The Wizards of Langley
Page 18
Wheelon, according to an NRO history, “either did not await DoD action or, more probably, had advance notice of Vance’s intentions.” On July 9, before Vance’s letter could reach McCone, the science and technology chief sent McMillan an outline of “the various tasks for which we require immediate NRO funding.” Wheelon’s task description went beyond feasibility studies to include funding for spacecraft, booster, and “assembly, integration, and checkout” contracts.108
That same month, the United States Intelligence Board formally called for development of a new search system—which still left open the question of the system’s characteristics and which agency should manage its development. On August 11, a meeting of Vance, McCone, Fubini, and McMillan addressed the CIA’s proposal. McCone accepted in principle a funding level of about $30 million and a set of Vance instructions on FULCRUM issued a week earlier, which was expanded to provide for some system design study work but under the aegis of the NRO.109
In a late 1964 presentation to the PFIAB, McMillan noted the Purcell panel’s report and apparently referred to its suggestion that the best thing to do was stay with CORONA. In a memo, the DS&T’s John McMahon contended that such an argument was misleading. He noted the contents of the briefings given by Reber and Lundahl as well as the panel’s being told that the new search system proposed as a follow-on to CORONA was a “10,000 lb. monster” that would require a Titan 3C booster. Thus, “the panel felt that rather than bankrupt the US Treasury . . . we turn to CORONA and make [it] work all the time” at nine-foot resolution. McMahon argued that the Purcell panel members failed to explain the documentation and presentations upon which they based their recommendations, and he suggested that if they were briefed on the current systems under way and the resolutions required, they would reach different conclusions than they did in July 1963.110
In any case, McMillan and the NRO were determined that FULCRUM, about which the CIA would tell them little, would not be the only candidate to succeed CORONA. A contract to begin studies for a system designated the S-2 was issued to Eastman-Kodak, whose approach McMillan recalled as “fairly conventional.” Some attention was also devoted to a proposal for a smaller system, designated MATCHBOX, that was advertised as being capable of producing equally detailed imagery.111
Then, on February 24, 1965, Itek made an announcement that stunned Wheelon and “the NRO Staff found hilariously enjoyable”—that it would undertake no further work on the FULCRUM program.112 Since Itek began serious work on FULCRUM, it had been faced with a CIA requirement that it felt unnecessary and unreasonable—that the camera be capable of photographing targets up to 60 degrees to the left or right of the satellite’s path above the earth—from horizon to horizon. The farther a camera is moved “off axis,” the more the atmosphere degrades its resolution, and thus a 35-degree capability in each direction had been the maximum demanded of any CORONA camera. As Walter Levison, a camera designer and senior Itek official at the time, recalled, Itek thought the decline in resolution that would result would be too great to justify attempting to produce a system that could scan 120 degrees.113
The difference in viewpoints had apparently led to some hard feelings between Special Projects Staff head Jack Maxey and Itek’s FULCRUM program manager John Wolfe. Maxey was, according to Frank Madden, the chief engineer for FULCRUM, “high-handed, demanding.” But the event that triggered Itek’s withdrawal, according to Levison, occurred at a meeting in Boston of the Land Reconnaissance Panel, attended by, among others, Levison, Itek president Frank Lindsay, Wheelon, and McMillan. The meeting featured a briefing by Leslie Dirks on FULCRUM. Dirks insisted that the requirement for the new satellite to scan 60 degrees in each direction was an Itek recommendation and not made at the CIA’s insistence. Levison’s reaction, in “the heat of the moment,” was “that tears it.” Later that afternoon, a meeting among Levison, Lindsay, and other Itek executives resulted in the decision to withdraw from the FULCRUM program.114*
After the meeting, Levison called NRO staffer Paul Worthman, who in a memo described Levison’s voice “as shaking throughout the conversation.” Levison informed Worthman of Itek’s decision and requested advice on how to handle the situation. Worthman suggested the first thing to do was inform McCone, which Levison said Itek president Frank Lindsay was trying to do at that moment. When McCone could not be reached, Lindsay called John Bross to give him the bad news.115 Worthman then called McMillan and told him to call Levison immediately. A meeting among Levison, Wolfe, McMillan, and Land followed, which left the latter two “stunned.” Levison told them that Itek felt it could not survive under “the domination of the CIA” and that the CIA had fostered an “immoral environment.”116
Itek’s announcement that it would no longer work on FULCRUM “hit us like a ton,” John McMahon recalled. It also led to suspicions on the part of several CIA officials that McMillan or Fubini had offered Itek an inducement to withdraw from the CIA program—such as a guarantee ofan NRO contract to build the next search system. McMillan denied any previous arrangement with Itek, and Levison recalled that “nobody made any promises to anybody.” McMillan did transfer the S-2 program from Eastman-Kodak to Itek after its withdrawal from FULCRUM. According to Levison, McMillan wanted to keep Itek working in the reconnaissance field, and Eastman-Kodak had plenty of work—including working on the KH-10 optical system for the Air Force’s Manned Orbiting Laboratory. Part of McMillan’s decision was apparently the result of the technical discussion about a new search system at the February 24 meeting.117
Back at Langley, McCone and Wheelon decided they would have to find another contractor, possibly Perkin-Elmer. Meanwhile, McMahon, along with two other CIA officials—Jim MacDonald and Henry Plaster—were sent to Itek, where they seized all records, brassboards, and engineering notebooks related to FULCRUM.118
During a visit to Perkin-Elmer, Wheelon asked Rod Scott for any ideas. Scott explained the concept of the “twister,” which would allow images to be recorded on film that was see-sawed back and forth—a radical departure from the practice of advancing the film frame by frame past the focal plane. The twister would permit placement of the cameras in the satellite so that they would be parallel to the satellite’s motion rather than perpendicular—which in turn meant the satellite could carry cameras of sufficient size to achieve the CIA’s resolution and scan objectives. (In other words, placing cameras of sufficient size across the width of the spacecraft would require building a spacecraft that would be too large for the nose of the launch vehicle, whereas placing them across the length would not.) The concept also involved rotating the focal plane rather than the camera.119
Wheelon next consulted Land, who preferred Perkin-Elmer’s design to Itek’s. According to McMillan, Land was supportive because he was not a systems engineer but a scientist who “liked nothing more than an innovative, clever device.” The twister “just knocked him off his chair.” Mc- Cone then put up $10 million–$30 million of agency money to keep the project going.120
In mid-July, McMillan made one last effort to slow down FULCRUM, sending Vance and Raborn a report in which he asked for a deferred review of progress. McMillan reported to Vance that the original S-2 system still appeared to be the most promising approach, adding that he proposed to select either Itek or Eastman-Kodak to develop an alternate camera configuration.121
The reaction from Raborn was similar to the reactions of McCone, Carter, and Wheelon to comparable proposals on similar occasions in the past. First, he politely protested McMillan’s apparent intention of unilaterally selecting a specific search system for development. Then he invoked the pending Land panel report as reason for not rushing to judgment. Finally, he made the point that only he and Vance could make the final decision on any specific search system.122
Vance had earlier cautioned McMillan to proceed cautiously in making program commitments to Itek, but McMillan, who was convinced that the S-2 system was by far the best prospect, had continued to invest in it. Since the Land p
anel had proposed no solution, only further study, Raborn suggested that McMillan had exceeded the authority entrusted to him.123
According to an NRO history, failure of the Land panel to make a choice was a disappointment for the NRO and McMillan. They had “hoped for selection of some system other than [FULCRUM], a development that would tend to choke off the CIA’s involvement in the creation of new satellite systems.”124
As a result of the August 1965 agreement, the CIA was given responsibility for managing the development of the new search system. But in September, a McMahon memorandum charged that McMillan had indicated that Itek had been selected to build the new search system, despite the fact that the competition was still ongoing.125
Perkin-Elmer ultimately won the competition, probably because of the coverage provided. The criteria for evaluation were written by McMillan’s successor as NRO director, Al Flax, and the scan angle received high priority. Its optical bar system provided horizon-to-horizon coverage, even though it was not used a majority of the time because the resolution was, as Itek had warned, severely degraded at the extremes. Normally, the camera scanned 30 degrees to either side rather than 60 degrees. The system was intended to yield resolution of three feet at the nadir (when the target was directly underneath) but produced no better than six feet at the extremes.126
On April 22, 1966, the USIB gave its blessing to development of a new search system along the lines of FULCRUM rather than its main competitor, the S-2. A new code name, possibly AQUILINE, was assigned to the program that same day, but it was replaced eight days later by the name it was subsequently known by, at least by those with the proper clearances—HEXAGON.127
Exactly four months after the USIB gave its approval to HEXAGON, Bud Wheelon wrote to DCI Richard Helms to confirm a conversation they had the previous week. In the letter he noted that “when I accepted John McCone’s invitation to join CIA in early 1962, it was with the understanding that I would serve . . . not more than four years. I believe that I have now accomplished all of the major objectives in creating a technical intelligence component for the Agency.” He also noted that “Hughes Aircraft Company has offered me a most attractive position. . . . As we agreed, 23 September 1966 will be my last official day.”128
A month and a day later, Wheelon put in his last day as Deputy Director for Science and Technology. There would be much more work to be done before HEXAGON or RHYOLITE flew, but they would become two of the three key reasons why, in 1994, he would be named the first winner of the R. V. Jones award. The satellites also provided the foundation for the CIA’s continuing role in the development of space reconnaissance systems, which would include a third major system whose origins could be traced to Wheelon’s tenure. The same CIA official who recalled Wheelon as “an acerbic son of a bitch” noted that without him, the Air Force “would have run away with the [reconnaissance] program.”129 He might have added that in that case, there would have been no RHYOLITE, HEXAGON, or KH-11.
(Left) Richard Bissell, photographed standing in front of the Brandenburg Gate, joined the CIA in 1954 as a special assistant to DCI Allen Dulles. During his tenure with the CIA (1954–1962), he oversaw the successful development of the U–2 and CORONA reconnaissance systems for the CIA. CREDIT: Fran Pudlo
(Below) Edwin Land, second from the left in the second row, with other recipients of the National Medal of Science and President Johnson. Land served as a scientific adviser to the CIA and NRO. He was a key figure in the decision to proceed with the U–2 and KH–11 systems and in the creation of the Directorate of Science and Technology. CREDIT: LBJ Library
(top left) Herbert Scoville Jr. joined the CIA as head of the Office of Scientific Intelligence and became Deputy Director for Research in 1962. He resigned a year later, largely because of the problems in his relationship with the National Reconnaissance Office. CREDIT: CIA
(bottom left) Albert “Bud” Wheelon replaced Herbert Scoville as OSI chief in 1962. When Scoville resigned from the CIA in 1963, Wheelon was asked to replace him. As Deputy Director for Science and Technology, Wheelon reestablished the CIA’s role in space reconnaissance and supervised the initial development of key imagery and signals intelligence systems. CREDIT: CIA
(above) Carl Duckett replaced Wheelon as head of the science and technology directorate in 1966 and served in that position until 1976. Under Duckett, two key programs initiated by Wheelon—the RHYOLITE signals intelligence satellite and the HEXAGON imagery satellite—became operational. A third, the KH–11, received presidential approval and was first launched in December 1976. CREDIT: CIA
(top left) Leslie Dirks joined the CIA in 1962 and became Deputy Director for Science and Technology in 1976. He played a key role in the KH–11 program from its conception, serving as its program manager prior to replacing Duckett. CREDIT: CIA
(bottom left) Lloyd Lauderdale served as Carl Duckett’s deputy before leaving government to join E-Systems. He is known as the “father of RHYOLITE,” having served as its program manager and guiding its technical development. CREDIT: Virginia S. Lauderdale
(above) Ruth David became the first Deputy Director of Science and Technology chosen from outside the CIA’s ranks. Her tenure (1995–1998) involved controversy over some of her decisions and a decline in the directorate’s status due to changes in the directorate’s role in space reconnaissance and imagery interpretation. She also increased the directorate’s work in information technology. CREDIT: CIA
Central Intelligence Agency, Headquarters, Langley, Virginia. CREDIT: CIA
Headquarters of the National Photographic Interpretation Center at the Washington Navy Yard. NPIC was established in 1961 as “service of common concern” to be managed by the CIA. In 1973, it became part of the DS&T. In 1995, it was merged with a number of other agencies to form the National Imagery and Mapping Agency, which was placed within the Department of Defense. CREDIT: Federation of American Scientists
(above) The Matomic Building (second from right) at 1717 H Street N.W. served as headquarters for the U–2 project staff from February 1956 until March 1962, when the staff moved to CIA’s new headquarters at Langley, Virginia. CREDIT: Federation of American Scientists
(below) The Ames Building in Rosslyn, Virginia, housed a number of CIA offices for several years, including the DS&T’s Office of Research and Development. The CIA moved out of the facility in the late 1990s. CREDIT: Federation of American Scientists
*Thirty-five years later, Levison wondered “What the hell difference did it make?” and noted that the amount of film a satellite could carry in the case of CORONA, had increased from 20 to 180 pounds. With that much film on board, the sensible reaction to wasting a few frames of film would be “Who cares?” (Telephone interview with Walter Levison, September 17, 1999.)
5
CHANGE OF COMMAND
When Bud Wheelon headed back to California in October 1966 to become vice-president for engineering at Hughes Aircraft, he brought along the Distinguished Intelligence Medal awarded to him shortly before his departure.1 He left behind a science and technology directorate that had become a key player in scientific and technical intelligence production, signals intelligence collection, and, most important, overhead reconnaissance.
In many ways, the DS&T was an intelligence community in itself—establishing requirements, developing and operating collection systems, and analyzing some of the data collected. What Ray Cline had seen as a problem—collectors and analysts reporting to the same boss—many in the directorate saw as a virtue. The priorities of those developing and operating collection systems flowed, without bureaucratic obstacles, from the priorities of the analysts.
Assuming command of the directorate, as acting deputy director, was former FMSAC chief Carl Duckett, who had become Associate Deputy Director for Science and Technology in mid-May 1966.2 In some ways, Duckett was no Wheelon; in others, Wheelon was no Duckett.
Although the directorate would relinquish its role in aerial reconnaissance operations to the
Air Force on his watch, Duckett’s strengths enabled the directorate to flourish and to reach its peak of power and influence. During his decade-long tenure, the objectives of the HEXAGON and RHYOLITE programs would be realized, another battle with the Air Force over development of an advanced satellite system would be fought and won, and collection operations in Iran and Norway would continue to yield vital intelligence on Soviet missile programs. The directorate would also expand its role in the CIA’s scientific and technical projects, assuming control over offices previously belonging to the intelligence and operations directorates.
A NEW BOSS
In contrast to Bud Wheelon, who had arrived at the agency with advanced degrees from prestigious institutions and a long list of scientific publications, Carl Duckett brought no impressive scholarly credentials to the job, not even a college degree. The lack of such credentials almost prevented Duckett from becoming Wheelon’s successor.
Wheelon himself did not think Duckett should succeed him, despite his having “many fine qualities”—and told Raborn and Helms so. Wheelon believed that the directorate was still in the “creation” stage, that it needed a strong technical person, but that individuals such as Leslie Dirks were too young. Thus, he had lined up his former boss at TRW, Frank Lehan, to replace him. Lehan was familiar with DS&T operations, having served as a key adviser from its birth. But a personal problem, in the form of a wayward daughter, forced Lehan to withdraw his name after Wheelon made it clear he could not stay on for another year.3