The Wizards of Langley

by Jeffrey T Richelson

  The optical system of the KH-11 (and its successors) scanned its target in long, narrow strips and focused the light onto an array of CCDs with several thousand elements. The light falling on each CCD during a short, fixed period of time was then transformed into a proportional amount of electrical charge. In turn, the electrical charge was read and fed into an amplifier, which converted the current into a whole number, between 0 and 256, representing a shade of color ranging from pure black to pure white. Thus each picture was transmitted as a string of numbers—one from each element.20

  More specifically, the CCD captured particles of energy, visible light, in an array of picture elements known as pixels. The pixels automatically measured the intensity of the particles and then would “send them on their way in orderly rows until they are electronically stacked up to form a kind of mosaic.” The standard CCD used in the Hubble Space Telescope has a total of 640,000 pixels arranged in an 800-by-800 format and occupies less than half a square inch.21

  The KH-11’s charge-coupled devices could not, however, do the job alone. Without a good mirror in front of them, even the best CCDs produce photographs with poor resolution. But the mirror for the first KH-11 was quite good and quite large—seven feet, eight inches wide. (Subsequently, mirror size increased). The secondary mirror, greater than one foot in diameter, narrowed the image coming off the primary mirror and sharply focused it.22

  Another key to the KH-11’s ability to produce high-quality photographs was its computer. About the size of a sleek VCR, the computer was fundamental to maintaining the KH-11 in a stable position, pointing the mirror and obtaining photographs of the desired targets.23

  Once the visible light was collected and transformed into an electrical charge, the signals were then transmitted to one of two Satellite Data System (SDS) spacecraft as the relay spacecraft passed slowly over the northern Soviet Union. The SDS orbit was identical to that first employed by Soviet Molniya satellites. Orbiting with a 63-degree inclination, the satellite approached to within 250 miles of earth when passing over the Southern hemisphere, and it moved as far away as 24,000 miles as it drifted over the Northern Hemisphere. A spacecraft in such an orbit took eight to nine hours to pass over Soviet territory, leaving it available to receive and transmit imagery for long stretches of time.24

  The SDS spacecraft, the first two of which were launched in June and August of 1976, performed a variety of functions. In addition to transmitting the KH-11 digital signals, they relayed communications to any B-52s flying on a polar route, served as communications links between the various parts of the Air Force Satellite Control Facility, and carried nuclear detonation detection sensors. Those other functions, Bud Wheelon noted, were “strictly a sideshow,” a nice bonus if they worked but a minor loss if they did not.25

  The SDS satellite then transmitted the KH-11 signals for initial processing to a ground station at Fort Belvoir, Virginia, about twenty miles south of Washington. The Mission Ground Site—a large, windowless, two-story concrete building—was given the cover title of Defense Communications Electronics Evaluation and Testing Activity and also was designated as Area 58.26

  Its method of transmitting data permitted the first KH-11 to remain in orbit for over two years—770 days. The next three lasted 1,166, 973, and 1,175 days, respectively.27 The KH-11 was not limited, as were the KH-8 and KH-9, by the amount of film that could be carried on board. In addition, it had a higher orbit, approximately 150 by 250 miles, that reduced atmospheric drag on the spacecraft.

  The new system had one serious initial limitation. Although it could transmit its data instantaneously, it could do so only for two hours per day. So much power was required to transmit the data to the relay satellite (via the KH-11’s traveling wave tube amplifier) that the system drained power far faster than it could be replaced by the satellite’s solar panels. Thus, the new model of the spy satellite fleet could be used only sparingly at first.28

  The KH-11 operated along with KH-8 and KH-9 satellites for several years before becoming America’s sole type of photographic reconnaissance satellite. Only a few, very select government officials were permitted to know of the KH-11’s existence or even see its product. The KH-11 was treated with even greater secrecy than usual in the black world of reconnaissance satellites—the photographs and data derived from them were not incorporated with data from the KH-8 and KH-9 systems. The decision to restrict the data to a very small group of individuals was taken at the urging of senior CIA officials, including COMIREX chairman Roland Inlow, but it was opposed by military officers who wanted the information to be more widely distributed throughout the armed forces.29

  Among the officials who did know about KENNAN in early 1977 were president-elect Jimmy Carter and his national security adviser, Zbigniew Brzezinski. The latter had been briefed extensively on the new system after Carter’s election in November. On December 30, 1976, Enno Henry “Hank” Knoche, the Deputy Director of Central Intelligence, along with John McMahon and other CIA representatives, met with Brzezinski. As part of the briefing, which covered human and technical collection operations, McMahon described the KENNAN system “at some length.” Knoche suggested to Brzezinski that he think of the new real-time capability in the context of new approaches to crisis management. He noted that a KH-11 could be tasked by crisis managers, “made responsive to live needs,” and be the “basis for re-thinking the organization of current crisis management.”30

  In the month between its launch and Carter’s inauguration, the KH-11 underwent checkout and testing and that day finally transmitted its first photos. By this time, Knoche was the CIA’s acting director, pending confirmation of the new President’s choice for DCI. George Bush had wanted to remain as DCI, but Carter would not extend the stalwart Republican’s tenure, not even until his own choice could be confirmed. Until that happened—and confirmation took longer than usual because Carter’s first nominee, former Kennedy White House aide Theodore Sorensen, withdrew when it became clear he could not win approval—Knoche was head of the world’s most technically accomplished intelligence establishment. Among Knoche’s first roles was to show Carter the capabilities of America’s newest spy satellite—one way of demonstrating the CIA’s value to the new President.31

  The most dramatic demonstration would have been for Carter to see the photos within moments of their arrival. But he was busy on Inauguration Day—taking the oath of office, strolling down Pennsylvania Avenue in subfreezing weather, and attending the various traditional celebrations. In addition, to have the most dramatic effect would have required either that Carter make the trip to Ft. Belvoir or that Ft. Belvoir come to Carter, neither of which was feasible. As a result, Knoche decided to wait a day before visiting the new President.32

  So it was 3:15 in the afternoon of January 21 when Knoche and Admiral Daniel J. Murphy, the DCI’s deputy for intelligence community affairs, began a fifteen-minute meeting with Carter and Brzezinski in the White House’s second-floor Map Room. Knoche had a handful of six-inch square black-and-white photos with him.33 McMahon had told Carter what the KH-11 could do; now Knoche would show him. Carter examined the photographs that Knoche spread on the map table. The photographs did not reveal some secret, nefarious activity on the other side of the world, but provided an overhead perspective of something much closer to home—Carter’s inauguration. After peering at the photos for a few moments, Carter looked up at Knoche, grinned, and then laughed appreciatively. He congratulated Knoche and Murphy on the apparent quality of their latest reconnaissance system and requested Knoche to send over some more samples for the next day’s National Security Council meeting, his first as President. “Of course,” Carter said as he turned to Brzezinski, “this will also be of value in our arms control work.”34

  Carter and Brzezinski knew they had something of immense value. Indeed, it was more obvious to them than to some in the CIA, for it was the president and his advisers who might be pressed in a crisis to make crucial decisions that could dramaticall
y affect the fate of the United States. Now they would be able to make those decisions with timely information.35

  PROSECUTION WITNESS

  By 1977, Leslie Dirks had been with the CIA for sixteen years. He had participated in countless meetings to discuss U.S. satellite reconnaissance activities—meetings that were conducted under strict security guidelines, often in vault areas. But in April 1977 and November 1978, Dirks would appear in a quite different setting—U.S. District Court. In each case, he would be the key prosecution witness in a trial that resulted from the sale of secrets about one of the CIA’s greatest technical accomplishments. The buyer was the Soviet Union. The secrets concerned RHYOLITE and the KH-11.

  The sequence of events that led to Dirks’s first court appearance began on July 29, 1974, when twenty-one-year-old Christopher John Boyce began work at TRW as a general clerk at a salary of $140 a week. On November 15, he was briefed on RHYOLITE, described by the briefer as “a multipurpose covert electronic surveillance system.”36 He was also told about the PYRAMIDER and ARGUS programs.

  The revelations were necessary because Boyce had been assigned to work in TRW’s “black vault”—a bank-style vault with a three-number combination and an inside door with a key lock. Within the vault, he monitored secret communications traffic relating to various CIA-TRW satellite projects. Less than six months after joining TRW, Boyce was using a boyhood friend, Andrew Daulton Lee, to sell the vault’s secrets to the KGB. In April 1975, Lee, who was more familiar with peddling marijuana than crypto cards, walked through the front door of the Soviet embassy in Mexico City and handed a typewritten note to the first official he encountered. It read, “Enclosed is a computer card from the National Security Agency, crypto system. . . . If you want to do business, please advise the courier.”37

  The KGB, naturally, was interested. And Boyce had a wealth of material to choose from, with fifty to sixty messages a day passing through his hands, messages kept on file for a year. Altogether, Lee would make seven trips to meet with KGB officers. On March 15, 1976, he arrived in Vienna and delivered ten rolls of film containing a month of ciphers; RHYOLITE communications traffic among TRW, the CIA, and Pine Gap; and a thick technical report on the proposed ARGUS system.38

  In early October 1976, Boyce joined Lee, for the first time, in Mexico City. Once there, it became apparent that as much as KGB officials valued the intelligence Boyce was providing on U.S. satellites, they thought he could be more useful elsewhere. The KGB would be willing, Boyce was told, to provide $40,000 to pay for college and graduate school. They envisioned that Boyce would become a Soviet or Chinese specialist and find a job with the State Department or CIA. Before the month was out, Boyce had applied for admission to the University of California at Riverside.39

  But the KGB’s plan to turn Boyce into a mole was torpedoed by Lee’s impulsive behavior. On one occasion, such behavior had led his KGB contact, Boris, to pack Lee in a car and toss him out onto the road. But the lesson did not stick. On January 6, 1977, eager to get money to purchase drugs from a Mexican supplier, Lee tried to get Boris’s attention by throwing a Spanish-American dictionary, on which he had marked “KGB,” onto the embassy grounds. Fearing that Lee was a terrorist and the dictionary a bomb, the Mexican police immediately arrested him and discovered a sealed envelope with microfilm strips inside the dictionary. The strips contained 450 frames concerning PYRAMIDER.40

  Within a few months, Boyce and Lee were standing trial, though separately. On April 20, 1977, Leslie Dirks began his testimony in a Los Angeles courtroom, which would conclude the prosecution’s case. After eliciting a detailed description of his background, the prosecution went on to establish Dirks’s authority to classify projects, including PYRAMIDER. Dirks described the origins and some of the details of the PYRAMIDER studies and his decision that they should be classified as top secret.41

  He also testified that even though no operational system ever emerged from the studies, U.S. intelligence would sustain significant damage if the Soviet Union had access to information about PYRAMIDER. It would, the science and technology chief argued, provide the KGB with valuable information about the state of covert communications technology as well as the options and strategies that had been considered.42

  Such information could be used to compromise the communications of both agents and emplaced sensors. Such communications were vital, for the “plans and intents [of foreign governments], in the hands of an agent, does no good. Those plans and intents must be communicated and frequently it’s important to do so rapidly to ourselves, the CIA, and ultimately to the President of the United States so that he may be forewarned of a plan or an intent to initiate . . . hostilities in some part of the world.” Without a link between U.S. agents and the President, “the national security would indeed be exposed to great damage.”43

  Dirks’s testimony for the prosecution concluded with his observation that among the reasons the PYRAMIDER studies needed to be classified was that “they reveal the requirements for covert communications which are still . . . valid requirements.” Then, according to author Robert Lind-sey, “in more than four hours of cross-examination, [defense attorney] George Chelius did not knock any significant holes in Dirks’s testimony. He had left a solid impression with the jury that Chris’s actions involved grave harm to the United States.”44

  That impression undoubtedly carried great weight with the jury, which returned a guilty verdict. On July 18, Lee, who had also been convicted, received life imprisonment. Boyce, who had been granted the clearances and had the responsibility for protecting the information, received forty years. Throughout the trials, convictions, and sentencing, there was not a word uttered about RHYOLITE or ARGUS, or the fact that based on Boyce’s revelations the Soviet Union began encrypting the telemetry that RHYOLITE had been intercepting, topics the CIA and NRO considered too sensitive for discussion in open court. Further, Boyce’s and Lee’s disclosure of the PYRAMIDER material was enough to seal their fate. At the time, the very existence of the SIGINT satellite program was highly classified. Out of public view, at least one change was made—the code name RHYOLITE was replaced by AQUACADE.45

  In November 1978, Dirks was back in court, again as the result of the compromise of a key satellite system. But rather than being in Los Angeles in early spring, Dirks was in Indianapolis in early winter. This time the defendant was a CIA rather than a contractor employee. And the trial revolved around the key secrets that had been divulged. Finally, the recipient was not the KGB but the GRU—the Chief Intelligence Directorate of the Soviet General Staff.

  On trial was William Kampiles, who in March 1977 had begun his brief and undistinguished CIA career as a watch analyst in the Operations Center. Watch analysts, although junior officers, received information from the full range of intelligence community sources—CIA agent reports, KEYHOLE imagery, NSA communication intercepts, as well as Defense attaché and Foreign Service reports. They monitored the incoming intelligence reports from around the world and routed them inside the agency.46

  Among the documents provided to help them understand the significance of the incoming data was Copy 155 of the KH-11 System Technical Manual. Once someone was within the vaulted confines of the watch office, there was no further restriction on access to the manual, and it was not locked up. Its normal resting spot was on a shelf alongside a copy of an almanac in an unlocked cabinet, known as the CONSERVA file, beneath an ordinary copying machine.47

  Despite access to such documents, Kampiles found the work tedious. He had envisioned using his fluent Greek in clandestine work overseas. But his undistinguished early record in the agency did not stimulate interest in him within the operations directorate. As his performance deteriorated, all hopes of a transfer disappeared. An operations center supervisor recommended that Kampiles be fired before his probationary period was up. He pawed the women in the office, bragged of his sexual exploits, and “stood out markedly from the rest of the people; he was a bullshitter.” In November 197
7, after only eight months in the CIA and receipt of a formal letter indicating dissatisfaction with his work, he resigned.48

  Before departing for good, Kampiles pilfered the KH-11 manual, which was not regularly inventoried. One day, he stuffed a copy of the manual into his sports jacket and carried it out of the building. Doing so was not particularly difficult. Removal of classified documents from the agency by employees who wanted to do some work at home had reached such proportions that DCI Stansfield Turner had sent a memo to agency employees ordering a halt to the practice.49

  On February 19, 1978, Kampiles left for Greece. Three or four days later, he went to the Soviet embassy and offered to deliver documents. He provided a GRU officer with two to three pages of the KH-11 manual—its table of contents, summary, and an artist’s conception of the satellite. At a second meeting, Kampiles demonstrated his lack of skill as a covert operator—turning over the rest of the manual for a mere $3,000.50

  Back in the United States, having received guidance from the GRU about topics of interest and instructions concerning further meetings, Kampiles was anything but discreet. On April 29, 1978, Kampiles sat on a bench outside CIA headquarters telling George Joannides, a friend and employee of the General Counsel’s office, how he had conned the Soviets out of $3,000—apparently believing that would make him more attractive to the operations directorate he still longed to join.51

  That talk was followed by a letter Kampiles wrote to Joannides at the latter’s suggestion, which he then passed on to an officer in the Soviet division, Vivian Psachos. She suggested that Kampiles, who was in Indiana, return to Washington to discuss the matter. Attending the discussions on August 14, along with Kampiles, were Psachos, the FBI’s Donald Stukey and John Denton, and Bruce Solie of the CIA’s Office of Security. Those officials were aware of the CIA’s conclusion that the KH-11 had been compromised and also knew, from a GRU officer, of the sale of the satellite manual to the Soviets in Greece and when it was sold—as well as when Kampiles had been in Greece. They were also aware that the Soviets never paid money for promises, only for hard information or documents. Thus, they were more than skeptical about Kampiles’s claim of having swindled the GRU.52