by Jason Fagone
I challenge anybody who dares ignore the solid work done through our agencies. . . . Our men worked, suffered. Still suffering. . . . We have succeeded in laying foundation for future work . . .
The two young codebreakers ended up solving the whole trunkful of messages for Scotland Yard, revealing an intricate separatist plot by Hindu activists living in New York to ship weapons and bombs to India with the help of German funds and assistance: dates, times, places, names. Several conspirators were charged in San Francisco, and prosecutors summoned William to testify in open court about how he broke the codes. Elizebeth wasn’t asked. She hated staying in Illinois while William went on an exciting trip to the West Coast—she thought she deserved to be called as a co-witness—“but someone had to stay behind and sort of oil the machinery at Riverbank.” She didn’t speculate about why William was chosen instead of her, perhaps because the answer was obvious: prosecutors thought the jury would more easily believe a male expert. As it turned out, the trial erupted in spectacle: Before William had a chance to say his piece on the witness stand, an Indian man in the gallery stood up, pulled a handgun from his vest, and shot one of the defendants in the chest. He yelled a single word—“Traitor!” Then a U.S. marshal fired at the gunman over the heads of the shocked spectators, killing him. The shooter apparently thought the defendant had snitched to the government, betraying his friends by revealing the code. He didn’t know about William, Elizebeth, and the science of codebreaking.
For the first eight months of the war, as incredible as it sounds, William and Elizebeth, and their team at Riverbank, did all of the codebreaking for every part of the U.S. government: for the State Department, the War Department (army), the navy, and the Department of Justice. And the broader scientific insights of the Riverbank Publications emerged directly from these day-to-day puzzles solved under wartime pressure. The pair would solve a cryptogram and realize they may have stumbled onto some more general method. Then they would test the method on additional examples, trying to see where it broke, what its limits were, aiming to strengthen a one-time solution into a universal principle and to share that knowledge with others.
It had long been known that the frequencies of letters in a cryptogram provide clues to its solution. Knowing this, cryptographers had invented many ways of obscuring the letter frequencies, making messages harder for adversaries to break. It was possible to encrypt a message with multiple cipher alphabets instead of just one (a poly-alphabetic cipher). It was possible to rely on a secretly chosen novel or dictionary to generate a code message, in the style of the Hindu revolutionaries. If the sender and recipient happened to select a key in their book that was exactly the same length as the message—known as a “running key”—the message became even harder to break. The War Department considered running-key messages to be indecipherable. And these methods could be mixed and matched to further frustrate the codebreaker. For instance, a plaintext word like “strawberry” could be turned into a block of code like WUBCW, then those letters transformed with a cipher into LWJIJ—a process of “enciphered code.”
Each of these techniques placed a wall between the message and the codebreaker—sometimes a pane of frosted glass, sometimes a sheet of metal or stone. Elizebeth and William invented new tools for destroying these walls—hammers, corrosive acids, explosives. They learned to identify and solve several different kinds of substitution ciphers: straight alphabets, direct alphabets, reversed alphabets, poly alphabets, mixed alphabets. They developed general techniques of solving book ciphers without needing a copy of the book. They taught themselves to solve messages enciphered with running keys. Together in Engledew Cottage they strolled through cities of text with their wrecking kit, swinging hammers with glee, blowing up brick, melting steel, the sound of breaking glass echoing out into the prairie. Then they wrote the scientific papers, the Riverbank Publications, documenting exactly how they did it, and how other people could do it, too, if they followed the same steps.
This part was crucial. The test of a scientific discovery is if others can replicate it and get the same results. Mrs. Gallup had never passed this test. Elizebeth and William wanted to pass. They later wrote, “What Colonel Fabyan failed to realize, throughout his campaign to ‘sell’ Mrs. Gallup’s decipherments, was that no demonstration, however good, can take the place of experiments which can be repeated and will produce identical results.”
To drive home this point, William even invented a new word: “cryptanalysis,” synonymous with “codebreaking.” The new Riverbank methods were not magic but a species of analysis, similar to the analysis performed by a chemist or an astronomer or an engineer designing a bridge.
Serendipity still played a role in codebreaking. “Many times,” the pair wrote, “the greatest ally the mind has is that indefinable, intangible something, which we would forever pursue if we could—luck.” Epiphanies happened. Insights that seemed to come from nowhere, bolts from the blue, guesses that made more progress on a problem than days of dreary labor. Mrs. Gallup had always called this “inspiration.” Elizebeth and William preferred to speak of “flexibility of mind” or “intuitive powers” because these phrases sounded less magical. Intuition, to them, was like a hard-earned internal compass, a grooved-in sense of how to move forward that came from patience, skill, and experience. It could be cultivated.
Starting here at Riverbank and continuing throughout their lives, people tended to describe the brains of Elizebeth and William in gendered ways, as if her style of solving puzzles had a starkly different texture than his. Elizebeth’s was usually said to be the more intuitive mind, William’s the more mathematical. He was supposed to be better with machines and she with languages—Elizebeth was rapidly picking up German and Spanish, and learning pieces of other tongues. There may have been some truth in it. But the reality is that they were both mathematical neophytes, even William. A future colleague of William’s, Lambros Callimahos, a classical flutist and trained mathematician, idolized William to the point of copying his personal habits; upon learning that William liked to use tobacco snuff, Callimahos took up snuffing. But Callimahos recognized that whatever made William good had little to do with math. He described William as a man “cursed by luck,” writing, “Even if he computed odds incorrectly, it didn’t make any difference because he would forge ahead in his blissful ignorance and solve the problem anyway. On several occasions he told me that if he had had more of a mathematical background, he might not have been able to solve some of the things he did.” If William had been older or better trained, “he could have been ruined. His definition of a cryptogram was simply a secret message that was meant to be solved, just that.”
To those who had a chance to watch them both work, the minds of William and Elizebeth appeared equally amazing and equally incomprehensible. Their brains were Easter Island statues, stony and imposing. Colleagues resorted to mystical analogies. William was like a latter-day King Midas: “Everything he touched turned to plaintext.” Elizebeth’s gift for puzzles was “God-given,” “an effect without a discernible cause.” Who was the better codebreaker, William or Elizebeth? People gave up trying to figure it out. A raffish young army officer from Virginia, J. Rives Childs, met William and Elizebeth at Riverbank in November 1917; they taught him the science of codebreaking, and he went on to serve with distinction in the war. Childs found it impossible to tell if William was smarter than Elizebeth or if it was the other way around: “I was never able to decide which was the superior.”
Elizebeth and William sometimes played into the stereotype that he was the mechanical male thinker and she the sensitive female thinker. It was a helpful shorthand for explaining the inexplicable.
There’s a now-famous story that encapsulates how they thought about their own differing brains. One day during the war, a series of five short messages arrived at Riverbank from Washington. It was a test of sorts. The messages had been encrypted with a small hand-operated device recently invented by the British army to make their f
ield communications more secure. The device was a kind of cipher disc, with two alphabets printed on rings that rotated with respect to each other, but with a twist: while the outer ring had the usual 26 letters, the inner ring had 27. The extra letter introduced a degree of irregularity, making it harder for a codebreaker to visualize the alphabets sliding against each other. The device also allowed the cryptographer to change the alphabets quickly and easily.
The British had already concluded that the device was unbreakable. So had experts in France and a few in America. But to be certain, an official in Washington had used the device to encrypt five test messages, using two alphabets of his choosing. He then sent the messages to the Colonel, to see if Riverbank could solve them.
William looked at the messages. He had been given a description of the device that produced them, but not alphabets. His only chance to solve the messages was to reverse-engineer the alphabets the Washington official had used.
He began with the assumption that the official wasn’t an expert cryptographer: a safe assumption, because almost no one is. Therefore the official might have made any number of common blunders that people often make when trying to communicate securely. The strength of a cryptographic system usually has less to do with its design than with the way people tend to use it. Humans are the weak link. Instead of changing keys or passwords at regular intervals, we use the same ones over and over, for weeks or months or years. We repeat the same words (such as “secret”) at the start of multiple messages, or repeat entire messages multiple times, giving codebreakers a foothold. We choose key phrases that are easy to guess: words related to where we live or work, our occupation, or to whatever project we’re working on at the moment. A couple of human mistakes can bring the safest cryptographic system in the world to its knees.
It struck William that the Washington official, in preparing this important test of a cryptographic device, might have used key words related to the practice of cryptography. So William tried words like “cipher,” “alphabet,” “indecipherable,” “solution,” “system,” and “method.” After two hours of intense focus, he was able to piece together what he thought was the alphabet on the outer disc, which seemed to use the key phrase cipher. William now assumed that if the official had been careless enough to use a guessable key phrase in one alphabet, he had probably used a similar key phrase in the second alphabet too, on the inner disc. But this one proved tougher to crack. William tested all sorts of key phrases; nothing worked. He turned to Elizebeth for help.
“I was sitting across the room from him,” she recalled, “busily engaged on another message”:
He asked me to lean back in my chair, close my eyes and make my mind blank, at least as blank as possible. Then he would propound to me a question to which I was not to consider the reply to any degree, not even for one second, but instantly to come forth with the word which his question aroused in my mind. I proceeded as he directed. He spoke the word cipher, and I instantaneously responded, “machine.” And, in a few moments Bill said I had made a lucky guess.
Later, in writing and interviews, Elizebeth would try to explain the “springlike elasticity” of her mind in this moment. What led her to blurt out the word “machine”? Where did it come from? All she would say is that because the British device was small and hand operated, “it did not occur to [William’s] meticulous mind to use the word machine. But to me it was a machine.” Thanks to Elizebeth’s guess, she and William were able to solve the five test messages in less than three hours.
William attributed Elizebeth’s insight in this case to the fact that she was a woman. He later said in a lecture, “The female mind is, as you know, a thing apart.” He appears to have made a joke about sex as well. He and Elizebeth had just gotten married when this story took place. William recalled in his lecture to a roomful of men, “I came to the end of my rope and said to the new Mrs. Friedman: ‘Elizebeth, I want you to stop what you are doing and do something for me. Now make yourself comfortable’—whereupon she took out her lipstick and made a few passes with it.” Imagine laughter.
To hear Elizebeth tell it, William was the brighter one. Before they were married and before they were a courting couple, she was already starting to praise his abilities in a way that minimized or overlooked her own, setting the pattern for the rest of her life, the moments when she would describe William to friends and to reporters as a man of history and destiny, “a wonderfully warm man, with the broadest of minds and intelligence,” and even “the smartest man who ever lived.” All the same, she was competitive by nature, and at times the two of them indulged a cheerful rivalry.
Once, in a dusty 1896 issue of the literary magazine Pall Mall, William and Elizebeth discovered an article about ciphers used by anarchist opponents of the old Russian czars. The article included a brief cryptogram at the end. In general, the shorter the cryptogram, the harder it is to solve, the same way a song is harder to identify by three notes of its melody instead of twenty. This “Nihilist” cryptogram consisted of only a few numerals and two question marks:
(Hathi Trust Digital Library)
No solution was given. “The meaning of the cipher which now follows will never be solved by any one,” the author wrote, concluding that the lock “has now closed and firmly shut its fastenings.”
Naturally, William grabbed a pencil and began trying to pick the lock. “Well,” Elizebeth writes, when William “met up with that message, he took the challenge and set his teeth into the tough nut with a snap. And would you believe it, he deciphered the message, short as it was, and the key, in 15 minutes!” The key relied on a single repeated word: “courage.” The plaintext read: “He who fears is half dead.”
It’s a convincing piece of testimony to his greatness: William Friedman, the smartest man who ever lived. But instead of ending the story here, Elizebeth goes on: “Of course, when I learned that, I too had to try my hand” at the cryptogram. “I unlocked the forever-to-be-hidden secret in 17 minutes.”
By the spring of 1917, William was in agony. He had known Elizebeth for eight or nine months now and he wanted her all the time. He was afraid to say it out loud because he didn’t know if she would reciprocate. She had never called him anything but a friend. But he was sure he loved her. It was getting hard to sit with her all day and pretend to be thinking about work. In the moments when he appeared to be scratching away at a puzzle, he was really wondering what her hair would look like if he reached behind her neck and removed the pin, her beautiful loosened hair if he pulled her close.
He imagined a life with her, a house, children, and at the same time he could not imagine it. He knew that his Jewish family and friends in Pittsburgh would not approve of him marrying a non-Jew. The community there had always seen marriage between Jews and Gentiles as a kind of betrayal, a weakening of Jewish resistance to a hostile and bigoted American culture. When William was growing up, Pittsburgh’s Jewish Criterion weekly newspaper made the case against intermarriage in repeated articles and editorials:
The glacial undercurrents of racial antipathy between Jew and non-Jew cannot be tepified by even the hottest fiercest rays of the sun of love! Statistics and the divorce courts prove this.
A part cannot become merged into a whole without ceasing to be a part. The Jews don’t want to merge.
WILL THE JEWS COMMIT SUICIDE THROUGH MIXED MARRIAGES?
He feared what his people would say. But desire trampled the fear.
Soon enough, but not yet, he would tell Elizebeth what he thought of her: that “your soul and spirit and heart are as fine, sweet, and pure as your body is beautiful”; that the perfection of her body was matched only by the clockwork of her mind, “brilliant and quick and clever”; that she had him “skinned to a frazzle” in the brains department. He marveled at her ability to escape the bounds of a problem, to strike the flint of her thoughts against different rocks, against history and math and logic and against William, too, shooting out sparks, ribbons of flame. “You’re lots sma
rter than I am in ciphers. You can soar away into the clouds and still remain planted firmly upon solid ground and reason. You can dream and be practical.” He would tell her, over and over, that he couldn’t express what was in his heart: “Oh Divine Fire Mine, I adore you, how futile are words!”
Divine Fire. A nod to the Bible of his father. The devouring God of the Old Testament, whose fire is in Zion, and his furnace in Jerusalem.
One of the mysteries of falling in love is that it makes you inarticulate and eloquent at the same time. You lose the ability to speak and write in normal ways (How futile are words!) even as you develop, with this person you love, assuming this person loves you back, a shorthand of glances and gestures. At first it seems like your beloved is “speaking in code”; later, maybe, it’s like the two of you are sharing a secret code.
This feeling may have a deep scientific explanation. In the 1930s and ’40s, before the digital computer was invented, a young scientist from rural Michigan named Claude Shannon wrote two papers that were like magic beans for the computing revolution, growing the great beanstalks of IBM, Apple, Silicon Valley, the Internet. As a graduate student at the Massachusetts Institute of Technology, Shannon realized that electronic circuits could be arranged to solve logic problems and make decisions, and that 0s and 1s could encode all the world’s information, from a song to a Van Gogh. He didn’t create the first computer, but he was one of the first to grasp the immensity of what digital computers could do.