The Wizards of Langley

by Jeffrey T Richelson

  The next day, in an interview at the Washington field office, James Murphy told Kampiles he didn’t believe him, and that if he didn’t believe it, “nobody in the whole world would believe it.” Kampiles, who had failed two polygraphs, slumped down in his chair and buried his head in his hands. After several seconds, Kampiles looked up at Murphy. “You’re right,” he said. “I didn’t get the $3,000 for nothing. I sold them the document.”53

  Less than three months later, Kampiles was facing a jury, charged with espionage. The trial was the product of a Carter administration decision to prosecute espionage cases, even at the risk of further disclosures.54 In addition to the testimony of Murphy and Stukey, who recounted their interrogation of Kampiles and his ultimate confession, the prosecution relied heavily on the testimony of Leslie Dirks.

  Dirks told the jury that U.S. national defense could be “seriously harmed” if the Soviet Union had access to the KH-11 manual. He said knowledge of the manual would suggest ways that the Soviet Union could hide its nuclear and military capabilities from the satellite.55 He explained that the “KH-11 system is a photographic satellite with associated ground facilities for controlling [the] satellite and distributing its products” and “one of the principal intelligence collection sources used to verify that the Soviet Union is indeed living up to the terms of their [SALT] agreement with the United States.”56

  Disclosure of the top-secret manual to a hostile foreign power, Dirks testified, would do serious harm to national defense. The manual described “the characteristics, capabilities and limitations of the satellite” and described the “process of photography employed by the KH-11 system and illustrates the quality of the photos and the process used in passing the product along to the users of the system.” Further, the manual detailed the “responsiveness and timeliness in the delivery of the ‘product.’” Page eight of the sixty-four-page manual described the satellite’s “limitations in geographic coverage.”57

  In addition, possession of the manual would “put the Soviet Union in a position to avoid coverage from this system. For example, by rolling . . . new aircraft [under development] into hangars when the system passes overhead, thereby preventing photographs of the new airplanes.” Knowledge of the quality of the photographs could enable the Soviets to devise “effective camouflage.”58 Compromise of the KH-11 system would, according to Dirks, cost the United States the advantage of being able to produce “accurate and current information” on Soviet capability for the U.S. president “in a time of crisis.”59

  The jury deliberated for just ten hours before returning their verdict—guilty on all counts. Two counts dealt with espionage and the defendant’s intent to injure the United States and carried the possibility of a life sentence. Other counts, carrying lesser penalties, concerned passing U.S. documents to unauthorized persons and the sale of U.S. documents valued at more than $100.60

  At his December 22 sentencing, Kampiles said, “First of all, Your Honor, I’m sorry for everything that has happened. Not at any time did I want to injure my country in any way. I only wanted to serve my country.” Prosecuting attorney David T. Ready suggested a “substantial sentence.” Kampiles “chose to casually disregard the safety and well-being of 200 million Americans.”61

  Judge McNagny sentenced Kampiles to forty years in prison stating that “This case is a complete tragedy for a young man who has never been in trouble before” but that “the United States has suffered a severe setback because of the sale to the Russians.”62

  MERGERS

  At the beginning of 1976, the Office of ELINT was not the only CIA component with a signals intelligence mission. In the operations directorate, Division D continued its mission of collecting COMINT from outposts in various U.S. embassies and consulates as well as from vans loaded with eavesdropping equipment. The primary mission of those outposts was “close support” of CIA operations—including monitoring the communications of the local security service.63

  But in February 1977, Division D and OEL were merged into the Office of SIGINT Operations (OSO), which would be part of the science and technology directorate. The consolidation may have been a means of eliminating the “constant battles” that former OEL chief Robert Singel recalled being fought by the two organizations. But according to Roy Burks, who became head of the OSO operations group when it was formed and then its director in September 1981, the merger was a friendly one, with no opposition from the operations directorate.64

  Heading up the new office during its first fifteen months was Edward Ryan, a veteran of the operations directorate whose previous assignments had included chief of station in Stockholm, chief of base in Berlin, and, until the merger, head of Division D. In May 1978, he was succeeded by another operations officer, David Barry Kelly, whose CIA service had included stints in Nepal, Vietnam, and Moscow.65

  Kelly would play a key role in the creation of a new and still classified secret signals intelligence organization, which merged the embassy eavesdropping operations of the DS&T and NSA. In 1976, Charlie Snod-grass, staff director for the House Armed Services Committee, conducted a study on U.S. SIGINT activities—and didn’t like what he found: too much duplication, not enough coordination, and lack of clear lines of authority in key areas. It had been the practice under Duckett for the CIA to conduct SIGINT activities as it wished, ignoring whatever edicts came from NSA.66 As a result of Snodgrass’s study, and the pressure from Congress that followed, the CIA was forced to acknowledge the NSA as the national SIGINT authority. NSA was quite willing to take over responsibility for all SIGINT. A memorandum of agreement between the two agencies followed, covering liaison, overhead collection, and a number of other subjects.67 In addition, agreement was reached for the CIA and NSA to merge embassy intercept operations. The CIA really had no choice, as the congressional oversight committees’ perception of too much overlap and competition between CIA and NSA embassy operations led Congress to cut off funding for the CIA operation.68

  The issue was somewhat more complex than that, according to Roy Burks. The CIA viewed the primary purpose of its embassy sites, not surprisingly given Division D’s origin, as to assist CIA stations and their officers in the field. For NSA, the mandate was different. It had no operatives in the field to support, so the focus of its intelligence activity was to support national and military policymakers.69

  The embassy operations, both CIA and NSA, had produced some valuable intelligence. In the late 1960s and early 1970s, the Moscow operation, code-named BROADSIDE, intercepted the radiotelephone conversations of Soviet Politburo members—including General Secretary Leonid Brezhnev, President Nikolai Podgorny, and Premier Alexei Kosygin—as they drove around Moscow. Traffic from the interception operation was transmitted to a special CIA facility a few miles from the agency’s Langley headquarters.70

  Originally, the conversations simply needed to be translated, since no attempt had been made to scramble or encipher the conversations. After columnist Jack Anderson disclosed the operation in 1971, the Soviets began enciphering their limousine telephone calls to plug leaks. Despite that effort, the United States was able to intercept and decode a conversation between Brezhnev and Minister of Defense A. A. Grechko that took place shortly before the signing of the SALT I Treaty. Grechko assured Brezhnev that the heavy Soviet SS-19 missiles under construction would fit inside the launch tubes of lighter SS-11 missiles, making the missiles permissible under the SALT treaty.71

  In general, however, the intelligence obtained, code-named GAMMA GUPY, was less than earthshaking. According to a former intelligence official involved in the operation, the CIA “didn’t find out about, say, the invasion of Czechoslovakia. It was very gossipy—Brezhnev’s health and maybe Podgorny’s sex life.” At the same time, the official said that the operation “gave us extremely valuable information on the personalities of top Soviet leaders.”72

  When the United States opened a liaison office in Beijing in late May 1973, CIA station chief James Lilley set up limousi
ne watches to monitor high-level meetings at the Great Hall and Zhongnanhai, the walled compound that has served as the seat of government in the Communist era. Subsequently, advanced eavesdropping equipment was brought in by diplomatic pouch and placed on the office’s roof. The equipment was used to monitor Chinese aircraft movements and intercept both military and civilian communications. It was not long before the liaison office, by monitoring the arrival of flights from the provinces, was able to determine when a Central Committee meeting was imminent.73

  Details of the merger of embassy intercept operations were worked out between Vice-Adm. Bobby Inman, who became NSA director in July 1977, and OSO chief Barry Kelly. It was agreed that the joint enterprise, to be called the Special Collection Service (SCS), would be initially headed by a CIA official who would serve a two-year term. The deputy director of the SCS would be selected from NSA, and an NSA official would become director after the CIA official completed his term. The director’s job would continue to alternate between CIA and NSA officials, with the director’s deputy succeeding him.74

  Things did not get off to a smooth start. Many in OSO were not happy with the formation of the SCS and didn’t want to be part of a group that would be managed by an NSA official half the time. Appointed to be the first head of the new service was Roy Burks. From NSA, Bill Black, a senior operations official, was selected to serve as deputy. Neither man, Burks later recalled, was involved in the bitter fighting that preceded the merger, a factor that gave them a better chance to make the new arrangement work.75

  According to Burks, there were people on both sides who seemed to want to make things difficult. Among the obstacles SCS faced was getting the CIA to courier documents to its College Park headquarters, forcing Burks to send a cleared secretary. The CIA did not want to send material to College Park because NSA people, at that time, were not polygraphed. Only when NSA began routinely giving polygraphs to its employees did the CIA feel comfortable in sending classified documents to SCS headquarters.76

  But by the end of 1983, joint CIA-NSA Special Collection Elements would be present in about a third of U.S. embassies abroad. The teams, which might consist of only two or three people, produced excellent intelligence, particularly if the embassy was located on high ground or near the foreign or defense ministries or other key offices in the capital. The sites were particularly effective in East European capitals.77

  ROCKET SCIENTISTS

  The early years of the operational KH-11 program also brought a further infusion of talented people into the development and engineering office, many of whom would go on to serve as senior office or NRO officials. Replacing Dirks as director of the development and engineering office was Donald L. Haas, an OD&E veteran who had been serving as director of ORD. He left in August 1978 to become deputy director of the NRO. Five months later, Haas’s position was filled by Bernard Lubarsky.78

  Lubarsky, who knew relatively little about the world of spies before joining it, was recruited from outside the intelligence community because the logical candidates to succeed Haas had left the CIA for jobs in industry and those remaining were not considered ready to assume the directorship. In 1947, after receiving his doctorate in mathematics (his bachelor’s and master’s degrees were in electrical engineering) from Case Institute of Technology, Lubarsky went to work at the National Advisory Council on Aeronautics (NACA) and became an employee of the National Aeronautics and Space Administration (NASA) when that agency was created in 1958 and absorbed NACA. While at NASA, he got to know Hans Mark, the director of the space agency’s Ames Research Center. In 1977, Mark became Under Secretary of the Air Force and director of the NRO, and in 1978 he recommended Lubarsky as a possible replacement for Haas. In January 1979, Lubarsky assumed command of the 500-person office.79

  His impression of the people who worked for him was that they were “just outstanding . . . no humpty-dumpties.” His deputy, Bert Aschen-brenner, who had served as acting director during part of the period between Haas’s departure and Lubarsky’s arrival, was a “very dedicated guy.” Heading up the office’s two satellite reconnaissance programs were Bob Kohler, who “was very aggressive [and] played hardball,” and managed the KENNAN program, and Julian Caballero Jr., who managed the AQUACADE (née RHYOLITE) program. Ed Nowinski served as Koh-ler’s systems engineering chief. All three eventually became directors of OD&E.80

  Many people who worked for Kohler and Caballero would go on to senior positions in the world of reconnaissance. In 1978, Jeffrey K. Harris, a twenty-five-year-old graduate of the Rochester Institute of Technology, transferred from NPIC to OD&E, eventually becoming manager of the KH-11 program and subsequently director of the NRO (1995–1997).81 Dennis Fitzgerald received a master’s from Johns Hopkins University in applied physics, mathematics, electrical engineering, and space technology. He joined the science and technology directorate in 1974 as a member of the OD&E Systems Analysis Group, working on developing new concepts for intelligence collection. In 1980, he became involved in collection systems procurement. Fitzgerald went on to hold senior positions in NPIC and the development and engineering and the research and development offices. In November 1997, he became the head of NRO’s signals intelligence directorate.82

  In 1981, David Kier left NASA, where he was responsible for “aeronautic interfaces” with the Defense Department and intelligence community, to join OD&E, where he would work on a highly secret aerial reconnaissance program (discussed in Chapter 8). In 1997, he became the NRO’s deputy director.83

  Modifications and improvements to the KH-11 and AQUACADE systems were part of the work of the rocket scientists. One proposal, made sometime early in the Carter administration, to modify the KH-11 system reflected the continuing competition between the Air Force and CIA. In response to a 1979 Air Force proposal to develop a dedicated radar imagery satellite to allow the collection of imagery at night and in the presence of cloud cover, OD&E proposed adding a radar imagery payload to the KH-11. The CIA suggested attaching a radar imagery payload developed by the Air Force, code-named QUILL, and orbited only once—in December 1964. The 1964 version returned its images in a recoverable capsule; the version proposed in 1979 would relay its data back to earth as the KH-11 did. The plan was opposed by Secretary of Defense Harold Brown, who objected to placing so many reconnaissance eggs in a single basket. The issue, with Brown and DCI Stansfield Turner on opposite sides, went to President Carter, who sided with Brown.84

  But the development wizards did create other improvements and modifications. When problems in the performance of an orbiting KH-11 were discovered, the models still on the ground were modified to avoid a recurrence. Before leaving the CIA, Jeffrey Harris led the successful effort to design an advanced version of the KH-11, which had higher resolution and greater swath width than the original model and thus could perform both high-resolution and area surveillance missions. Harris also oversaw the addition of an infrared imagery capability, code-named DRAGON, to the advanced spy satellite.85

  NEW ENEMIES, NEW ALLIES

  By the beginning of 1979, the TACKSMAN I facility consisted of four major units—a command center built into a hilltop, a radar antenna inside a thirty-foot-high dome, a radio monitoring device atop a steel tower, and a relay station pointed upward for communication with U.S. satellites. The radio monitoring device, built by Scientific Atlanta Inc., was a Pedestal Model 310—a device with four eight-foot-long arms studded with quill-like protrusions. The arms were almost joined at the front and pointed toward the Caspian Sea. Nearby was Scientific Atlanta’s “power with Pedestal Model 300-L,” a dish-shaped device fifteen feet high that pointed upward. There were 100 U.S. technicians who operated the equipment.86

  The TACKSMAN II facility at Kabkan, Iran, forty miles east of Meshed, was described by a CIA staff member as a twenty-first-century facility with advanced electronic equipment. The facility itself was in stark contrast to the surrounding area—a remote mountainous locale inhabited by nomads. However, it was not the loca
l environment that interested the CIA but rather the fact that Kabkan was only 650 miles south of the Tyuratam space center and ICBM test facility and was capable of eavesdropping on the ABM test center at Sary Shagan.87

  President Carter considered the sites sufficiently important that he told his ambassador to Iran, William Sullivan, that intelligence cooperation between the CIA and Iran should continue despite the Shah’s poor human rights record.88 But in January 1979, the Shah of Iran, in the wake of increasing protests and riots, fled the country he had ruled so autocratically for twenty-five years.

  The Shah’s end took the U.S. intelligence community by surprise. A sixty-page CIA study completed in August 1977, Iran in the 1980s, had asserted that “there will be no radical change in Iranian political behavior in the near future” and that “the Shah will be an active participant in Iranian life well into the 1980s.” A year later, a twenty-three-page CIA Intelligence Assessment, “Iran After the Shah,” proclaimed that “Iran is not in a revolutionary or even a ‘prerevolutionary’ situation.” A month later, the DIA issued an Intelligence Appraisal stating that the Shah “is expected to remain actively in power over the next ten years.”89