Folly follows the script

Why all the nonsense over an undeciphered text? asks Jacques Guy.

This is a light book about a serious subject: a copiously illustrated medieval manuscript written on vellum in an unknown script and an unknown language, which has resisted all attempts at decipherment since its discovery in 1912 by Wilfrid Voynich, a dealer in rare books and manuscripts, who eventually deposited it at Yale University. Mary d'Imperio called it "an elegant enigma" in her 1978 monograph. It certainly is an enigma that cries out for an elegant solution, but so far not even one half-sensible decipherment has been offered.

Instead, the Voynich Manuscript has become a monument to human folly and self-delusion. Much of what has been written on the subject would have been laughed out of court under normal circumstances. Why it has not been is another enigma. If, after 90 years of concerted efforts, we are still unable to interpret this manuscript - which, judging from the artwork and penmanship, probably originated in north Italy around 1450 - what chance do we stand of making sense of putative alien transmissions from light years away, the object of the Seti project? We cannot even tell whether the Voynich Manuscript contains meaningful text or mere gibberish, written glossolalia. All this should have provided the matter for a serious book, but the authors appear to have chosen to dazzle rather than educate.

There is, for instance, the matter of the "Chinese hypothesis". It was a hoax of mine, concocted in 1991. My point was to demonstrate how the absurd can be dressed in sensible garb. I suggested that Marco Polo brought back two Chinese scholars, who, once in Venice, set about transcribing their knowledge in an encyclopedia - the manuscript discovered by Voynich - in a south Chinese dialect, which they wrote in a Roman-looking script, elaborated ad hoc . Absurd though this seems, the arguments in its favour appear solid. Judging from its illustrations, the Voynich Manuscript consists of a large herbal, containing herbal medicine recipes, as well as anatomical diagrams and an astronomical or astrological section, while the "letters" of its "language" exhibit distributional properties strongly reminiscent of Chinese when written alphabetically.

Gerry Kennedy, one of the authors, was much taken by my hoax - so much so that it was about all that survived of our 25-minute telephone conversation in the run-up to his BBC Radio 4 programme on the manuscript. The conversation was all leading questions and play-acting. Kennedy was to ask me about the Chinese hypothesis; I was to tell him it was a hoax; he was to feign surprise; I was to confirm the hoax. You will find my confession in this book, quoted in full with my permission. Since I had said it, and written it, I could not honestly retract it.

What you will not find, however, is what followed the Chinese canard. In 1997, a new member of our Voynich interest group, Jorge Stolfi, a professor of computer science, came up with statistical evidence that the structure of "Voynichese" was very similar to that of Mandarin Chinese. Highly sceptical, I set about finding flaws in his data or his reasoning. I could fault neither, and I wrote as much, just as emphatically as I had confessed to my Chinese hoax. There was then a lively exchange lasting some two years, with evidence mounting for the Chinese-like structure of Voynichese.

These exchanges are easily accessed now that Google has indexed the huge Voynich archives, but they are reduced in this book to a longish paragraph suggesting that the "Chinese theory" is safely dead and buried. "According to Stolfi's theory, the manuscript was written in Chinese (or another East Asian language)I To date the theory rests undisturbed in the sward."

This is immediately followed by another theory, by Beatrice Gwynn, who claimed that the writing is merely mirror-writing, like Leonardo da Vinci's notebooks, and that the language is Middle High German. That is preposterous, and any serious book on the manuscript should loudly say so, and why. But here none of Stolfi's observations is explained in detail, nor is Gwynn's decipherment. By listing one theory immediately after the other, the book serves to paint the former in the same potty colours as the latter.

Generally speaking, the authors seem quite out of their depth, or else they do not care to be cautious or plausible. They spend some 20 pages on the tenuous link between the Voynich Manuscript and John Dee, astrologer to Queen Elizabeth I, even mentioning the infamous occultist Aleister Crowley in relation to Enochian, the "Angelic language" fabricated by Edward Kelly, Dee's medium and a complete fraud. But they make no reference to the one authoritative work on Enochian, Donald Laycock's The Complete Enochian Dictionary , which is far from a mere dictionary. Dee is regularly mentioned in relation to the Voynich Manuscript because he happened to be in Prague at about the same time as the manuscript might have been - as part of some scholar's library, or perhaps in the library of Emperor Rudolph II himself.

On this basis, Pontius Pilatus might as well be considered a possible author of the Gospels because he happened to be in the right place at just about the right time.

This illustrates another reason why the Voynich Manuscript should be treated seriously - for the way it brings out the folly in human beings and stifles their reason. Consider the story of its first hopeful decipherer, William Romaine Newbold, a respected professor of philosophy at the University of Pennsylvania. Voynich had found a letter between the pages of the manuscript, dated 1665, from the rector of the University of Prague to the Jesuit polymath Athanasius Kircher, which said in substance: "My excellent friend Athanasius, I send you this book which only you can decipher. It was bought by Emperor Rudolph for 600 gold ducats. It is believed to be the work of Roger Bacon."

This started Newbold down the Bacon track. The rest is a saga of mind-boggling self-delusion. Using a magnifying glass, Newbold noticed that the individual letters of the manuscript were formed of tiny signs resembling Greek or Roman shorthand. It never occurred to him that these tiny signs were just cracks in the ink - or perhaps it did, but he was so intent on his discovery that he shut his mind to the obvious. Even those cracks were mostly self-delusions. Plate XIIIB in Newbold's posthumously published The Cipher of Roger Bacon shows a photographic enlargement of a five-letter Voynichese word that looks a bit like "8oa29", under each letter of which is Newbold's breakdown into shorthand signs. Opposite is plate XIIIC, a larger, unaltered enlargement of the same section. None of the supposed shorthand strokes is identifiable in it.

Newbold's delusion did not stop at this: he reasoned that Bacon could have written those tiny strokes only under a microscope. Therefore, Bacon had invented the microscope. Any serious book on the Voynich Manuscript should of course mention Newbold's folly. But it should dismiss it forthwith for what it is, and then only, perhaps, delve deeper into it. Newbold's folly is mentioned in this book, of course, but no clear stance is taken, so careless readers may end up thinking that there is something to it. The only clear statement of Newbold's delusion appears in a short paragraph that could easily misinterpreted as a mere opinion. Worse: there is no proper reference. A footnote refers you to " Speculum , p.350". Speculum is the journal of the Medieval Academy of America, which has been in existence since 1926. Guessing at the name of the author is not so difficult: he is probably John M. Manly, mentioned in the previous paragraph. Guessing at the year of the journal is another matter. Putting the question to Google will find you the likely reference: "J. M. Manly, Roger Bacon and the Voynich Manuscript, in: Speculum 6, 1931, p.345." There are many more instances of such useless bibliographical entries, which can be hunted down only by querying search engines.

So just what is the utility of this book? It has pitifully few photographs of the manuscript. Search the web, and you will find many medium to high-quality images, in colour as well as black-and-white - almost the entire manuscript, in fact. Search for "Voynich" together with the names of Jim Gillogly, Gabriel Landini, Jim Reeds, Dennis Stallings, Jorge Stolfi or Rene Zandbergen and see for yourself. They write serious, solid stuff, with a wealth of proper references, the complete transliteration of the manuscript, and software tools to analyse it. There is nothing of the kind in this book, not even a single reference as to how one can tell vowels from consonants in a text written in an unknown alphabet, which was the object of several articles in Cryptologia .

If it were ever to be deciphered, the Voynich Manuscript would likely prove to be a dreadful disappointment. Its interest lies not in its contents, but in the methodology of its decipherment. Without a guiding methodology - which we lack and which we need to develop - its study leads only to follies worthy of Newbold's.

As recently as two months ago, in its July issue, Scientific American published an article proposing that the manuscript is meaningless gibberish, a hoax. The proof? The author, Gordon Rugg of Keele University, claims he has found a way, "using a simple coding tool that was available in the 16th century", of generating random text with the same statistical properties as Voynichese; and therefore the manuscript is random gibberish.

How can the fallacy have escaped him and Scientific American ? Stolfi showed our group how anyone can fluently generate Chinese-sounding gibberish. Does that make Chinese literature nonsense? Rugg concludes that his approach may even be applied to the problem of Alzheimer's disease. "Then a medieval manuscript that looks like an alchemist's handbook may actually prove to be a boon to modern medicine." The spirit of Newbold lives on.

Jacques B. M. Guy is a computer scientist with a long-standing interest in the Voynich Manuscript. He holds a PhD in linguistics from the Australian National University.