June 25, 2013 at 12:58 PM ET
By Rossella Lorenzi
The Voynich Manuscript, an enigmatic book that has frustrated codebreakers and linguists for a century, contains a genuine message, according to a new computer analysis.
The study analyzed the unintelligible scripts that fill the about 250-page-long manuscript and extracted clusters of “keywords” which could serve as a good starting point in cryptographic attempts.
Described as “the world’s most mysterious manuscript,” the book takes its name from the rare-book dealer Wilfrid Voynich, who discovered it in 1912 in the Villa Mondragone near Rome. He claimed the book had belonged to the 16th-century Habsburg emperor Rudolf II.
Radio carbon dating established the manuscript was penned on 15th-century parchment pages.
The book’s estimated 250,000 characters are totally alien and make “The Da Vinci Code” pale by comparison: arranged in groups like words and sentences, some resemble Latin letters and Roman numerals; others are unlike any known language.
Moreover, the puzzling handwriting is surrounded by intricately drawn illustrations: plants that can’t be identified, astrological symbols, elaborate networks of pipework and naked ladies dancing or bathing in a strange green liquid.
Modern scholars have thematically divided the manuscript into five sections: Herbal, Astrological, Biological, Pharmacological and Recipes.
“In spite of its unmistakable medieval-codex look, the origin, purpose and contents of the Voynich manuscript remain a deep mystery,” Marcelo Montemurro, a theoretical physicist from the University of Manchester, UK, and Damian Zanette, of the Statistical Physics Group at Centro Atómico Bariloche, Río Negro, Argentina, wrote in the journal Plos One.
“Since the seventeenth century, numerous attempts at deciphering the script have led to a few claims of success, but none of them has been convincing,” he added.
Speculations ranged from the manuscript being the secret work of a religious sect, the only remaining document from a forgotten language, an unbreakable secret code, and the recipe for the “elixir of life.”
Montemurro used a computerized statistical method to analyze the text at a large scale. Focusing on patterns of how the words were arranged, the study extracted “clusters” of cryptic words such as shedy, cthy, chor, qotedy and qokeey.
“The identification of these words is crucial for our analysis,” Montemurro told Discovery News.
The key idea behind the extraction of the keywords relies on the non-homogeneous use of content-bearing words in any text, with certain words being used with a higher-than-average frequency when a specific topic is discussed.
“At the statistical level this means that these words that define the topics in the text end up being used in a sort of clustered pattern. On the contrary, words that are not related to any specific topic, like for instance function words like ‘or’, ‘and’, ‘an’, have a much more uniform rate of use,” Montemurro said.
According to the researcher, if the words in the Voynich text are real encoded words, then this method could provide clues on which are the keywords that are more related to the book’s topics.
He noted that the manuscript presents a complex organization in the distribution of words that is compatible with those found in real language sequences.
The analysis also unveiled a relationship between the sections of the text that is consistent with similarities between the images.
“For instance, at the level of language, the Herbal and Pharmaceutical sections of the text are the most closely linked. Interestingly, these two sections are the ones sharing the lavish plant illustrations of the book,” Montemurro said.
Several experts dismissed the manuscript as a deliberate hoax, possibly forged by John Dee, an English mathematician and astrologer at Rudolph’s court.
In 2003 computer scientist Gordon Rugg demonstrated that text resembling that in the book could be generated with a Cardan grille, an encryption device invented around 1550.
But according to Montemurro none of the proposed hoaxing methods has so far reproduced all the level of statistical structure of the Voynich text.
“The grille and table method proposed by Gordon Rugg as the one used to produce the hoax was introduced around 1550. However, carbon dating of the Voynich manuscript’s vellum indicates a date between 1404 and 1438,” said Montemurro.
“This fact seriously weakens even further the hoax hypothesis since it would imply that the expensive complete vellum was kept unused for more than a century, until the grille and table hoaxing method became available,” he concluded.