Labyrinths of Reason (32 page)

Some symbols (those labeled
A, I, L, M, N
, and
O)
are similar to lowercase script versions of the corresponding Roman letters. According to Bennett, other symbols look like letters of the Cyrillic, Glagolitic (old Bulgarian), and Ethiopian alphabets. The symbol labeled
Y
looks Chinese.

To add to the confusion, folio 17 contains a tiny note in Middle High German, not necessarily by the original author, talking about the Herbal of Matthiolaus. Some astrology charts in the manuscript have the months labeled in Spanish. What appears to be a cipher table on the first page has long since faded into illegibility.

About 40 pages of the manuscript are missing. Originally it contained 17 quires of 16 pages each. The last pages of the book contain pictures of stars in the margins, but no writing. That hints that the pictures were done first and the writing filled in later. In that case the pictures may be merely decorative. Many have nevertheless tried to read meaning into the illustrations. Some have guessed that the number of stars, women, or flowers on a page encodes something. The pictures of plants could mean the text discusses medicinal or magical uses of herbs, maybe an elixir of life or some such thing. Botanists have met with mixed success in identifying the plants depicted. The picture on folio 93 may or may not be a sunflower. A fruit on folio 101 suggests a capsicum pepper. Both are American plants unknown in Europe until Columbus returned in 1493. That was a couple hundred years after Roger Bacon’s death.

All sorts of exotic scenarios have been suggested. The manuscript
is in a lost dead language; it purposely avoids the most common letters of the source language to frustrate decipherers; it is a meaningless forgery (by John Dee? the Jesuits? Voynich?) created for monetary gain; it is the work of a medieval James Joyce, who invented his own words; it is the furious ravings of a long-forgotten madman. The Voynich manuscript is reminiscent of (and conceivably inspired) Borges’s short story “Tlön, Uqbar, Orbis Tertius.” In the story an eccentric millionaire financed a conspiracy of scholars to write an encyclopedia of the imaginary world “Tlön.” The first drafts were in English, but the plan was to translate the encyclopedia into Tlön’s (equally imaginary) language and calligraphy, creating a wholly inscrutable work.

The Voynich cipher has become a sword in the stone to cryptographers. Many of the most talented military code breakers of this century have tried to decipher it as a show of prowess. Herbert Yardley, the American code expert who solved the German cipher in World War I and who cracked a Japanese diplomatic cipher without knowing the Japanese language, failed with the Voynich manuscript. So did John Manly, who unscrambled the Waberski cipher, and William Friedman, who defeated the Japanese “purple code” of the 1940s. Computers have been drafted into the effort in recent years, to no avail.

The fact that computers have failed to decipher the Voynich manuscript may surprise some. As a practical matter, the task of cracking a cipher is mainly one of finding “weak spots.” Just as a diamond is cut along dislocations in the crystal, ciphers are broken by exploiting telltale regularities. The Voynich manuscript appears to be an intractable cipher, a string of symbols purged of all the usual statistical earmarks of language. Attempts to solve it have been as useless as trying to drive a chisel through the geometrically perfect part of a diamond crystal.

Unless it is a fabrication, (and we will see later that it is almost certainly not), the text of the Voynich manuscript meant
something
to its author. It meant that something in part because of what the author was thinking at the time he or she wrote it. But is the meaning also inherent in the pattern of symbols? Or in a lost key to the cipher? Or in a combination of all of these? The possibility of our deciphering it depends on the meaning being “contained” in the pattern of symbols as well as in the now unknowable mental processes of the author.

It is doubtful that Bacon or any other medieval author single-handedly created a cipher more secure than the scores of military
codes of later centuries that
have
been cracked. Some see this as evidence that the Voynich manuscript is nonsense. A set of symbols doesn’t
have
to mean anything. Is there any way of telling if a collection of symbols contains a message? This question is one of the most difficult ones in the study of knowledge.

Imagine someone in a far-future age digging up a time capsule containing a newspaper from our era. By then, English is a forgotten language; even the Latin alphabet is unknown. One archaeologist looks at the newspaper and decides it must be a form of writing. He hopes to decipher it and learn about the lives of the people who buried the time capsule. A second archaeologist says, “Don’t waste your time! That’s wallpaper! People glued it to the walls of their houses. Those little black squiggles are a decorative design that was popular back then.”

You might think the first archaeologist would have an easy time demonstrating that the newsprint is writing and not a wallpaper design. There would be regularities in the newsprint—common letters, common words, periods at the ends of sentences—that would mark it as writing. The trouble is, there are regularities in decorative motifs too. It is hard to say offhand just how the regularities in an unknown design
necessarily
differ from those in an unknown script. The more alien the script or decorative art, the less confident one could be of any such determination.

Nor could the archaeologist necessarily expect to decipher the newspaper and prove his case that way. Egyptian hieroglyphic writing was never “deciphered” on the strength of internal clues. Only the serendipitous Rosetta stone revealed it to the modern world.

There is a bittersweet appeal to the puzzle of the Voynich manuscript. It’s not just the prospect of discovering a medieval diary, a magic text, or a forbidden book of erotica. By its very inscrutability, the Voynich manuscript is a comment on the frailty of knowledge.

A pair of Bacons pioneered the scientific method: Franciscan monk Roger Bacon of the thirteenth century and Elizabethan statesman Sir Francis Bacon of three centuries later (1561–1626). Roger Bacon is by far the more mysterious of the two. Little is known of his life outside of what may be conjectured from his writings. We know he was an educator who lectured at Oxford and Paris. At some point in his career, he joined the Franciscan order and took a vow of poverty.

About 1247 Bacon grew dissatisfied with the faith his contemporaries had in Aristotelian science. He felt that direct observation and experiment was superior to reliance on established authorities. He credited this emphasis on experiment to Durand de Saint-Pourçain, a French Dominican philosopher of whom little else is known. In 1267 Bacon reported having spent more than 2000 Parisian pounds over the years on experiments and “secret books.” From one of these rare books he learned the formula for gunpowder. He described the preparation of explosives in cryptic form.

Bacon’s originality created friction between him and the Franciscan hierarchy. Fortunately, Bacon had befriended the man who became Pope Clement IV. Clement, on hearing of Bacon’s ideas for a philosophical encyclopedia, ordered Bacon to send him a copy. The Pope thought that the work already existed. Actually, it was just an idea that Bacon had partially sketched in letters to friends. Rather than explain, Bacon set to work. He concealed the project from his Franciscan brothers, working without copyists. A year and a half later he had a trilogy:
Opus Majus, Opus Minus
, and
Opus Tertium
.

These works made Bacon notorious for speculations on future technology. He described a telescope (but did not have a practical model). He envisioned automobiles and, less accurately, airplanes. Bacon was thinking of human-powered flight, in which human arms flapped artificial wings. He also concluded that balloons might be made to float by filling them with gases lighter than air.

Bacon believed the earth was round.
Opus Majus
described a sea voyage west from Spain to India. Cardinal Pierre d’Ailly plagiarized this passage in his
Imago Mundi
(published 1480), where Columbus read it and cited it in a letter to Ferdinand and Isabella of Spain.

Eventually Bacon’s reputation as a miracle worker overwhelmed him. The Franciscans imprisoned him around 1278 for “suspected novelties.” The story about Bacon’s enemies destroying his books after his death is apparently false. As far as we know, all his major works survive.

The Voynich manuscript has driven some if not to madness then to extraordinary self-delusion. More than one person has gone to his grave
thinking
he decoded the Voynich manuscript.

In 1921 Professor William Romaine Newbold of the University of
Pennsylvania announced that he had deciphered the Voynich manuscript and would reveal his findings at a meeting of the American Philosophical Society. Like many, Newbold attributed the manuscript to Roger Bacon. He thought it proved that Bacon had built both a microscope and a telescope centuries before Galileo and van Leeuwenhoek. The illustration on folio 68, Newbold thought, was the spiral nebula in Andromeda as seen through Bacon’s secret telescope. Newbold even reported that the telescope’s mirrors had cost Bacon the equivalent of $1500. Other illustrations showed spermatozoa and ova. Newbold’s revelations briefly excited the press and public. One woman was so sure that Newbold had uncovered Bacon’s black incantations that she traveled hundreds of miles to ask Newbold to cast out the demons that had possessed her.

It is now sadly clear that Newbold himself was possessed. He was reluctant at first to give away too much of his discoveries. The more he made public, the more it became obvious that he was reading his own hunches into a still uncracked cipher. Newbold had Bacon observing the spiral structure of the Andromeda nebula with a reflecting telescope. Astronomers pointed out that the nebula’s spiral structure isn’t visible in
any
telescope—only in time-exposure photographs. Even Newbold didn’t have Bacon inventing the camera. From the earth, the Andromeda nebula is seen almost edge on. Whatever is being depicted in folio 68 is face on, so that its outline is a circle.

The cipher Newbold attributed to the manuscript was a masterwork of wishful thinking. He found a barely legible “key” on the last page of the manuscript. (More than one scholar has supposed that this inscription may be a key. Some think the handwriting of the “key” is different and that this inscription was added later by someone other than the author.) Newbold claimed that the symbols translated to the Latin
A mihi dabas multos portas
(“Thou wast giving me many gates”). This he took to mean that more than one cipher was used.

According to Newbold, Roger Bacon encoded an original Latin text with a “biliteral” cipher. In a biliteral cipher, a pair of letters in the visible writing encodes one letter of the message. This was ingenious by the standards of thirteenth-century cryptography, and should have ensured the secrecy of whatever the author was writing.

But that, insisted Newbold, was just one of a series of cryptographic Chinese boxes. In a normal biliteral cipher, the encoded message (“ciphertext”) is twice as long as the original message
(“plaintext”). To make the ciphertext more concise, Newbold thought that Bacon chose letter pairs so that the last letter of one pair was always the same as the first letter of the next pair. If Bacon wanted to encode the Latin word
unius
, he might have
or
representing
u
, and
ri
meaning
n
, and so on:

Then he would strike out the repeated letters and get
oritur
. To make the cipher yet more baffling, more than one letter pair could encode a given letter, and letters that sounded similar—such as
b, f, p
, or
ph
—could be encoded by the same letter pair.

Confused? So was Newbold’s audience. Those cryptographers who were still following what Newbold was saying realized that such a cipher would be hopelessly impractical.