The Mammoth Encyclopedia of Unsolved Mysteries (65 page)

We do not know how the manuscript came to be in Prague, but the likeliest possibility is that it was taken there from England by the famous Elizabethan “magician” Dr John Dee, who went there in 1584; one
writer speculates that Dee may have obtained it from the Duke of Northumberland, who had pillaged monasteries at the behest of Henry VIII. The English writer Sir Thomas Browne said later that Dee’s son Arthur had spoken about “a book containing nothing but hieroglyphics” which he had studied in Prague. Marci believed the mysterious book to be by the thirteenth-century monk and scientist Roger Bacon.

The Voynich manuscript (as it came to be known) is a baffling mystery because it
looks
so straightforward; with its drawings of plants it looks like an ordinary medieval “herbal”, a book describing how to extract healing drugs from plants. One would expect astronomical or astrological diagrams in a herbal, because the plants were often supposed to be gathered by the full moon, or when the stars or planets were in a certain position.

Kircher obviously had no success with the manuscript; he finally deposited it in the Jesuit College in Rome, whence it came into the hands of the Jesuits of Frascati.

Voynich was fairly certain that the manuscript would not remain a mystery once modern scholars had a chance to study it. So he distributed photostats to anyone who was interested. The first problem, of course, was to determine what language it was in – Latin, Middle English, perhaps even Langue d’Oc. This should have been an easy task, since the plants were labelled, albeit in some sort of code. But most of the plants proved to be imaginary. Certain constellations could be recognized among the astronomical diagrams but again, it proved impossible to translate their names out of code. Cryptanalysts tried the familiar method of looking for the most frequent symbols and equating them with the most commonly used letters of the alphabet; they had no difficulty recognizing 29 individual letters or symbols, but every attempt to translate these into a known language was a failure. What made it so infuriating was that the writing didn’t
look
like a code; it looked as if someone had sat down and written it as fluently as his mother tongue. Many scholars, cryptanalysts, linguists, astronomers, experts on Bacon, offered to help; the Vatican Library offered to throw open its archives to the researchers. Still the manuscript refused to yield up its secret – or even one of its secrets.

Then in 1921 a professor of philosophy from the University of Pennsylvania, William Romaine Newbold, announced that he had solved the code; he explained his discovery before a meeting of the American Philosophical Society in Philadelphia. What he had done, he explained, was to start by translating the symbols into Roman letters,
reducing them in the process from 29 to 17. Using the Latin
conmuto
(or
commuto
: to change) as a key word, he then went on to produce no less than four more versions of the text, the last of which was (according to Newbold) a straightforward Latin text mixed up into anagrams. These merely had to be unscrambled and the result was a scientific treatise which revealed that Roger Bacon was one of the greatest intellects of all time.

This had, of course, always been suspected. It was Bacon who had inspired Columbus to seek out America by a passage in his
Opus Majus
in which he suggested that the Indies could be reached by sailing westward from Spain. In the days of alchemy and a dogmatic and muddled science derived from Aristotle, Bacon advocated learning from nature by experiment and observation, and was thrown into prison for his pains. In rejecting the authority of Aristotle he was also by implication rejecting the authority of the Church. In his
City of God
, St Augustine had warned Christians to shun science and intellectual inquiry as a danger to salvation. Roger Bacon, like his Elizabethan namesake Francis, could see that such an attitude was tantamount to intellectual suicide. Yet when all this is said, it has to be admitted that Bacon was very much a man of his time, and that the
Opus Majus
is full of statements that a modern scientist would regard as gross errors and superstitions.

But if Newbold was correct, Bacon was one of the greatest scientists before Newton. He had made a microscope and examined biological cells and spermatozoa – these were the tadpole-drawings in the margins – and had made a telescope long before Galileo; he had even recognized the Andromeda nebula as a spiral galaxy. Newbold translated a caption to what he claimed to be a sketch of the nebula: “In a concave mirror I saw a star in the form of a snail . . . between the navel of Pegasus, the girdle of Andromeda and the head of Cassiopeia”. (It is known that Bacon understood how to use a concave mirror as a burning-glass.) Newbold declared that he had no idea of what he would find by looking in the region indicated, and was surprised to find that the “snail” was the Andromeda nebula.

But in
The Codebreakers
cipher expert David Kahn has pointed out one of the basic flaws in Newbold’s system. Newbold’s method depended on “doubling up” the letters of a word, so that, for example, “oritur” became or-ri-it-tu-ur, and this text was solved with the aid of the key word “conmuto” and the addition of a q. But how would this process be carried out in reverse – in other words, when Bacon was turning his original text into a cipher? Kahn says: “Many one-way
ciphers have been devised; it is possible to put messages into cipher, but not to get them back out. Newbold’s seemed to be the only example extant of the reverse situation”.

Newbold died in 1926, only sixty years old; two years later his friend Roland G. Kent published the results of Newbold’s labours in
The Cipher of Roger Bacon
. It was widely accepted – for example, by the eminent cultural historian Étienne Gilson.

But one scholar who had been studying Newbold’s system was far from convinced. He was Dr John M. Manly, a philologist who headed the department of English at Chicago University, and who had become assistant to the great Herbert Osborne Yardley – described as the greatest codebreaker in history – when US Military Intelligence set up a cryptanalysis department in 1917. Manly had produced the definitive edition of Chaucer in eight volumes, comparing more than eighty versions of the medieval manuscript of the
Canterbury Tales
. One of his most remarkable feats was the deciphering of a letter found in the baggage of a German spy named Lothar Witzke, who was captured in Nogales, Mexico, in 1918. In three days of non-stop application Manly had solved the twelve-step official transposition cipher, with multiple horizontal shiftings of three and four letter groups finally laid out in a vertical transcription. In a military court he was able to read aloud a message from the German minister in Mexico beginning: “The bearer of this is a subject of the empire who travels as a Russian under the name of Pablo Waberski. He is a German secret agent . . .” It was the spy’s death warrant (although President Wilson commuted it to life imprisonment).

Now Manly studied Newbold’s
Cipher of Roger Bacon
, and concluded that in spite of his undoubted integrity, Newbold had been deceiving himself. The weak point of the cipher was the anagramming process. Most sentences can be anagrammed into a dozen other sentences, a method by which admirers of Francis Bacon have had no difficulty proving that he wrote the plays of Shakespeare (see chapter 51). With a sentence involving more than a hundred letters, there is simply no way of guaranteeing that some particular rearrangement provides the only solution – David Kahn points out that the words “Hail Mary, full of grace, the Lord is with thee” can be anagrammed in thousands of different ways.

Newbold had also made certain “shorthand signs” a basic part of his system of interpretation. When Manly looked at these through a powerful magnifying glass he found out that they were not “shorthand” at all, only places where the ink had peeled off the vellum. By the time he
had pointed out dozens of cases in which Newbold had allowed his interpretation to be influenced by his own twentieth-century assumptions, Manly had totally demolished Newbold’s claim to have solved “the cipher of Roger Bacon”.

Since that time, 1931, there have been many attempts to decipher the Voynich manuscript. In 1933 a cancer specialist, Dr Leonell C. Strong, published his own fragments of translation, and proved to his own satisfaction that the work was a herbal by an English scholar, Anthony Ascham; he even published a recipe for a contraceptive which apparently works. But Strong failed to explain the method by which he arrived at his translations, so they have never achieved wide acceptance.

William F. Friedman, who organized a whole group of specialists to work on the problem in the last year of the Second World War, was frustrated by the end of the war and the disbandment of his group. But Friedman pointed out that the Voynich manuscript differs from other codes in one basic respect. The inventor of a code attempts to frustrate would-be cryptanalysts by trying to remove repetitions that would give him away (for example, a repeated group of three letters would almost certainly be “and” or “the”). The Voynich manuscript actually has far more repetitions than an ordinary text. This led Friedman to hypothesize that the text is in some artificial language which, because of a need for simplicity, would inevitably have more repetitions than a highly complex “natural” language. But this presupposes that Roger Bacon (or whoever wrote the manuscript) was so anxious to conceal his meaning that he went to far greater lengths than even a code-expert would consider reasonable. And for a thirteenth-century monk, who had little reason to fear code-breakers, this seems unlikely . . .

And this, of course, is the very heart of the mystery. We do not know when the manuscript was written, or by whom, or in what language, but even if we knew the answers to these questions it is difficult to think of any good reason for inventing such a baffling code. The earliest ciphers in the Vatican archive date from 1326 (when Roger Bacon was a boy) and these are merely “coded” names relating to the struggle between Ghibellines and Guelphs. These were respectively supporters of the Holy Roman Emperor and the pope; the Ghibellines are called Egyptians and the Guelphs Children of Israel. (It is easy to guess what side the inventor of the code was on.) The earliest Western “substitution” cipher dates from 1401. The first treatise on codes, the
Polygraphia
of Johannes Trithemius, was not printed until 1518, two years after the death of its author. So it is hard to imagine why Roger Bacon or anyone within a century of his death should have gone to so much trouble to
invent a code of such apparent sophistication when something much simpler would have sufficed.

Kahn offers one clue to why the author of a herbal (which is what the Voynich manuscript looks most like) should want to conceal his meaning when he speaks of one of the earliest encipherments, a tiny cuneiform tablet dating from about 1500
BC
. “It contains the earliest known formula for the making of glazes for pottery. The scribe, jealously guarding his professional secret, used cuneiform signs . . . in their least common values”. The author of the Voynich manuscript may have been a highly skilled professional herbalist who wrote down his secrets for his own use and those of his pupils, and was determined to keep them out of the hands of rivals.

This view would have struck the antiquarian bookseller Hans Kraus as altogether too commonplace. When Ethel Voynich died at the age of ninety-six, in 1960, Kraus purchased the manuscript from her executors and put it up for sale at $160,000; he explained that he thought that it could contain information that might provide new insights into the record of man, and that if it could be deciphered it might be worth a million dollars. No one took it at that price, and Kraus finally gave it to Yale University in 1969, where it now lies, awaiting the inspiration of some master-cryptographer.

38

Joan Norkot

The Case of the Bleeding Corpse

When Sir John Mainard, known as “a gentleman of great note and judgment of the law”, died in 1690, there was found among his papers an account of an incredible case in which a woman’s corpse accused her murderers. The whole document was printed in
The Gentleman’s Magazine
of July 1851, a copy having been taken by a lawyer named Hunt. The events it describes are so extraordinary that the magazine took the precaution of heading it: “Singular Instance of Superstition”. Yet Mainard’s account makes it clear that the event was witnessed by a crowd of people standing in a graveyard, including two clergymen.

The trial in which the evidence was heard took place at Hereford Assizes in Herefordshire in “the fourth year of the reign of King Charles I” (1629), more than sixty years before Mainard’s death. Mainard’s account is printed in full in the Reverend Montague Summers’s
The Vampire in Europe
(1929). An account written by Valentine Dyall in
Unsolved Mysteries
(1954)
¹⁶
is also based on the court records, which it admits are “tantalizingly scanty”. The account has an irritating lack of dates, and even of names, but the general outline is clear enough.

There had been, it seems, some trouble between Arthur Norkot and his wife, Joan, one suggestion being that he suspected her of infidelity. It may have been simply overcrowding that caused their disagreements, for Joan lived in a two-room cottage with her husband and baby, as well as with her mother-in-law, Mary Norkot, and her sister and brother-in-law, Agnes and John Okeman.

One morning Joan was found dead in her bed, with her throat cut; the baby was lying – unharmed – beside her. A bloodstained knife was sticking in the bare floorboards.