A Book Written for No One

Picture this: sometime in the early fifteenth century, someone takes an expensive piece of vellum — not paper, calfskin, the kind that cost real money — and covers it in a script that exists nowhere else on earth. Not before. Not after. Not in any monastery, not in any archive in Europe. Six hundred years of searching, and not a single trace of anyone continuing this tradition. One manuscript. One writing system. One author whose name nobody knows.

That's the real mystery of the Voynich Manuscript. Not that no one can read it. That it exists at all.

Radiocarbon dating puts the vellum somewhere between fourteen-oh-four and fourteen-thirty-eight — the early Renaissance, the era of Jan Hus and the first humanists. And right in that window, someone invents a unique alphabet, writes out roughly a hundred and seventy thousand characters, something like thirty-five thousand "words" — and vanishes. No sequel. No students. Not so much as a footnote in anyone else's letters.

But here's what's striking about the text itself: it isn't chaos. Certain characters only ever appear at the start of a word, others only at the end. There are recurring combinations, "rare" words and "favorite" words — exactly the way a living language behaves. Physicist Marcelo Montemurro, at the University of Manchester, showed that in the botanical and pharmaceutical sections, the same "terms" cluster together, as if the author were writing a coherent treatise with a specific topic per chapter. Random gibberish doesn't behave that way. Faking that level of structure would take at least as much grammatical know-how as any medieval scholar had.

And there are at least two hands in the manuscript — call them "A" and "B." Slightly different handwriting, noticeably different vocabulary, almost like two dialects of the same language. Both scribes draw their characters confidently, in one or two continuous, unhesitating strokes — the way people write in their own native alphabet, not one they've just invented. The illustrations, though, are colored more crudely, painted on top of the finished text, as if a third person — less skilled — added the color after the main writing was already done.

Picture the pages: nude female figures in strange pipes and tubs — the so-called "biological" section. Botanical drawings of plants nobody can definitively identify. Astrological diagrams. Something that looks like a book of recipes or remedies. All of it bound into one volume, as if the author were building not just a text, but a layered artifact.

The manuscript's own history is just as tangled. The first documented mention comes in sixteen thirty-nine: a Prague alchemist named Georg Baresch writes to the Jesuit scholar Athanasius Kircher in Rome, saying he owns a "mysterious book" covered in unreadable writing, and hoping Kircher — an expert in Coptic and other Eastern languages — might make sense of it. By all accounts, Kircher couldn't. The manuscript then resurfaces among alchemists and collectors of curiosities. Legend has it — and it is only legend, with no documents to back it up — that Emperor Rudolf the Second paid six hundred ducats for it, believing he was holding a lost work by Roger Bacon. In nineteen twelve, a rare-book dealer named Wilfrid Voynich pulled it out of a chest in a Jesuit college near Rome — and the world sat up again.

US Navy cryptographer William Friedman — the man who would later crack Japan's Purple diplomatic cipher — spent years on it. He built frequency tables, compared symbols. And near the end of his life, he didn't say "I couldn't solve it." He said something different: that in his judgment, the problem might simply be unsolvable. That's a rare admission from someone at the top of his field. Almost a century later, physicist Andreas Schinner built a mathematical model of the character sequences and reached the opposite conclusion: the patterns are too regular for a real language, but they fit an elaborately constructed fake text quite well. Friedman and Schinner, arguing across a hundred years — and neither one has won.

In twenty-eighteen, Canadian computational linguist Greg Kondrak ran the text through machine-translation algorithms and announced the language was close to medieval Hebrew, the words encoded anagrams. In twenty-nineteen, British researcher Gerard Cheshire confidently declared it a proto-Romance language — a women's health manual, compiled by nuns for the Queen of Aragon. What all these theories share is that none of them can be independently reproduced. The text acts like a mirror: hold up almost any flexible enough hypothesis, and it reflects it back.

In the margins, researchers found faint pencil marks — barely visible attempts to match the characters to Latin letters. The style suggests the work of mid-twentieth-century cryptographers, likely from the Second World War era. Correspondence preserved in Yale's library archive shows military codebreakers used the manuscript as a training exercise. So the book traveled from an alchemist to an emperor, from a book dealer to the cryptanalysts of a superpower. It's as if every generation has to rediscover, on its own, that this book is smarter than they are.

Today the manuscript lives at Yale's Beinecke Library, catalogued as MS 408. Every page has been scanned in high resolution — you can zoom in on every single stroke of the pen. And still: silence.

There's something almost unbearable in the fact that the most mysterious book on earth was written in a steady, calm hand. No hesitation. No corrections. As if the author knew exactly what he was saying — and knew, just as certainly, that we never would.

If you leave with more questions than you came with, that was the idea. The answers and the map live at moygid dot online.