by Jason Fagone
Hoover got his wish in June 1940, with a presidential directive that represented a historic expansion of the FBI’s power. For the first time, the bureau was free to dispatch agents into other countries. He created a new division called the Special Intelligence Service (SIS) and began recruiting agents for duty in South America.
Their mission would be to find and monitor the secret mail drops and radio stations used by the spies; to map the structure of their organizations and communications networks; to determine the true identities of the enemy agents; and to cooperate with local State Department officials and police in arresting the spies, seizing the radio stations, and destroying the rings.
A tall order. The first five SIS agents were dispatched to the continent in September 1940, one each to Peru, Uruguay, Argentina, Brazil, and Venezuela. Pale and corn-fed, they stepped off their planes into the lacerating sun of another continent. They wore snap-brim hats and looked like detectives that South Americans had seen in newspapers and movies. The agents knew little about codes, ciphers, or radio, these crucial tools of their adversaries, and didn’t speak the local languages. The SIS man sent to Brazil had been given a crash course in Spanish. When he arrived, he realized, to his frustration, that the language of Brazil was actually Portuguese.
Hoover’s men in South America were so unprepared that they had almost no chance of catching the spies through old-school gumshoe tactics: interviewing associates, recruiting confidential informants, developing leads. They needed to know what the spies were saying to one another in private. They needed codebreaking. And this was exactly the problem.
To break codes, you need intercepts and you need codebreakers to solve the intercepts. The FBI had neither. It had no intercepts because it had no listening stations; when the bureau wanted intercepts it was forced to obtain them from the coast guard and the Federal Communications Commission (FCC). And when the FBI got these intercepts, it couldn’t read them, because the FBI had no codebreaking unit. What it had instead was a Technical Research Laboratory, essentially a crime lab, a place where bureau technicians analyzed bullets, fingerprints, threads of fabric, and blood samples.
All of this spelled trouble for J. Edgar Hoover. At the very moment he was launching a hemisphere-wide hunt for spies who communicated in code, his bureau had no ability to discover what they were saying.
Around this time, Elizebeth received an unusual order from her Treasury bosses: They asked her to visit FBI headquarters. She was to teach codebreaking to an agent named W. G. B. Blackburn, an employee in the Technical Lab. Elizebeth proceeded to train Blackburn in codes and ciphers, much as she had trained her own junior colleagues, and Blackburn established a small Cryptographic Branch at the FBI, which would grow to the size of a handful of employees over the next several years, all of them codebreaking novices.
This still wouldn’t do for Hoover’s purposes. The Invisible War demanded a level of technical firepower and prowess that his Technical Laboratory simply did not command. What he required was the full assistance of a mature codebreaking organization, whether they wished to help him or not. He needed Elizebeth and the coast guard.
She read pacifist poetry. It resonated. She thought of her kids. Barbara was in her last year of high school and planned to attend college at Radcliffe, and John Ramsay was a fourteen-year-old freshman at Mercersburg Academy, an elite boys’ school in rural Pennsylvania. He wasn’t young enough to be safe from a military draft. War would scatter her family. She also worried about the fate of her team at the coast guard. She had built this little organization and it was good and she wanted to protect it from disruption. Codebreaking is delicate work. You have to look at the page and get all the letters aligned just right, then you have to look at your team and get all the people aligned just right, so that the flow of intercepts and records and ideas and solutions becomes as efficient as possible.
Elizebeth escaped Washington for a week in June 1940, traveling to Mexico on a quick vacation with daughter Barbara and sister, Edna. It was the last time in the next five years she would get a break, a chance to pause and look around and spend time with the women closest to her. They drove a beat-up rental car through the farmlands of Oaxaca and the mountain ranges of Puebla Cordoba, descending into canyons on the backs of burros. Elizebeth wrote to William, “All Mexico is so full of resounding cockcrows, pig-grunts, burro-brays, and church bells that all sleep is intermittent, at best.” The two sisters had a great time and woke up early each day; Barbara wanted to sleep in and complained that the altitude made her knees wobbly. She was a gorgeous girl of seventeen now, six inches taller than her mother, confident and voluptuous. One day, when they were all on a plane above Oaxaca, Elizebeth happened to fall asleep in her seat, and when she woke, she saw Barbara up in the cockpit, next to the pilot. Wait, what kind of airline was this? Are girls just allowed to ride in the cockpit without their mother’s permission? Isn’t that unsafe? She felt like a mom.
The news of the war got rapidly worse while she was in Mexico. She had to stop reading the papers in the morning because it was too depressing. Nazi tanks were said to be plowing through the French countryside on the way to Paris. The Mexican papers seemed to think America was bound to join the war. The peso was rising, eating into Elizebeth’s meager trip budget of fifty dollars. She airmailed William a letter about the rising cost of goods. In his reply he begged her not to spend more money than was absolutely necessary, “or we shall never never climb out of this morass of debt.”
She wasn’t sure if William was okay. He sounded sad and mopey in his letters. He said it had been rainy in Washington, and in the evenings he had been sitting alone with a pencil and a pad, listening to the rain on the roof, writing a technical paper on cryptology. He told her, “There won’t be anybody [to] read this thing, I imagine, at least not for some centuries,” and added a lament about the shackles of secrecy: “I wish I could write about forbidden subjects. What a story could be told.”
By the time she got back to Washington—to home, husband, and job—the Nazis had entered Paris, hanging the swastika flag from the Arc de Triomphe.
CHAPTER 2
Magic
One day in September 1940, inside the windowless vault of William Friedman’s army codebreaking unit, one of the two female codebreakers on the team, Genevieve Grotjan, stood at her desk and let the men know that she might have found something.
The men called her Gene. She was twenty-eight and quiet and wore rimless glasses. She had a background in statistics; she would later teach math as a professor at George Mason University in Virginia.
Gene Grotjan had been looking at raw Purple intercepts for hours, weeks, months when she called the men over. The team had not been able to penetrate the garbled Japanese text on the intercept sheets that had been streaming into the Munitions Building. But now Grotjan thought she had noticed two patterns that others had missed—subtle cycles of repetition, loops of letters in the text, much like the ones discovered by the coast guard in Enigma messages. Frank Rowlett, one of William’s deputies, came over and looked at Grotjan’s worksheets. Then he looked at her. He got the sense that her eyes were beaming through her glasses, that she was struggling to contain her emotions. Others started to crowd around the desk. Rowlett started jumping up and down.
“That’s it!” he shouted. “That’s it! Gene has found what we’ve been looking for!” Another man busted into a funny little dance. He threw his arms up in a victory pose. “Whoopee!”
Grotjan was a modest person. “Maybe I was just lucky,” she said later in an NSA oral history. “I perhaps had a little more patience” than some of the other workers. She didn’t become as animated as Rowlett because “I regarded it more as just one step in a series of steps.”
William Friedman heard the commotion and shuffled into the room from his nearby office: “What’s all this noise?”
Rowlett showed Gene’s worksheets to the boss, and William agreed within seconds that they revealed a loose thread in the code.
There was still more to do, but it was clear that the thread, if pulled, would allow the team, with great grinding effort, to recover the daily keys and consistently read the Japanese messages.
The men were bouncing and laughing with the excitement of the discovery. Friedman seemed almost sad. “Suddenly he looked tired,” Rowlett later recalled, “and placed his hands on the edge of the desk and leaned forward, resting his weight on them.” Rowlett knew that Friedman had been under a lot of stress, working sixteen-hour days for weeks, months, years. The younger man pulled out a chair for his elder. Friedman sat quietly for a few moments. Everyone was looking at him, waiting for his reaction. He turned to the codebreakers. “The recovery of this machine will go down as a milestone in cryptologic history,” he said in a formal, distant voice. Then he left the room.
In codebreaking, the larger the success, the more it must be suppressed. Any leak might reach the enemy and cause them to switch to a new code system, destroying the value of the break. Heroes celebrate briefly and in secret. Someone went and got Cokes.
A few minutes later, the group dissipated, and everyone returned to their desks to explore the new textual terrain they had unlocked—everyone except Rowlett, who went looking for Friedman. The young man was still shaking, still full of adrenaline, wanting some kind of catharsis. The boss’s lack of enthusiasm confused him. Rowlett found Friedman in his office, “sitting at his desk, studying some notes he had made on a pad. When I entered the room, he sat quietly, merely looking questioningly at me.”
Over the next hours and days, the team kept applying pressure to the hairline fracture in the code until it shattered and the first bits of plaintext revealed themselves. Five days after Grotjan’s discovery, on September 25, 1940, the team produced their first full decrypted message. It was a big moment. William and the rest of the codebreakers had never been able to look at the Japanese machine, or touch it. They had never seen a drawing of it, or a patent illustration, or a photo. Yet they now understood how it worked and how to recover the daily key for a given set of messages. People had reverse-engineered cipher machines and devices before, but nothing at Purple’s level of complexity. Today historians of cryptology believe that in terms of sheer, sweaty brilliance, the breaking of Purple is a feat on par with Alan Turing’s epiphanies about how to organize successful attacks on German Enigma codes.
Once William and his colleagues fine-tuned their bootleg Purple machine, the one they were building to help them read Japanese messages, they demonstrated it for the unit’s commanding officer, typing out a sample of ciphertext and then decrypting it as he watched. A sheet of fresh plaintext inched its way out of the machine, and after the man grabbed it and looked for a few seconds, a smile lit up his face, and he congratulated the codebreakers on their “magnificent achievement.” Then he rushed off to get his commanding officer, Joe Mauborgne, William’s old friend. When they returned together, the first officer pointed to the machine and said to Mauborgne, “Last night your magicians completed the reconstruction of the new Japanese cipher machine,” and the codebreakers repeated the demonstration for Mauborgne. “By God, it really works beautifully!” Mauborgne said.
Your magicians. It really did seem like that. Like magic.
MAGIC became the top-secret moniker for these Japanese decryptions, for the astonishing fountain of secrets that would keep gushing up all through the war, secrets of Japanese strategy and Nazi tactics that flowed across Japan’s encrypted circuits and were tapped by the U.S. Army (and later by the U.S. Navy too, which solved Japan’s naval cipher), giving Allied planners the drop on their foes. The first handful of decrypted messages turned into 20, into 100, into 1,000, piped directly from William’s unit through the corridors of power. MAGIC beguiled all who touched it. Men read the daily “MAGIC Summaries” with bulging eyes and could not quite believe they were reading the authentic words and orders of imperial Japan. It was almost too good to be true. The president read MAGIC, and Army Chief of Staff George C. Marshall, and Secretary of the Navy Frank Knox, and, eventually, Prime Minister Winston Churchill in Britain, who insisted on getting MAGIC raw, unsummarized by his generals.
MAGIC led directly to bombs falling on imperial ships at Midway and other decisive naval battles. It caused the deaths of hundreds of thousands of Japanese and saved the lives of unknown numbers of Allies. MAGIC changed the war. It was also one of the great secrets of the war, exactly like ULTRA, the Enigma codebreaking program. These were tremendous military advantages that could not be revealed to the enemy lest the enemy get wise and cut off the stream of intelligence. The advantage “would be wiped out almost in an instant if the least suspicion were aroused regarding it,” George Marshall wrote later in a classified letter that captured the value of MAGIC in the Pacific War against Japan:
The Battle of the Coral Sea was based on deciphered messages. And therefore our few ships were in the right place at the right time. Further, we were able to concentrate our limited forces to meet their advances on Midway, when they otherwise would certainly have been some 3,000 miles out of place. We had full information of the strength of their forces in that advance, and also of the forces directed against the Aleutians, which finally landed troops on Attu and Kiska. Operations in the Pacific are largely guided by the information we obtain of Japanese deployments. We know their strength in various garrisons, their rations, and other stores available to them. And what is of vast importance, we check their fleet movements and the movements of their convoys.
The triumph over Purple would turn out to be William’s last hurrah as a hard-core codebreaker, his final death-defying climb. From now until the end of his life, he would serve America as an inventor of cipher machines and an architect of intelligence institutions (and ultimately a critic of them as well). He had reached his peak. Elizebeth, though, was still climbing, and she couldn’t see him up there, across the gap between their two towers, starting his descent. She couldn’t share this victory with him, because on the day he and his team broke Purple, a historic achievement that had required all of his battered brain, all he had learned in his unexpected life of exploration with the woman who meant everything, he said nothing about it to her when he came home. He did not seem different to her than he did on any other evening. He said hello and asked what was for dinner.
Heavy air attacks on London had begun that month. The Blitz. On September 7, 1940, shortly after 5 P.M., a thousand German planes appeared in the sky above London. It was a bright blue afternoon. The planes arranged themselves in vertical formations, fighters and bombers. The fighters had bright yellow noses and tails. The bombers were black and set their sights on industrial facilities that lined the Thames River. The bombs destroyed factories, shock waves and oily smoke rippling out. British Spitfires gave chase to the German planes. “The sky seemed full of them,” one British pilot later said, “packed in layers thousands of feet deep. They came on steadily, wavering up and down along the horizon. ‘Oh golly,’ I thought, ‘golly, golly. . . . ’ ”
The bombings of London continued for fifty-six straight days. Sirens and shelters, blackouts at night. The Axis was growing bolder in the final months of 1940. Japan invaded Vietnam, expanding its empire in East Asia. The Nazis confiscated the private radios and telephones of Jewish families and cordoned off the Warsaw Ghetto with barbed wire, trapping 400,000 adults and children, most of them Polish Jews.
America didn’t want war. Both major political parties still supported neutrality. The aviation pioneer Charles Lindbergh argued in popular radio speeches that it would be foolish and hypocritical to fight Germany. He said America had no standing to accuse the Nazis of aggression and barbarism because America had sometimes been aggressive and barbaric itself. Later he argued that American Jews were a “danger to this country” on account of their “ownership and influence in our motion pictures, our press, our radio and our government.” Lindbergh became the public face and champion of an antiwar group called the America First Committee. “America First,” a c
ampaign slogan of Woodrow Wilson, had been adopted by the Ku Klux Klan in the 1920s. Within a year the America First Committee was holding rallies at Madison Square Garden.
The worsening of the war in Europe, combined with U.S. reluctance to fight, was about to drag Elizebeth into the orbit of a highly motivated and capable group of British spies. The British were afraid. They knew they didn’t have the money, the people, or the weaponry to sustain a long fight against the Nazis. They needed America to join the war. Their survival as a nation depended on it.
In the early summer of 1940, British officers began to arrive in America on a covert mission. Some went to Washington, making the rounds of embassy cocktails and dinner parties, looking for all the world like bright young chaps out for a good time, and others worked in the heart of New York, in a Fifth Avenue skyscraper, the thirty-fifth and thirty-sixth floors of Rockefeller Center. The group included Ian Fleming, a handsome lieutenant with blue eyes and a smart blue naval officer’s jacket, and twenty-three-year-old Roald Dahl, a tall, elegant Royal Air Force fighter pilot who looked a bit like Gary Cooper. These guys would both become famous fiction writers after the war; Fleming invented the character of James Bond, and Dahl wrote children’s books about chocolate factories, flying peaches the size of zeppelins, and foxes who outwit monstrous humans. For now, though, Fleming and Dahl were spies. Dahl was a particularly good spy. He seduced actresses and heiresses in Washington, gathering gossip in bed, and he charmed the president and the first lady, becoming a regular guest at their Hyde Park, New York, home, where they spoke so freely with the young pilot that he had difficulty maintaining his composure: “I would do my best to appear calm and chatty,” he later wrote, “though actually I was trembling at the realization that the most powerful man in the world was telling me these mighty secrets.”