The Billion Dollar Spy: A True Story of Cold War Espionage and Betrayal

by David E. Hoffman

  In the note, Tolkachev returned to the contentious topic of money, but he was apologetic. He said there had been misunderstandings, “my mistake.” He acknowledged creating confusion when he “started talking about a six digit figure when I had in mind six zeros.” Tolkachev explained that he was certain Belenko had received $6 million and that had influenced his thinking, but he had never demanded a specific sum. He said he would let the CIA decide how much to pay him, based on the value of his espionage. His expectations for a large paycheck were still high. He asked the CIA to put his money in hard currency in an escrow account, as long as he could withdraw it at any time.22

  Guilsher expressed sympathy with Tolkachev on the money question in a note to headquarters. “We can easily imagine the difficult position CKS finds himself in when negotiating due reward for his efforts,” Guilsher wrote. “He obviously fears he will be taken advantage of (large government vs. one helpless individual) when he has committed treason and already passed invaluable materials to us.” Guilsher added, Tolkachev was still talking “in terms of millions.”

  That spring, the air force reported that Tolkachev’s material had dovetailed with other intelligence on Soviet weapons systems, and much of the Tolkachev material was quite damaging to Moscow. The air force said Tolkachev had provided a wealth of science and technology information, and his primary value was providing a detailed picture of new Soviet weapons systems that would not be available from other sources for many years, if ever. In March, Guilsher and Gerber raised the question with headquarters of what Tolkachev was really worth. In a cable, they pointed out that previous evaluations from the military had been glowing. The evaluations had praised Tolkachev’s materials as the “first information” about some weapons, as the “only information,” “first hard information,” and declared, “Time saved on research and development of U.S. countermeasures to these systems has been reduced by minimum of 18 months, for one system as much as five years.” Another called Tolkachev’s intelligence a “gold mine” and “one hundred and eighty degree change in seventy million dollar project.”

  Guilsher and Gerber inquired if the intelligence community could put a dollar value on it all:

  For instance, how much can we expect to save on R and D? Have we discovered vulnerabilities in our systems that we can now correct? Can we develop new countermeasures against Soviet systems? Do we have true picture of Soviet capabilities in field? Has CKS provided information on systems for which we had no data before?23

  The answer to all the questions was that Tolkachev had provided intelligence that was, as headquarters put it, “essentially invaluable.” Headquarters added, “We suspect cksphere himself, fully aware of the extraordinary value of his materials, realizes we cannot pay him the exact amount the production is worth, even if we could calculate it.”24

  Next, headquarters sent the Moscow station an internal CIA evaluation of Tolkachev’s materials. The internal evaluation declared, “The timeliness of these reports is especially significant in terms of savings to the intelligence community. With other systems such detail and understanding are not obtained until years after deployment. The definitive data in these reports will save many years of analysis and debate in the intelligence community.”25

  Although the 150,000 rubles that Guilsher had given Tolkachev in December 1979 quieted his anxiety, the CIA had never really settled on how much to pay him over the long haul. Tolkachev was waiting for their decision. In March 1980, Gerber wrote to Hathaway, now the Soviet division chief, saying that before the next meeting with Tolkachev they would have to “make some realistic plans for future” and “decide on commitments we will live up to.”26 The Moscow station suggested a total compensation package of $3.2 million. It was still a tiny fraction of the value of Tolkachev’s intelligence to the U.S. government, but much more than he had been given so far.

  Hathaway knew well how troublesome the question had become; he had wrestled with it himself as station chief. In early April, Hathaway responded, “The problem is basically one of not so much how much his information is worth, rather how much can he handle, and frankly how much can we reasonably pay him. I say reasonably since, if we took the total value of his information, we would frankly be talking about astronomical figures. I do not mean to be cruel, but we are not a business and we can compensate him reasonably without paying the maximum.”

  Hathaway also wanted to slow down Tolkachev. “We here are obliged to do everything to slow him down,” he wrote. “A reasonable sum will, I think, help slow him down, whereas large (indeed justified) sums may egg him on even more.” While the CIA would build up Tolkachev’s account, “we do think your final figure of 3,200,000 is high.”

  Soon after writing this, Hathaway told Gerber he would go to the CIA director with a plan to pay Tolkachev $200,000 for his work through 1979, $300,000 for 1980, $400,000 for 1981, and half a million dollars for every year thereafter. It would be, he said, “the highest salary ever paid to any single individual in the history of this organization.”27

  Once again, Turner put on the brakes. On May 10, 1980, he approved a pared-down package: $200,000 for 1979 and $300,000 for every year after. Hathaway told Gerber and Guilsher that he was sure they would be disappointed. “I share your disappointment,” he said. But “it is an enormous amount of money.” He expressed confidence that Guilsher could “use his persuasive talents and his special relationship with cksphere to get this across.”28

  Another factor crowding in on Guilsher was exfiltration. Tolkachev had earlier said he would never consider leaving the Soviet Union. But in his February ops note, he suddenly expressed interest in the idea. “I never thought about the possibility of exiting the USSR,” he said, but “if a realistic possibility exists for exfiltrating me and my family, then, no matter what the inherent risk connected with it, I would like to take advantage of this possibility. My family does not know about my activity.” Tolkachev asked for more details—how long would they have to prepare?29 Guilsher felt this new interest in exfiltration was a signal that Tolkachev was growing pessimistic about his future in Moscow.30 Perhaps if the CIA gave a positive nod to exfiltration, Guilsher thought, it would ease Tolkachev’s demands for millions of dollars. At headquarters in May 1980, Turner approved giving Tolkachev “the commitment to exfiltrate him, his wife and his son if and when circumstances dictate, with resettlement in the West.”31

  Guilsher began to draft the ops note he would give Tolkachev at their next meeting. He knew that Tolkachev had been shattered by the decision on the L-pill. Now he had to talk Tolkachev through more unpleasant news. Guilsher had “absolutely no doubt” that Tolkachev would be “highly shocked and unhappy” with the pay package.32 Guilsher tried to break the news gently. “The evaluation of your materials is very high and it has been decided to grant you the highest salary that our organization has ever paid,” Guilsher wrote, laying out the sums that Turner had approved. There were a lot of zeros, but not millions. Guilsher wanted to “cushion” the blow by promising exfiltration, but that was also problematic. By merely discussing it, they would raise Tolkachev’s expectations for departure. The CIA did not want to exfiltrate Tolkachev unless there was an emergency. They wanted a productive, deeply embedded agent like Tolkachev to remain in place as long as possible. Guilsher didn’t promise anything swift. He reminded Tolkachev that he would have to tell his family about exfiltration, and if they were to hesitate, “problems may arise.” He outlined for Tolkachev what the CIA would do for him in terms of resettlement in the United States.

  The Tolkachev operation was now growing quite complex—the issues of exfiltration, money, photography, security, and the suicide pill were all interlocked. The Moscow station and headquarters chewed over each of them in detail, with cables back and forth, but every move had the potential for error and to upset Tolkachev, as happened with the suicide pill. The CIA had attempted to manage Tolkachev through carefully calibrated offers of money and e
xfiltration, but looming over it all was the reality that Tolkachev did not always listen to his handlers. He had an unshakable determination to steal as many secrets as he could. He did not want to be slowed down.

  Tensions between the Soviet Union and the West deepened considerably in the spring of 1980. The United States boycotted the Moscow Summer Olympics in response to the invasion of Afghanistan, sharply offending the Soviet leadership. When the dissident physicist Andrei Sakharov spoke out against the invasion, he was detained and exiled to Gorky.33 The Olympics were scheduled to open in July, and for weeks beforehand the streets were crawling with extra militiamen, many of them brought to Moscow from the provinces. All this made John Guilsher’s next move even more risky and intricate.

  On the evening of May 12, Guilsher put on a disguise. It was his first experience with the procedure known as identity transfer, a trick that played on a KGB weakness. Although the KGB scrutinized all Americans in Moscow, they could not surveil them all, so they ignored many who were considered ordinary workers, not involved in intelligence. A CIA officer could disguise himself as one of the embassy workers the KGB had ignored and slip out of the compound without attracting attention. The trick worked. Guilsher escaped the compound without the KGB’s spotting him. He was hoping to make an impromptu call to Tolkachev followed by a meeting on the street. But when Guilsher called, he heard a voice that was not Tolkachev’s and aborted the attempt.34

  The next attempt to avoid the KGB was an elaborate ruse. Guilsher notified the Soviet authorities that he planned a short trip abroad. He flew out of Moscow as planned, but then he abruptly returned, earlier than he had notified them. He hoped to meet Tolkachev before the KGB realized he was back. But when Guilsher got to passport control on his premature return, he noticed special attention was being paid to him, and he decided not to go ahead with the meeting.

  Two missed meetings and blanket surveillance on the streets did not deter Gerber. He wrote a cable to Hathaway after Guilsher’s second attempt failed on May 21. Although “not optimistic” about a natural break in surveillance, Gerber said it was essential for Guilsher to meet with Tolkachev once more. Guilsher was due to end his Moscow tour that summer. With the delicate issues of money and exfiltration, Gerber cabled, “we prefer have cksphere discuss these critical matters with familiar face rather with total stranger in fall.” Although Gerber believed that the station’s case officers should be interchangeable in operations, Guilsher was the only case officer he could send now, the only one known to Tolkachev.35

  The two failed attempts to meet with Tolkachev in May led to an important new dimension in the operation. The CIA’s prowess with technology was growing, and it dreamed of using electronics to evade the KGB and carry out espionage unfettered. Through the 1960s, communications with agents in hostile areas were largely carried out with a small number of proven techniques, like secret writing, microdots, radio broadcasts, and dead drops. But starting with the 1970s, thanks to the revolution in microelectronics, technology began to transform the way case officers communicated with their agents, and the CIA attempted to maintain a cutting edge. An example of this was the first in a family of small electronic devices known as the SRAC, for short-range agent communications. An early version of the device, called Buster, had been given to Dmitri Polyakov, the general in Soviet military intelligence, the GRU, who volunteered in New York in 1961 and was code-named tophat by the FBI. Later, he was run by the CIA in Rangoon and New Delhi and now was back in Moscow, where he rose to become commander of the GRU training academy. The new device, it was hoped, would make it easier for Polyakov to communicate with the CIA and avoid the KGB. The device was a handheld communications system, which consisted of two portable base stations—each about the size of a shoe box—and one agent unit that could be concealed in a coat pocket. With a tiny keyboard one and a half inches square, the agent would first convert a text message into a cipher code, then peck the code into the tiny keypad. Once the data were loaded—Buster could hold fifteen hundred characters—the agent would go somewhere within a thousand feet of the base station and press a send button. The base station could be moved around, placed in an apartment window, or a car; the agent would have to be told approximately where. In essence, Buster was a primitive text messaging system. Its advantage was safety: the agent could communicate without actually meeting a case officer on the street. However, there were problems with using such equipment in the field. Buster was clearly spy equipment and would fatally compromise an agent if caught with it.

  Hathaway, now division chief, was a technology enthusiast, eager to deploy new gadgets to frustrate KGB surveillance. Ideally, Buster could ease the risk of personal meetings with an agent in parks and on dark street corners. “Think of how we’ll save this poor bastard from the danger of going out and meeting somebody from the agency,” Hathaway said, a reference not only to Tolkachev but to any CIA agent who could make good use of the gadget.

  Over time, Buster underwent major improvements. The next model was called Discus. It was also handheld but much easier for the agent and the case officer to use. The Discus eliminated the need for the bulky base station and could transmit to a case officer holding a second small unit hundreds of feet away. Most significantly, Buster had required the agent to encrypt his message by converting it to cipher code, painstakingly by hand, before typing into the device, while Discus used automatic encryption. The keyboard was larger, and it could transmit significantly more data. Moreover, Discus had a verification system so the agent knew the message had been received. The Discus was way ahead of its time; there were no consumer handheld devices available then, nothing remotely like the BlackBerry or the iPhone.36

  In June, Hathaway urged the station to consider giving a Discus unit to Tolkachev. He said Tolkachev could use the device to signal the CIA when he was available for a meeting and perhaps “select important portions of documents and transmit them to us.” He speculated that Tolkachev could use Discus to alert the Moscow station about where to pick up film and other materials in a dead drop. Hathaway expressed confidence the Discus would “enhance security and production of this operation.” The Discus beckoned as a kind of invulnerable magic carpet that would soar over the heads of the KGB. While the traditional method of dead drops usually took a day or longer to signal, emplace, and collect, the electronic communicator could transmit urgent intelligence almost instantly.37

  Gerber, however, thought the Discus was ill-suited for Tolkachev. He had experimented with the device and knew that operating it was far more complicated than just pressing a button. He tested the Discus in a Moscow vegetable market while looking over cucumbers and tomatoes as his wife, Rosalie, held the second unit in another section of the market. For an exchange, the Discus required both sender and receiver to be static. Gerber tried to send a signal. He immediately realized that an agent would have to be looking down into his pocket until the red verification light flashed, or else he would not know the message went through. The red light flashed only after a pause. Was peering into one’s pocket, watching for the light to flash, the kind of body language that would give away an agent? Was it worth the risk? Moreover, Gerber surmised that the agent would have to give some kind of warning to the CIA that he was about to transmit on Discus. That signal was another operational act. Also, sites had to be selected in advance for Discus transmissions and tested; the radio waves tended to bounce around, and not all locations were suitable. What’s more, the testing also put the signal in the air, briefly, where the KGB might notice, and it required case officers to be exposed. Was it worth the risk?

  Gerber doubted that Discus could ease the security dangers for Tolkachev and thought it might bring new ones. “Do not think now is the time to discuss topic with him,” Gerber responded to Hathaway. He added, the “value of cksphere production is in voluminous reproduction of entire detailed documents, not just in few tidbits” that could be transmitted by Discus. Gerber insisted that they continue
personal meetings for the large amount of material Tolkachev was delivering. Gerber wanted a sure, steady process. He said that giving Tolkachev a Discus might spur him to speed up and take risks. He also pointed out that Tolkachev’s own preference was for meetings, not dead drops. Tolkachev “has strong psychological need for direct personal contact.”38

  On June 11, Gerber sent a second strong cable to Hathaway, saying he had raised using the Discus with the case officers at the Moscow station. “We conclude that in this operation we have more to lose than to gain,” he wrote. The device wasn’t very useful in sending or receiving complex messages, such as those concerning “requirements,” or what secret materials Tolkachev might gather. Moreover, the Discus would bring Tolkachev and CIA case officers into frequent and close proximity to each other on the street, which the KGB might notice as a pattern. To evade the KGB, different sites would have to be selected for every transmission. It just wasn’t worth it. Gerber thought to himself that perhaps Turner would like to eliminate human intelligence and just rely on technology, but it could not be done.

  They needed to look the agent in the eye, and Tolkachev needed to shake the hand of a case officer he could trust.39

  9

  The Billion Dollar Spy

  On June 17, 1980, Guilsher put on a disguise once again and headed out to the street. The KGB watchers didn’t see him; the disguise worked exactly as intended. After a surveillance detection run, he met Tolkachev at 10:55 p.m. for their seventh rendezvous. The sky was luminescent. In good spirits, Tolkachev told Guilsher he had not encountered any difficulties. He was relieved they could meet before his summer vacation because he had been very busy. He had good news: in February, the tightened security arrangements had suddenly been abandoned. Once again, it was no longer necessary to deposit the building pass when checking out documents. The reason was a bureaucratic logjam: the clerks in the department, mostly women, were swamped with all the building passes coming in and could not get out for lunchtime, a break when they usually searched for food and goods. So the director of the institute went back to the old rules: documents could be taken in the morning and brought back in the late afternoon.1