and held at the George C. Marshall Foundation, reveal the full scope of his frustration.

The Statistical Evidence

One of the most important outcomes of the codebreaking efforts has been a detailed statistical picture of the Voynich text. And here’s the thing that keeps researchers coming back: the text is demonstrably not random.

The character distribution follows patterns consistent with natural language. The text obeys Zipf’s law — the linguistic principle that in any natural language, the most frequent word occurs roughly twice as often as the second most frequent, three times as often as the third, and so on. Random or meaningless text does not follow this distribution. The Voynich text does.

Word-length distributions, character-pair frequencies, and entropy measurements all fall within the ranges expected for natural language text. The manuscript has distinct “dialects” — the statistical properties of the text shift between sections, as if different subject matter is being discussed using specialized vocabulary. There are even apparent labels on illustrations, short words or phrases that seem to identify the depicted objects.

All of this suggests — though it doesn’t prove — that the text carries genuine linguistic content. It’s not gibberish. It’s not a string of random symbols. Somebody was saying something. The question is what, and in what language.

Modern Computational Attempts

The advent of computers and later artificial intelligence has brought new tools to bear on the problem, but so far without a breakthrough. In 2016, researchers using neural networks claimed to have identified Hebrew as the likely source language, but their methodology was criticized and the claim hasn’t held up. In 2019, Gerard Cheshire of the University of Bristol published a paper claiming the manuscript was written in “proto-Romance” — an early, unattested precursor to modern Romance languages — and that he’d decoded it. Medievalists responded with something between skepticism and exasperation. Lisa Fagin Davis, then executive director of the Medieval Academy of America, called the claim “not backed up by the evidence” and noted that Cheshire’s supposed translations were essentially unfalsifiable — you can read almost anything into an undeciphered text if your linguistic framework is flexible enough.

The pattern of claimed solutions followed by immediate debunking has repeated itself so many times that Voynich scholars now greet new claims with weary suspicion. As one researcher put it, “The Voynich Manuscript is a mirror — people see in it whatever they’re looking for.”

The Hoax Hypothesis

If the manuscript can’t be read because it doesn’t actually say anything, a lot of the mystery evaporates. The hoax theory has serious proponents and comes in several flavors.

Edward Kelley and the Elizabethan Con

The most historically grounded hoax theory centers on Edward Kelley, a notorious 16th-century English occultist and confidence man. Kelley was the scryer (crystal gazer) who worked with the mathematician John Dee, claiming to receive messages from angels in a language he called “Enochian.” Kelley spent time at the court of Rudolf II in Prague during the 1580s and 1590s — exactly the period when Rudolf is supposed to have acquired the manuscript.

The theory runs like this: Kelley, always in need of money and possessing both the audacity and the esoteric knowledge to pull it off, fabricated the manuscript and sold it to Rudolf as a genuine work of Roger Bacon. Rudolf, who collected everything from bezoard stones to clockwork automata, would have been an ideal mark. Kelley had motive, means, and opportunity.

The problem with this theory is the carbon dating. In 2009, researchers at the University of Arizona used radiocarbon dating to test the manuscript’s vellum (the prepared animal skin it’s written on) and dated it to between 1404 and 1438 — over a century before Kelley was born. The inks were also tested and found to be consistent with the same period. Unless Kelley got his hands on a large supply of blank 15th-century vellum (unlikely but not impossible — old vellum was sometimes scraped and reused), he didn’t write it.

Wilfrid Voynich Himself

A more modern hoax theory accuses Voynich himself of fabricating the manuscript, or at least the text. Voynich was a book dealer; he stood to profit enormously from the discovery of an impossibly rare medieval manuscript. He had access to old vellum and period-appropriate inks. He was known to be shrewd and ambitious.

But this theory has its own problems. The Marci-to-Kircher letter is genuine and has been authenticated through multiple independent lines of evidence. If Voynich faked the manuscript, he somehow also faked — or planted — a 17th-century letter with a verifiable provenance chain. Moreover, the manuscript’s text exhibits sophisticated statistical properties that would have been essentially impossible to fake in 1912. Voynich would have needed knowledge of information theory and computational linguistics decades before those fields existed.

The Gordon Rugg Experiment

In 2004, computer scientist Gordon Rugg of Keele University demonstrated that text with Voynich-like statistical properties could be generated using a Cardan grille — a 16th-century encryption device consisting of a card with holes cut in it, placed over a table of syllables. By shifting the grille across the table, one could produce large volumes of text that looked structured but was actually meaningless.

Rugg’s experiment was a genuine contribution: it showed that the Voynich text’s statistical properties didn’t necessarily prove it was meaningful language. But it didn’t prove the manuscript was a hoax either. Showing that something could have been faked is not the same as showing it was faked. And Rugg himself noted that producing the full manuscript using a Cardan grille would have been an enormous undertaking, requiring weeks of tedious mechanical effort — possible, but hardly the quick con that most hoax theories imagine.

Competing Theories

Beyond code and hoax, the Voynich Manuscript has generated a spectacular ecosystem of alternative explanations.

A Lost or Unknown Language

Perhaps the simplest explanation is also one of the hardest to evaluate: the manuscript is written in a real language that has simply never been identified. Hundreds of languages have existed throughout history without leaving written records. If the author belonged to a small linguistic community — a isolated regional group, a secretive guild, a heretical sect — their language might have vanished without a trace, leaving the Voynich Manuscript as its sole surviving document.

This theory is consistent with the text’s statistical properties but faces a practical objection: without a bilingual text (a Rosetta Stone equivalent), a truly unknown language written in a unique script may be fundamentally undecipherable. We might never have enough data to crack it.

Encoded Natural Language

The manuscript could be a natural language — Latin, Italian, German, or something else common in 15th-century Europe — encoded through a sophisticated cipher. This was Friedman’s terrain, and the fact that neither he nor his successors could break it suggests that if it is a cipher, it’s extraordinarily good. Some researchers have proposed that it could use a nomenclator system (where symbols stand for common words and names rather than individual letters) or a polyalphabetic cipher, either of which would be far ahead of its time for the early 15th century. The Alberti cipher disk, generally considered the first polyalphabetic cipher, wasn’t invented until the 1460s — after the manuscript’s vellum was produced.

Glossolalia and Automatic Writing

A more left-field proposal suggests the text is the product of glossolalia — “speaking in tongues” — or a similar state of altered consciousness. Under this theory, the author entered a trance state and produced text that felt meaningful but wasn’t connected to any actual language. This would explain both the text’s language-like statistical properties (the unconscious mind might naturally produce structured output) and its resistance to decipherment (there’s no underlying code to break).

This theory is virtually unfalsifiable, which is both its strength and its weakness. It can explain everything, which means it explains nothing.

Extraterrestrial or Temporal Origin

At the far end of the speculation spectrum sit the theories that the manuscript was produced by non-human intelligence — aliens who visited Earth and left behind a record, or time travelers who brought back a document from a future (or alternate) civilization. These theories enjoy significant popularity online but rest on no evidence whatsoever beyond the manuscript’s resistance to conventional explanation. The vellum is demonstrably from 15th-century animals. The inks are period-appropriate. The binding is medieval. Everything physical about the manuscript says it was made on Earth, by humans, in the early 1400s.

The Roger Bacon Theory

The attribution to Roger Bacon — the 13th-century Franciscan friar who wrote about optics, alchemy, and the scientific method centuries before it became fashionable — was the first major theory about the manuscript’s authorship and persisted for decades. Voynich himself promoted this idea vigorously, and it had a certain romantic appeal: Bacon was known to have used ciphers, he had wide-ranging scientific interests, and he had been imprisoned by the Church for his unorthodox ideas, giving him plausible motive to encode his work.

The 2009 radiocarbon dating killed this theory. Bacon died in approximately 1292. The manuscript’s vellum dates from 1404 to 1438 — a gap of over a century. Unless Bacon was writing on vellum that wouldn’t be manufactured for another hundred years, he didn’t produce the Voynich Manuscript.

It remains possible that the manuscript is a copy of an earlier Baconian work, transcribed onto 15th-century vellum. But this is speculation without supporting evidence, and it doesn’t explain the unique script, which has no known precedent in Bacon’s authentic writings or anywhere else.

Why It Matters

The Voynich Manuscript occupies a unique position in the landscape of historical mysteries. Unlike most conspiracy-adjacent puzzles, it’s not a matter of disputed evidence or competing interpretations of known facts. The manuscript is a physical object. It exists. You can visit it at Yale (though you’ll need to request access through the Beinecke’s reading room). High-resolution scans are available online for anyone to study. The mystery isn’t about access or suppression — it’s about genuine, stubborn incomprehension.

This makes it unusually honest as mysteries go. There’s no shadowy organization hiding the truth. There’s no government coverup. The manuscript has been publicly available for over a century, studied by some of the finest analytical minds in history, and it simply will not yield. In a world where conspiracy theories often thrive on the claim that someone is hiding something, the Voynich Manuscript is a rare case where the something is right there in front of everyone, and nobody can figure out what it is.

It also raises uncomfortable questions about the limits of modern knowledge. We live in an age where we’ve sequenced the human genome, detected gravitational waves, and photographed black holes. The idea that a medieval book can defeat all of our computational and linguistic firepower is, frankly, humbling. Either someone in the 15th century was far more sophisticated than we give them credit for, or there are categories of knowledge that our current frameworks simply cannot access. Neither possibility is particularly comfortable.

Timeline

• c. 1404–1438 — Radiocarbon-dated range for the manuscript’s vellum production
• Late 1500s — Emperor Rudolf II of the Holy Roman Empire reportedly purchases the manuscript for 600 gold ducats
• 1665–1666 — Johannes Marcus Marci sends the manuscript to Athanasius Kircher in Rome
• c. 1666–1912 — The manuscript sits in Jesuit archives, apparently unexamined
• 1912 — Wilfrid Voynich purchases the manuscript from Villa Mondragone near Rome
• 1919 — William Romaine Newbold claims to have partially deciphered the manuscript; his solution is later discredited
• 1921 — Voynich begins publicly exhibiting the manuscript and soliciting decipherment attempts
• 1930 — Wilfrid Voynich dies without a solution
• 1940s–1960s — William Friedman leads intermittent study groups at the NSA; all efforts fail
• 1961 — Hans P. Kraus purchases the manuscript from Voynich’s estate
• 1969 — Kraus donates the manuscript to Yale’s Beinecke Rare Book and Manuscript Library
• 2004 — Gordon Rugg demonstrates that Voynich-like text could be generated with a Cardan grille
• 2009 — University of Arizona radiocarbon dates the vellum to 1404–1438, ruling out Roger Bacon’s authorship
• 2014 — Stephen Bax claims partial decipherment using linguistic analysis; results disputed
• 2019 — Gerard Cheshire claims “proto-Romance” solution; immediately rejected by medievalists
• 2020s — Multiple AI and machine learning studies attempt decipherment; none produce accepted results

Sources & Further Reading

• Primary Source: Yale University Beinecke Library, MS 408. High-resolution scans available online
• D’Imperio, Mary. The Voynich Manuscript: An Elegant Enigma. National Security Agency, 1978. (Declassified NSA monograph — the most comprehensive early study)
• Kennedy, Gerry, and Rob Churchill. The Voynich Manuscript: The Mysterious Code That Has Defied Interpretation for Centuries. Inner Traditions, 2006.
• Rugg, Gordon. “An Elegant Hoax? A Possible Solution to the Voynich Manuscript.” Cryptologia 28, no. 1 (2004): 31–46.
• Amancio, Diego R., et al. “Probing the statistical properties of unknown texts: application to the Voynich Manuscript.” PLoS ONE 8, no. 7 (2013).
• Tucker, Arthur O., and Rexford H. Talbert. “A Preliminary Analysis of the Botany, Zoology, and Mineralogy of the Voynich Manuscript.” HerbalGram 100 (2013): 70–85.
• Zandbergen, René. voynich.nu — The most comprehensive online resource for Voynich Manuscript research.

Related Theories

The Voynich Manuscript connects to broader questions about ancient advanced technology and whether significant bodies of historical knowledge have been suppressed by institutional authorities. While the manuscript itself doesn’t require any conspiratorial explanation — it may simply be undeciphered rather than suppressed — its sheer resistance to modern analysis feeds into narratives about the limits of conventional knowledge and the possibility that the past held secrets we’ve lost the ability to access.