In the Enemy's House

by Howard Blum

  Sasha told himself he was doing spy work. He was at last part of the rezidentura. But at the same time he could not help reminding himself that his transmissions were just a bit of redundancy, the Center’s wanting to make sure that the unscrupulous Americans were not tampering with the consulate’s cables sent through commercial offices such as Western Union or RCA. Each day he spent in his attic hideaway in front of a radio transmitter he felt as if his reserves of ingenuity and commitment were being squandered. He wanted to be in the field, running flesh-and-blood agents in the streets of New York.

  SASHA FINALLY GOT HIS CHANCE because his best friend was color-blind.

  It was the silly season in the running battle between the Bureau’s SE squad and the Soviets. As soon as anyone suspected of being a KGB hood walked out of the consulate on East Sixty-First Street, the Bureau, or so it seemed to the rezidentura, would be breathing down their necks. Sasha made a game of following the watchers on his trail, and in his estimation the FBI pavement artists were a pretty motley crew.

  On a saunter across town to a steamship company to purchase tickets for some Russians heading home, he had no trouble spotting the opposition. The soldier riding behind him on the subway escalator. The man in shirtsleeves and no tie leaning on the counter in the ticket office. The man in the black raincoat and fedora in the cafeteria where he’d stopped for lunch. And that same soldier again as he walked back to the consulate. The enemy kept changing the guard, but despite the impressive show of manpower, the Bureau’s handwriting, he judged with a professional’s vanity, was entirely amateurish.

  It was so heavy-handed that Sasha couldn’t resist the urge to have some fun at the opposition’s expense. There he was on a busy midtown street, and across the way a fellow in a topcoat and hat was staring at him. There was not even the pretense of artfulness. So Sasha, all mischief, crossed over and confronted the FBI man. “He looked,” a bemused Sasha would always remember, “like a pickpocket caught in the act.” The agent recovered, though, and walked off, ducking into the first doorway he could find. Sasha followed. The desperate agent now twisted the handle—but the door was locked. That’s when Sasha started to laugh out loud. He was still laughing as the FBI man ran in embarrassment full-speed down the block.

  But while Sasha, still on the operational sidelines, was enjoying these spy-versus-spy encounters, Yatskov was actually running agents. He couldn’t afford to be observed. And that was why he turned to his best friend for help.

  It was a meet at a coffee shop on the Upper West Side, and Yatskov worried about the drive uptown. He was color-blind, unable to tell the difference between the red and green traffic signals. He would need to focus all his attention on the flow of cars in front of him. There’d be no opportunity to glance in his rearview mirror to see if he was being tailed. So he asked his friend to drive with him, and keep a sharp eye out for the enemy. And when they reached the coffee shop, he had Sasha stand on the corner and babysit, as security operations were known in the trade. If he spotted the FBI, his instructions were to go to the counter and order a coffee, and Yatskov and his contact would know to run for the hills.

  The meet went off without complications. After that, Sasha was given more opportunities to babysit for his friend. Then, one evening, without a hint of what was in store, Yatskov brought him to a restaurant on the East Side, near Thirtieth Street. Kvasnikov was already seated at a table in the back of the room. Speaking in a whisper so soft that Sasha had to lean across the table to hear him, Kvasnikov told him about Operation Enormoz.

  At that moment he became a member of the KGB’s most important covert team. He and his best friend would be running the networks stealing America’s atomic secrets.

  CONVINCED OF THE URGENCY OF his mission, Sasha went to work. He discovered that there was an army of possible recruits—scientists, intellectuals, and Party members—who believed it would be a disaster if America had sole possession of an atomic weapon. Some put their hopes for peace in a more equitable balance of power. Others were committed to the Soviet cause. And Sasha, playing caring friend, stern adviser, and shrewd accomplice, as if he had known no other life, was there in the shadows to help them.

  But all at once, in the aftermath of the succession of defections in the fall of 1945, Moscow Center ordered that the ongoing operations in the United States come to an abrupt halt. Enormoz, although ripe with tantalizing promise, was shut down. And Sasha, for all operational purposes, was shut down, too.

  IT WAS DURING THIS DORMANT period that a classified paper titled “A Review of the Uranium Problem” made its way to Laboratory Number 2 in Moscow. It had fallen into the New York rezidentura’s net—whether the credit should go to Sasha, or Yatskov, or to some other operative remains a mystery hidden in still-classified files—just before Enormoz had ground to a halt. It wasn’t a scientific treatise, but rather a memo written for British and American policy makers, and so perhaps that explained why it had taken so long before it was shared with the physicists laboring on Pyzhevskii Lane.

  But the young director of the Soviet atomic project, Igor Kurchatov, found the paper on his desk. Picking it up, he began reading but hadn’t gotten very far before he abruptly stopped. Full of his sudden discovery, he dashed off an excited personal letter addressed to General Fitin at Moscow Center.

  The brief note called attention to “the extremely curious remark on page 9” of the purloined report. And making sure the busy general would focus on the correct reference, Kurchatov patiently explained that this was a description of a secret facility in America called “Laboratory V” where scientists were calculating the physical properties of uranium-235 and plutonium “in connection with manufacturing a bomb.”

  And with that, the scientist’s letter came to its brusque end. There were no specific demands, no recommendations about how Moscow Center should proceed. But arguably this was a shrewdness since clearly none were necessary. After Viktor had warily turned it over in his mind, a cable went out to New York: Kvasnikov and his team were to find out all they could about the mysterious Laboratory V.

  Operation Enormoz was reactivated. And at the same instant, Sasha was back up and running, too.

  11

  AS COINCIDENCE WOULD HAVE IT, not long before Operation Enormoz was secretly reactivated, at Arlington Hall the orders were issued, with similar stealth but a lot less importunity, to get the wranglers back working on the Blue Problem. This decision, just like the sudden one that had shut it down, seemed more a bureaucratic whim than a well-considered strategy. Nevertheless, the Army Security Agency got moving again on its attempts to break the Russian code.

  Gene Grabeel, to her mystification, was abruptly pulled from her duties on the German desk and reassigned to the Blue Problem. And now she was joined by eight new female recruits. Like the industrious Grabeel, most were former schoolteachers, but the youngest had a leg up on the rest: she had taken a correspondence course in cryptanalysis during her senior year at the Mississippi College for Women.

  The newly assembled unit went back to work sorting the intercepted Russian cable traffic—“discrimination,” the more veteran cryptanalysts called it—into trade, diplomatic, and military groupings. They worked in a room with two long tables and a shoulder-high wall of filing cabinets that separated them from the impressively large teams dealing with the German and Japanese codes. Under orders, they talked to one another only in a hushed whisper, but even this sort of communication was rare. They spent each day staring blankly at thin pieces of paper filled with typed blocks of numbers—an incomprehensible language. The impossibility of their task left them enveloped in a nearly terminal listlessness. Yet with a numbing regularity, new piles of intercepted cables kept arriving on their desks, as many as 4,000 in a single week.

  After eight dismal months, a frustrating period during which little more than a careful sorting of the Russian traffic had been accomplished, the team finally dared to ask for help. “The aim is to break the system and a staff of experts w
ould be of value to the unit,” it was tactfully suggested in a memo to the Arlington Hall brass.

  It was a tempered plea, but it was also fortuitously blessed in its timing: the wranglers had already broken the back of the German and Japanese codes; and, more incentive to move on, the end of the war was in sight. So perhaps that was why it bore such quick fruit. But for reasons that the Army never bothered to explain, a half-dozen or so experienced cryptanalysts, most with impressive academic credentials, were soon assigned to the Russian unit.

  And slowly, like the turning of a dial on a radio receiver, first tentatively, then more determined as it homed in on a signal, the new team began to make progress.

  LIEUTENANT RICHARD HALLOCK HAD AN orderly mind. One of the new additions to the unit, he had studied archaeology at the University of Chicago and, more because he enjoyed poking into puzzles than for any practical scholarly purpose, he’d taken it upon himself to translate texts from various ancient Babylonian dialects. The key to making headway, he’d discovered after all his late nights, was to focus on repetitions: a word or phrase appearing in one dialect would have approximately the same meaning in another. It was a tool he employed time after time to wrench open whole sentences. And as he began to grapple with the Russian cables—blocks of numbers that were, at first perplexed glance, more abstruse than any language conceived by the ancients—he was guided by his own self-taught wisdom: hunt down the repetitions.

  Only it was not enough. No sooner had he fixed on this strategy, than he realized he’d be setting off on a quest that was as unlikely to succeed as it would be unending; there were hundreds of thousands of pages, each filled with blocks of numbers. He needed to find a way to whittle down the odds. And that’s when Hallock had his second insight.

  The Russians—not unlike the stilted royals at the Babylonian court—often had standard, introductory doggerel at the start of each message. Time after time, they could be counted on to begin a cable with a formal phrase that, more or less, read, “Reference your message # . . .” And the sign-off was pretty formulaic, too.

  Sensing an opening, Hallock rolled up his sleeves and went to work. Arlington Hall had one of the first IBM processing systems, and he set the staff to preparing punch cards for the beginnings and endings of 10,000 messages; the bulk of them were back-and-forths between Moscow and Washington sent years earlier, in 1942 and in the winter of 1943.

  The chugging machine scanned the perforated rows and columns not unlike the way a player piano read musical scores. And the song the IBM wound up playing was music to Hallock’s ears: there were seven cases where the Russians, going against all their rigid coding procedures, had enciphered two unrelated messages with the same additive block of numbers.

  Could this have been a statistical accident, one of the vagaries of life, a happenstance akin to someone’s picking the exact Irish Sweeps number five years in a row? The more mathematically gifted of the cryptanalysts did the sums: the odds of one-time pad numbers—random additives spewed forth by a cavalier machine in Moscow Center—being repeated seven times were about a billion to one.

  And if that wasn’t sufficient proof that the Russians’ security procedures had indeed broken down, further scrutiny removed any lingering doubt. More duplicates, more examples of disparate Russian cables using the identical five-figure additives, were identified.

  Which should have been impossible.

  The entire point of one-time pads—and the source of their impenetrable security—was that they were used only one time. The cipher clerk sending the coded message added the random blocks of numbers taken from a page in his pad, and the recipient, using the same page, subtracted them. When each clerk was done, his page was to be incinerated, and that specific string of numbers would go up in smoke, gone forever.

  Unless, against all their training, all their knowledge, all their institutionalized instincts, the Russians in the early years of the war had distributed duplicates of their one-time pads to their embassies and consulates.

  Could that have happened? What had gone wrong back then?

  THE GERMAN ARMY WAS AT the gates of Moscow. It was November 1941, and with the bone-chilling winter blowing in off the Russian steppe, the ground quickly froze. The tanks of the Wehrmacht spearheads, fifty-one divisions that had already blitzkrieged their destructive way across the Soviet Union, prepared to advance. The armored units would encircle the city, and then charge forward in unison to crush Moscow.

  Stalin remained defiant, determined to rally his people. Against the admonishments of his generals, who feared Moscow would be left perilously exposed, he diverted troops to form a defensive line along the outlying cities of Klin and Tula. It was a gamble, and a desperate one. If the enemy broke through, Moscow’s residents would need to take up arms. Yet with grim inevitability, Stalin’s gamble moved closer and closer to becoming a losing bet. In the first icy week of December, the German tanks drove on, boldly splitting the Russian line of defense. “The enemy, ignoring the casualties, was . . . willing to get to Moscow by any means necessary,” Marshal Georgy Zhukov, the chief of the Soviet General Staff, would recall. Stalin, his confidence now teetering, had no choice but to confront his generals. Could they keep the Wehrmacht out of the city? Could they successfully defend Moscow? With the German Seventh Panzer Division just twenty miles from the walls of the Kremlin, the generals’ assurances were guarded, fraught with doubts.

  It was at this perilous time, as the German tank commanders could see the twisting spires of the Kremlin in their field glasses, that the harried cipher clerks in Moscow Center who produced the one-time pads decided that the moment had its own unique priorities. With the Motherland imperiled, with so many pressing demands on resources, they decided it was necessary to risk a small accommodation to the usual code-manufacturing procedures.

  In the past, the pages of random blocks of numbers rumbling out of the coding machines had employed a single sheet of carbon paper to make a single copy. They now used three sheets of carbon. As a result, the output of pads was tripled.

  With that bit of time-saving, the one-time pads became another casualty of the war. Since there were duplicates, the blocks of additives no longer formed an unbreakable wall around the encoded message.

  And a few years later the Arlington Hall code breakers would begin tearing the wall down, brick by loose brick.

  AS THE WRANGLERS PROCEEDED WITH their arduous work, in one more stroke of good fortune another of the new recruits helped move things along by finding an ingenious shortcut. Cecil Phillips, a gangly nineteen-year-old, was an improbable candidate for such a fruitful discovery. He had wound up at Arlington Hall purely by chance.

  A sophomore at the University of North Carolina, he had been cooling his heels at home for the summer when his mother suggested he’d better get a job. Obediently, he went off to the U.S. employment service in Asheville to see what the government had to offer a chemistry major. They had nothing, but suggested he speak to the Signal Corps lieutenant who was recruiting at the local post office. Phillips took a short IQ test, responded that of course he knew what the word “cryptography” meant (his parents had bought him a Little Orphan Annie Decoder Pin as a present for his eleventh birthday), and on the spot he was offered a job at Arlington Hall.

  He did not return to college when the summer ended. Instead, for nearly the entire uneventful year, Phillips worked as a civilian clerk tackling the Japanese weather problem. Then, in May 1944, he was transferred to the Russian unit. Armed with paper and a well-sharpened pencil, he spent six diligent months scouring the pages lined with columns of numbers hunting for—well, something. And then he found it.

  But his eureka moment would not have been possible without the interference of a Russian spy.

  ELIZABETH BENTLEY’S LENGTHY CONFESSION, THE same 107-page document detailing Soviet espionage in America that had helped spark Bob Lamphere’s mounting concerns, contained a single brief, almost throwaway, comment about a well-placed government officia
l. Yet it was a revelation that left G-2 officers at Arlington Hall riveted.

  “I recall one occasion,” she had written, when one of the Soviet assets in the Washington ring, whose microfilms she’d been carrying to New York in her knitting bag, told her “he had informed Moscow that the United States was on the verge of breaking the Soviet codes.”

  The information was overly optimistic. The code breakers were a long way off. Yet as things worked out, the message the spy conveyed to his Soviet handler became, in time, a self-fulfilling prophecy.

  On April 25, 1944, in response to the warning, Moscow Center sent an urgent directive out “to all Residents.” As of May 1, the indicator system that showed the specific page from the one-time pad book the sender was using—the “key page,” the wranglers called it—would be changed. The old two-digit key indicator at the start of every message would be replaced by a five-digit group taken directly from the page being used.

  It was a tricky maneuver that Moscow Center thought would put the American code breakers off the scent. And it succeeded. For six months the Russian unit moaned that the sudden change in Soviet coding procedures had left them stumped.

  But then Cecil Phillips noticed something. He realized the Russians had done them a favor.

  “TOO MANY SIXES”—PHILLIPS EXCITEDLY ANNOUNCED to the rest of the outfit. At the start of some of the recent traffic sent from the Center, he studiously pointed out, there was a block of numbers that contained more sixes than seemed random. He had kept staring stubbornly at this oddity, and then it struck him: these numbers were not strings of enciphered code, but were naked, raw (en clair, as his more experienced colleagues put it) numbers. But why?