by DAVID KAHN
Then he rubbed one part of it with oile of nuts, to see if it were not written with the lee of a fig-tree: and another part of it with the milk of a woman giving suck to her eldest daughter, to see if it was written with the blood of red toads, or green earth-frogs: Afterwards he rubbed one corner with the ashes of a Swallowes nest, to see if it were not written with the dew that is found within the herb Alcakengie, called the winter-cherry. He rubbed after that one end with eare-wax, to see if it were not written with the gall of a Raven: then did he dip it into vinegar, to try if it was not written with the juice of the garden Spurge: After that he greased it with the fat of a bat or flitter-mouse, to see if it was not written with the sperm of a whale, which some call ambergris: Then put it very fairly into a basin full of fresh water, and forthwith took it out, to see whether it were written with stone-allum: But after all experiments, when he perceived that he could finde out nothing, he called the messenger, and asked him, Good fellow, the lady that sent thee hither, did she not give thee a staffe to bring with thee? thinking that it had been according to the conceit whereof Aulus Gellius maketh mention, and the messenger answered him, No, Sir. Then Panurge would have caused his head to be shaven, to see whether the Lady had written upon his bald pate, with the hard lie whereof sope is made, that which she meant; but perceiving that his hair was very long, he forbore, considering that it could not have grown to so great a length in so short a time.
Then he said to Pantagruel, Master, by the vertue of G—I cannot tell what to do nor say in it; for to know whether there be any thing written upon this or no, I have made use of a good part of that which Master Francisco di Nianto the Tuscan sets down, who hath written the manner of reading letters that do not appear; that which Zoroastes published, Peri grammaton acriton; and Calphurnius Bassus, de literis illegibilibus: but I can see nothing, nor do I beleeve that there is any thing else in it then the Ring: let us, therefore, look upon it.
And inside the ring, of course, is the message of reprobation from the lady. Half of this lusty episode—was ever a discussion of invisible inks rendered with more glee?—turns on real things, and illustrates Rabelais’ vast erudition, while the half that makes it an outrageous burlesque is purely Rabelais’ invention. The inks made with ammonium chloride, onions, alum, and tithymallus will all work, while the red toads, the ear wax, and the bat fat lampoon the exotic formulas and the unnecessary mystery with which the would-be magicians of the day, rather like Trithemius and, in a different way, like Porta, loved to surround themselves. The staff, of course, is a skytale, which the Roman writer Aulus Gellius describes in his Attic Nights. The head-shaving story kids Herodotus’ tale of Histaeius and his tattooing of a secret message on a slave’s head. The three books are purely imaginary. And the climax of the episode—with the message just where they should have looked to begin with—satirizes the whole business of unearthing secret messages. Here, in the volcanic writings of a titan, cryptology explodes into incandescent literary life.
The greatest writer of all, William Shakespeare, never mentioned cryptology as such, but did touch upon its older brother, interception. In Henry V, Henrys brother, the Duke of Bedford, is discussing with two other peers a conspiracy that three other lords were hatching against the king. “The King hath note of all that they intend, /By interception which they dream not of,” Bedford says. Not long thereafter Henry presents the three traitors with proofs of their guilt—probably the intercepted letters themselves. The effect is dramatic: “Why, how now, Gentlemen?/What see you in those papers that you lose/So much complexion?” Henry exclaims. “Look ye, how they change! /Their cheeks are paper. Why, what read you there/That hath so cowarded and chas’d your blood/Out of appearance?” They immediately confess, and, in the play, are promptly executed.
Though this was a historical play, Shakespeare’s sources—the chronicles of Raphael Holinshed and of Edward Hall—do not mention interception. This aspect must therefore have sprung from Shakespeare’s imagination, which may have drawn its inspiration from the knowledge, probably fairly common then, that letters were intercepted and opened by the authorities to obtain information.
It perhaps testifies to Shakespeare’s scope that discussions of cryptology in works of the imagination in the 200 years that succeeded his appear nonexistent, or at least were so rare and tangential to the main theme of a work that they left no traces in the critical literature. Then, in 1829, literary cryptology took a step sideways when Honoré de Balzac published The Physiology of Marriage, one of the works that make up his immense Human Comedy. It is a long, amusing, sardonic dissertation on marriage, and in a section on “Religion and confession, considered in their relation with marriage,” Balzac wrote: “La Bruyère has said very spiritually: ‘it is too much against a husband [to require both] devotion and gallantry: a wife must choose.’ The author thinks that La Bruyère is mistaken. In reality,”—and there followed four pages of the most confused typographical mishmash, with letters upside down, letters sideways, type standing on its head and printing on its feet, none of it making the least bit of sense, but with a final “en effet” slyly inserted near the end. In the errata, which serve “to caution you against the errors you have made in reading this work,” Balzac cited the four pages and observed: “To really understand the sense of these pages, a reader who is honest with himself ought to reread its principal passages several times; for the author has placed all his thought there.”
Could it be a real cryptogram? Commandant Bazeries wondered, and, taking a later edition, analyzed the cryptic text, found the cryptogram to be neither a transposition nor a substitution, and concluded that it was no cipher at all. He could have determined this more easily simply by comparing his edition with the first, or with any one of the successive editions—some of which appeared during Balzac’s lifetime. He would have seen that Balzac, evidently not wishing to reveal his feelings on religion and confession, had ordered his printer to set pi, for the “cryptogram” changed from one edition to the next!
Balzac’s fake cryptogram in The Physiology of Marriage, showing differences in different editions, beginning with the first edition in 1840 (upper left), a “new” 1840 edition (upper right), an 1847 edition (lower left), and an English 1901 edition (lower right)
A few years later, literary cryptology took its greatest step forward with the work of Edgar Allan Poe, whose story “The Gold-Bug” remains unequalled as a work of fiction turning upon a secret message.
That the early American writer should have become interested in cryptology seems almost inevitable. He urged exactness in thinking, talked about “ratiocination,” and wrote stories, like “The Purloined Letter,” that demanded a methodical logic. But he also wrote poems of an unearthly beauty and the macabre Tales of the Grotesque and Arabesque, and he looked into such irrational subjects as mesmerism and phrenology. Cryptology, more than other subjects, is split the same way. It beams the hard bright searchlight of reason upon the phenomena it investigates. At the same time it glimmers with the pale, eerie, indistinct moonshine of mysticism and spooky powers. This double aspect of cryptology played upon Poe’s dual nature—its science upon his intellect, its occultism upon his emotions—and aroused thereby a wholesouled response.
Evidence of Poe’s interests in writing and mystery appeared in The Narrative of Arthur Gordon Pym, when outlines on maps and flakings of cavern walls inexplicably spelled out words in Ethiopian and Arabic. His first known mention of cryptology itself occurred, however, in an article entitled “Enigmatical and Conundrum-ical” that appeared in the December 18, 1839, issue of Alexander’s Weekly Messenger, a Philadelphia newspaper. After printing a riddle that had baffled one of the paper’s subscribers, Poe wrote:
We sympathize with our correspondent’s perplexity, and hasten to remove it—especially as we have a penchant for riddles ourselves. In spite of the anathemas of the over-wise, we regard a good enigma as a good thing. Their solution affords one of the best possible exercises of the analytical facul
ties, besides calling into play many other powers. We know of no truer test of general capacity than is to be found in the guessing of such puzzles. In explanation of this idea a most capital Magazine article might be written. It would be by no means a labor lost to show how great a degree of rigid method enters into enigma-guessing. So much is this the case, that a set of rules might absolutely be given by which almost any (good) enigma in the world could be solved instantaneously. This may sound oddly; but is not more strange than the well known fact that rules really exist, by means of which it is easy to decipher any species of hieroglyphical writing—that is to say writing where, in place of alphabetical letters, any kind of marks are made use of at random.*
And the footnote to this read:
* For example—in place of A put † or any other arbitrary character—in place of B, a *, &c. &c. Let an entire alphabet be made in this manner, and then let this alphabet be used in any piece of writing. This writing can be read by means of a proper method. Let this be put to the test. Let any one address us a letter in this way, and we pledge ourselves to read it forthwith—however unusual or arbitrary may be the characters employed.
Replies poured in on Poe. At first they came just from Philadelphia and environs, but later they arrived from Alabama, Massachusetts, New York, Ohio, Indiana, and Iowa—the correspondent from Iowa being 17-year-old Schuyler Colfax, later to be Vice President in Grant’s first term. The cipherers used wild combinations of asterisks, question marks, numbers, paragraph marks, and once “the ugliest and drollest hieroglyphics imaginable (we having no type in our office which would come within a mile of them).” The volume increased so that he asked his readers: “Do people really think that we have nothing in the world to do but to read hieroglyphics? or that we are going to stop our ordinary business and set up for conjurers? Will any body tell us how to get out of this dilemma? If we don’t solve all the puzzles forwarded, their concocters will think it is because we cannot—when we can. If we do solve them we shall soon have to enlarge our sheet to ten times the size of the Brother Jonathan” (a “mammoth” New York newspaper about two feet by three). Things never got that bad, but the great number of messages afforded Poe a lucky opportunity to explain why he limited his challenge to ciphers that were—for him—so easy to solve: “Were we to engage in the solution of every kind of puzzle sent us, we should have our hands full. We said that we could and would solve every cipher, of a stipulated character, which we should receive, and we have kept our pledge more than ten times over.”
In the 15 numbers of Alexander’s Weekly Messenger in which he had articles on cipher, Poe published the ciphertexts and solutions of 11 cryptograms and the solutions only of 16; he merely stated that he had solved three others. Six he had not solved: one he had lost, one he had had no time to examine, one was written in pencil and defaced, two were “impositions” (false cryptograms), and one had 51 different cipher characters and hence lay outside the limits of strict monalphabeticity with no homophones that he had laid down in his original challenge.
During all these months, he never once revealed how he solved the cryptograms, though his readers begged, “Just let us into the secret, as we are fond of the marvelous.” Teased Poe: “Well, what will he give us for the secret?—it is a wonderful one and well worth paying for. Let him send us on a list of forty subscribers, with the money, and we will give him a full explanation of our whole method of proceeding.” After solving an alliterative message that used, apparently, a different alphabet for each line, he promised to reveal his method of solution. But he evidently realized that the mystery heightened interest, for in the next issue he recanted: “Upon second thought, we must decline giving our mode of solution for the present.”
The closest he ever came to doing so was when he demonstrated how he deduced that a challenge sent him by G. W. Kulp, of Lewiston, Pennsylvania, was a false cryptogram. He picked out three words in the cryptogram—MW, LAAM, and MLW. Since “all English words of but two letters consist of a vowel and a consonant,” he wrote, MW must be one of 30 words, which he listed. He then inserted every letter of the alphabet in the middle of all 30 words in an exhaustive trial process to see which letters would make a sensible word out of MLW. Here he found 13, including ash and tho’. Turning to LAAM, he observed, that “if MLW be ash, then LAAM will be a word of this form, s..a, in which the dots represent two unknown letters of the same kind.” He ran through his 18 words in this way, and found that the only one that gave a possible meaning for LAAM was h..t, or hoot, “LAAM is then hoot or nothing. But the hypothesis of the word hoot is founded upon that of the word tho’ …. We now arrive at a definite conclusion. Either Mr. Kulp’s puzzle is not genuine, or MW stands for to, MLW for tho’, and LAAM for hoot. But it is evident that this latter cannot be—for in that case both w and A represent the letter o. What follows?—why that Mr. Kulp’s puzzle is no puzzle at all. This demonstration is as absolutely conclusive as any mathematical one could be. The process of reasoning here employed is that employed also in the solution of the cyphers.”
All his Alexander’s cryptograms were simple monalphabetics with word divisions (except one which appears to have been some kind of Cardano grille, whose solution only he gives). None of Poe’s correspondents tried to stump him with puzzle-type cryptograms that use bizarre words and twisted syntax, which would have been perfectly within the limits of his rules. In fact, they often made it especially easy for him by using well-known pieces for their plaintext, such as the Lord’s Prayer or the opening lines of Twelfth Night. Identification of a word or two of these would give away the entire message, and indeed “a single glance” enabled Poe to read the Lord’s Prayer cryptogram. Most employed symbol alphabets constructed without any key, though one correspondent actually used an a = 1, b = 2, …, z = 26 substitution. Poe solved them quickly: “Our correspondent will know by the date of his communication, that we could only have received it on the morning when we go to press (Tuesday)—consequently we must have read his puzzle instanter” he wrote in giving his first solution in Alexander’s. He also solved some sloppily, giving solutions with errors or omissions. This rather suggests that he attacked his cryptograms inductively, guessing at words by trying combinations of letters, as in the Kulp demonstration, rather than carefully analyzing ciphertext frequencies and characteristics. This intuitive approach, which offers a bolder and easier solution, would accord well with his habits of thought and work.
Poe’s solutions for Alexander’s became the narrow foundation upon which he erected an exaggerated reputation as a cryptanalyst. He enlarged the foundation slightly in 1841 when he solved cryptograms in a polyphonic substitution, a little more difficult than ordinary monalphabetic substitution, submitted by readers of another magazine. It was his publicitywise skill at magnifying these rather ordinary deeds, however, plus the testimony of others, that created the legend of his almost supernatural cryptanalytical prowess. An article on “Quick Perception” in the Philadelphia Saturday Museum in 1843 reported how Poe was shown a cryptogram and “immediately” gave the answer. He supplied material for a biography of himself in the same issue which cited the most difficult cryptogram that he had ever solved as the easiest: “This cryptograph, however,” the article observed falsely, “was simplicity itself in comparison with others resolved by the subject of our memoir.” A friend reported in a letter how Poe solved a cryptogram “in a much shorter time” than it took to encipher it—and Poe promptly published the letter. The legend appeared in full flower less than a year after Poe’s death. A Massachusetts clergyman, telling how Poe read a cryptogram “in one-fifth of the time it took … to write it,” concluded that “The most profound and skilful cryptographer who ever lived was undoubtedly Edgar A. Poe.”
The myth has shrunk since then. But vestiges linger. Poe student W. K. Wimsatt, Jr., has contended that Poe’s cryptanalytic powers were “far beyond the ordinary.” Beyond the ordinary individual, of course, and even beyond the person who may have a curiosity about
ciphers. But not beyond the ordinary amateur. Most amateurs that do any solving at all solve monalphabetics with quite the same facility as Poe. Just a modicum of experience endows a solver with an apparently miraculous facility to recognize words and word-patterns at a glance—KVBK stands out as that, KVILI as there, AVTOV as which. The solution of such ciphers is not only simple but also becomes, upon the slightest practice, mechanical. On the available evidence, then, it is wrong to say that Poe’s cryptanalytic skill exceeded the ordinary. But what about his latent capacity, what Wimsatt called his “native power”? Other amateur cryptologists of the day—Babbage, Wheatstone, Kasiski—sometimes solved ciphers much more difficult than monalphabetic substitution. Would Poe have been able to do so too? The question is unanswerable. Poe limited himself to the simplest kind of cipher. To decide on the basis of no evidence whether he did so because he feared to tackle the more complicated systems, or because, as he said, he did not have time, would be arbitrary and useless.
Nevertheless, Poe’s reputed cryptanalytic superiority has colored the whole picture of the man. “Doubtless nothing contributed to a greater extent than did Poe’s connection with cryptography to the growth of the legend which pictures him as a man at once below and above ordinary human nature,” wrote Joseph Wood Krutch. For by associating himself with cryptology, Poe clothed himself in the spectral aura that has always shimmered about it.