Cracking the Voynich Code

Gordon Rugg’s Search Visualizer analysis of the manuscript text, showing the banded distribution of syllables. (Courtesy Gordon Rugg)

Between our first and second conversations, Rugg put some common Voynichese syllables into the Search Visualizer program. The results were staggering. While natural languages have an even distribution of common suffixes and prefixes, common prefixes and suffixes of the Voynich Manuscript are clustered in different parts of the text. But perhaps even more astonishing is the distribution of four parallel charts with four different syllables: The clusters all align. The banding of syllables—where they change frequency quite radically—all occur at the same place in the text. While this is radically inconsistent with natural languages, it is quite consistent with the table and grille method of producing words, in which, to create the illusion of a language, the writer would turn the grille on its axis to start creating a new frequency. Imagine trying to fool people into believing that you lived in a certain neighborhood. You might come into work every day bearing coffee from a café in that neighborhood. But they might catch on. So one day, after a month or so, you come in with lunch from an eatery in that neighborhood. You do that for a while, and then start purchasing books from a local bookstore. In the same manner, turning the grille gives a new set of prefixes and suffixes, so that the illusion is maintained, and words aren’t repeated too much.

“After a few weeks, we found something no one else had seen,” Rugg says. His book, The Blind Spot, is due out in May, and has a chapter on the Voynich.

***

In 2000, a second letter mentioning a mysterious book in code and addressed to Sphinx master Kircher was found. René Zandbergen, an engineer by trade with a website about the Voynich, discovered it in Kircher’s letters. Georg Baresch, an antiques dealer, wrote to Father Kircher in 1639 (for the second time), asking him to take an interest in a mysterious manuscript that he couldn’t decipher. He hoped that Kircher, who “burns with a publication of things which are good, will not disdain from revealing also those things which are good in his books, buried in unknown characters.” Like Marci, Baresch seemed to think Kircher alone capable of deciphering the text, “given that here there is nobody capable of lifting such a weight, which consists of such obscure material that it requires a special genius.” The letter suggests, “from the pictures of herbs, of which the number in the Codex is enormous, of various images, of stars and of other things which appear like chemical secrets, I conjecture that it is all of medical nature.” This led Zandbergen to conclude that the manuscript described was the Voynich. Because “all details he mentions (unknown writing, herbs, stars) fit as well, there can be no doubt at all,” he wrote in an email.

Voynich’s contention that the book was written by Roger Bacon came from Marci’s letter, which was inside the manuscript when Voynich presented it for the first time at a Chicago Art Institute exhibition in 1915. The letter mentions that “Dr. Raphael, tutor in the Bohemian language to Ferdinand III, then King of Bohemia, told me the said book had belonged to the Emperor Rudolph and that he presented the bearer who brought him the book 600 ducats. He believed the author was Roger Bacon, the Englishman.” A signature on one of the pages also seems to suggest that the manuscript was once in or around the court of Rudolf II. The name “Jacobus de Tepenec” appears on the first page. Tepenec was a pharmacist of Rudolf’s.

Born in 1552, King Rudolf II of Bohemia was prone to bouts of melancholia, which led him to consort with doctors of the occult, such as Edward Kelly, a known alchemist and spiritualist, and John Dee, a consultant in Queen Elizabeth’s court on all things mathematical, astronomical, and alchemical. The two worked closely to communicate with angels, Kelly transcribing whole books in the Enochian language with which they spoke to him. The relationship lasted until one day, while consulting with the spirits, “Kelly pretended to be shocked at their language, and refused to tell Dee what they had said,” according to Charles Mackay’s 1848 Memoirs of Extraordinary Popular Delusions and the Madness of Crowds. Upon Dee’s insistence, Kelly told him that, according to the angels, the two men “were henceforth to have their wives in common. Dee, a little startled, inquired whether the spirits might not mean that they were to live in common harmony and good-will? Kelly with apparent reluctance, said the spirits insisted upon the literal interpretation.” It marked the end of the friendship.

Interestingly, the parchment, when radiocarbon dated, revealed what seemed to be a different conclusion than the one suggested by the 16th-century paper trail. In 2011, physicist Gregory Hodgins of the University of Arizona sampled four of the Voynich Manuscript’s pages: the page with Tepenec’s signature, one of the foldouts, and two pages bearing the two handwritings noticed by experts. Because the unstable form of carbon, or C14, decays at a known rate from the day that an animal or plant dies, its measurement can yield a time frame of death, Hodgins explained to me patiently on the phone. This time frame is then compared to a database assembled of known C14 measurements gathered from trees, whose rings correspond to years. “Radiocarbon dating is not accurate, but it is precise,” Hodgins explained. “There is a true value to what we are measuring, even if we don’t know what our target is before we begin the process.”

What he found when sampling the Voynich Manuscript was even more precise than usual, due to a fortuitous accident of nature. C14 levels are contingent upon external factors as well, such as cosmic rays and changes in atmosphere. In the 16th century, for example, C14 had stabilized relatively, making it harder to radiocarbon date things within less than a 100-year time frame. The change in C14 was simply too slow. But the animal upon whose skin the Voynich Manuscript is written died in a century during which the rate of decay enabled a very precise window. Hodgins estimates with 95 percent certainty that the animal died between 1404 and 1438.

This date, roughly 150 years before that suggested by the Tepenec signature, has led many experts to conclude that the manuscript must have been written in the 15th century. “It’s just logic,” said Paula Zyats, assistant chief conservator of the Yale Library. “Velum was too expensive to leave untouched. It did not get wasted; the opposite—it was used over and over. Nobody lost a big chunk of parchment.” Zandbergen too thinks that the radiocarbon date provides ample evidence for an early-1400s date. In Hodgins’ experience, forgeries tend to get different results on different pages, whereas with the Voynich Manuscript, all four pages overlapped in a 34-year period. “It’s possible that they came from different skins, but the four samples are very closely tied together,” Hodgins said. “Why would someone buy two-hundred-year-old paper?” Stolfi asked me. “That would be equally mysterious.”

Continue reading: This wonderful crazy man

Batya Ungar-Sargon is a Brooklyn-based freelance writer. Her work has appeared in The Forward and City Limits, among other publications. Her Twitter feed is @bungarsargon.