The Wizards of Langley

by Jeffrey T Richelson

  Wheelon also had come to the conclusion that the work of OSI’s offensive systems division was not up to his standards and needed an infusion of better-trained personnel—individuals who understood telemetry and other technical issues. Furthermore, missiles and space were not within the expertise of OSI chief Donald Chamberlain. And he also had “too much on his plate” anyway, according to Wheelon. There was also, in Wheelon’s view, a need for a forceful leader, someone willing to battle the Air Force over issues such as the capability of a new Soviet missile.54

  Also needed was an organization that had no institutional conflict of interest in analyzing foreign efforts and had access to the full range of intelligence data. In Wheelon’s view, neither the Army’s missile intelligence unit nor the Air Force’s Foreign Technology Division satisfied both those requirements. FTD, in particular, was to Wheelon a “propaganda mill,” and one function of FMSAC would be to “keep FTD honest.”55

  One particular controversy that, to Wheelon, indicated a need for a second voice on missile matters was the continuing debate over the SS-8 missile. In 1961, using data collected from the first tests of the missile earlier that year, scientists working for the Air Force calculated that the SS-8 nose cone weighed around 25,000 pounds—sufficient to carry a warhead in the 100-megaton class. But there were doubters in other corners of the intelligence community, including CIA, and during 1962 the question of the nose cone’s size became a matter of intense disagreement and the focus of a major analytical effort. In 1962 and 1963, outside review groups and other members of the intelligence community, including Army intelligence, moved toward the view that the SS-8 warhead was small.56

  An October 1963 national intelligence estimate stated that the data available indicated that the SS-8, if large, could carry a nose cone weighing about 10,000 pounds, but the best estimate was that it had a payload similar to that of the SS-7—only about 4,500 pounds. The Air Force retreated but did not surrender—insisting that the evidence did not exclude the possibility of the SS-8 carrying a nose cone of up to 18,000 pounds.57

  Wheelon intended FMSAC to be a center where all incoming information relevant to missile and space activity would arrive and be analyzed, with results distributed to the White House, NASA, and other interested parties. It would also play a role, Wheelon expected, in influencing the development and deployment of collection systems.58

  Wheelon found a forceful leader for FMSAC in JAM SESSION colleague Carl Duckett, at the time the head of the Directorate of Missile Intelligence of the Army Missile Command at Huntsville, Alabama. Duckett differed from many of those who rose to high levels in the CIA. He grew up in rural North Carolina and never attended an Ivy League university. When he was seventeen, his mother presented him with a new pair of jeans, some money, and instructions to get a job at the mill down the road. Duckett didn’t stop for 200 miles, until he got a job at a radio station. He was drafted for service in World War II, and when the results of his IQ tests came in, it was apparent that the military had a genius on its hands. Duckett was then sent to study radio at Johns Hopkins University.59

  After the war, Duckett was assigned to White Sands Proving Ground and became involved in missile testing and telemetry analysis. From White Sands he moved on to Huntsville to work on range instrumentation. In 1957, he was brought to Washington as part of the JAM SESSION program.60

  Among those Duckett brought into FMSAC were David Brandwein, a veteran of TRW and EARSHOT, who would succeed Duckett as FMSAC director; M. Corley Wonus, a future head of the directorate’s SIGINT operations; and future DS&T chief R. Evans Hineman (commonly referred to as Evan Hineman).61 In 1956, Hineman, having obtained his degree in mechanical engineering and completed his Reserve Officer Training Course, was headed for two years of Army service. A course in technical intelligence, which he considered preferable to the alternative of learning to repair tanks, was followed by assignment to the Army missile intelligence unit at Huntsville. In his two years there, he had an inside view of Soviet space and missile efforts and was called on to brief notables such as Wernher von Braun and General John Medaris, head of the Army’s missile program.62

  When his two years at Huntsville were up, Hineman joined the Army Ordnance Technical Intelligence Agency at Arlington Hall, Virginia—an organization that would become part of the Army’s Foreign Science and Technology Center (FSTC) when it was formed in 1962. Hineman first met Carl Duckett around 1960. When Duckett was subsequently put in charge of the Army’s missile intelligence effort, he tried to get Hineman to come back to Huntsville, but both Hineman and his wife felt they had seen enough of the Alabama town. When he became head of FMSAC, Duckett called again, and this time, Hineman signed up.63

  Not everyone was as enthusiastic about FMSAC as Wheelon, Duckett, and Hineman. Among the least enamored were two powerful Air Force generals, Bernard Schriever, head of the Air Force Systems Command (AFSC) and chief of staff Curtis LeMay. In September 1963, LeMay had been alerted of Wheelon’s plans in a letter from Schriever’s deputy. In December, after the center’s creation and a briefing from Wheelon and Duckett, Schriever wrote to LeMay, urging that “immediate action should be taken to slow down or block CIA action to duplicate DOD missile and space intelligence.”64 By “DOD” Schriever meant “FTD,” which reported to his Systems Command.

  Schriever complained that “the establishment of this activity within CIA is most certainly the first step in competing with and possibly attempting to usurp the Services’ capabilities in this area of scientific and technical intelligence.” He also objected that the creation of FMSAC had “resulted in undesirable competition for special talent and special data.” Schriever characterized the Air Force capability in the area as representing “a significant investment in manpower and resources and . . . an extremely vital function which must not be lost or permitted to be eroded by another government agency.”65

  He recommended that Joseph Carroll, head of the Defense Intelligence Agency and thus the senior military official on the United States Intelligence Board, be encouraged to protest CIA activities in the area “at least until an agreement on respective responsibilities and mutual support can be reached.” But Schriever believed that since “the problem could not be solved through intelligence channels alone . . . I recommend that you and the Secretary act to protest this expensive and unnecessary duplication of DOD space and missile intelligence analysis by CIA.”66

  On January 2, 1964, LeMay asked the head of Air Force intelligence, Brig. Gen. Jack Thomas, and a colleague to prepare a memorandum that would serve as the basis of a JCS request to the Secretary of Defense for OSD action “to oppose the FMSAC program.”67 That was apparently followed by a memo from LeMay to Schriever stating that he shared Schriever’s concerns and that the memo he envisioned going to McNa-mara would request the Defense Secretary to consult with McCone “in an effort to prevent a major CIA effort competing with and largely duplicating activities well under way in Department of Defense agencies.”68

  Ultimately, the Air Force opposition proved futile. But the Defense Department’s review of missile and space intelligence activities, which was being conducted during late 1963, did result in a Defense Department competitor, at least in some respects, for FMSAC. Chartered by a Defense Department directive, the Defense Special Missile and Astronautics Center (DEFSMAC), a joint NSA-DIA operation, opened for business on June 1, 1964.69

  The center was to receive warnings of upcoming foreign missile and space launches, alert all relevant intelligence collectors and officials of the forthcoming launches, and provide initial assessments of the launches. Former DEFSMAC chief Charles Tevis recalled that there was a “great opportunity for these two . . . centers to fight . . . everybody likes to be the first one to get a current report out,” and there was a “kind of one-upmanship in reporting.” According to Hineman, there was a rivalry with DEFSMAC over “who’s going to get to the street first,” and the rivalry was probably good for the country. Since the intelligence communi
ty was working with incomplete data, it was useful to have organizations that could go down different analytical paths in pursuit of the truth.70

  There was ample Soviet missile and space activity to keep FMSAC and DEFSMAC busy during Wheelon’s tenure. The Soviets began orbiting a variety of military support satellites. Reconnaissance and meteorology spacecraft joined scientific satellites in orbit, usually with the uninformative Cosmos designation (which was also used for space probes that never made it out of earth orbit). In 1965, the first Soviet Molniya (“Lightning”) communications spacecraft was placed in its peculiar orbit—flying at a 63-degree inclination and reaching 24,000 miles above the earth when over the northern Soviet Union and descending to a mere 240 or so miles when it whizzed over the southern portion of the planet.71

  Manned missions were intended to claim space “firsts” for the Soviets as well as help prepare for a mission to the moon. The cosmonauts stayed close to earth initially; unmanned missions were intended to establish an ability to reach, orbit, and land upon the moon as well as send back photographs. Thus, the October 1964 Voshkod 1 mission was the first flight without spacesuits and the first with direct in-flight medical observations. The Voshkod 2 mission of March 1965 included the first space walk, a ten-minute stroll outside the capsule by Alexei Leonov.72

  Meanwhile, Soviet lunar probes were launched to orbit the moon or achieve a soft landing and in each case also to send back photographs. A string of failures in 1964 and early 1965 was followed by the success of Zond 3, which sent back pictures of the lunar surface taken during a flyby. That was followed by several more failed soft landings before Luna 9 was successful on January 31, 1966.73

  Probes fired at Mars or Venus included the April 1964 mission of Zond 1, which the Soviets apparently lost contact with before it passed by Venus. The Venera 2 and Venera 3 missions in November 1965 were partially successful but did not transmit any data because the first flew by and the second crashed into the planet.74

  One component of FMSAC, the Activities Interpretation Division, served as twenty-four-hour watch center. The time difference between Washington and Soviet launch sites, along with the urgency in determining the mission associated with each launch, required FMSAC to disrupt the sleep of key personnel. Evan Hineman remembers being called in during the middle of the night on several occasions to examine data on trajectories and orbits so that the White House, NASA, and other agencies could be told whether an earth satellite, interplanetary probe, or lunar mission had been launched.75

  Among the important elements of these inquiries was trajectory analysis. Establishing the launch time to the nearest minute, through tracing the trajectory back to the launch point and incorporating data on velocity, permitted a great deal to be inferred about the objectives of the mission as well as the technology employed. Thus, the launch time of Luna 1 on January 2, 1959, had enabled U.S. analysts to conclude that the Soviet claim that it was a solar satellite was most likely an attempt to cover up a failed attempt to send the spacecraft crashing into the moon’s surface.76

  Quick assessments were only part of FMSAC’s work. Wheelon and Graybeal noted in their 1961 article that although “gross features of a Soviet space shot can usually be . . . established within the first few hours by an experienced technical man,” the “variations and nuances of a given flight, however, which can be equally important, may require weeks of concentrated effort by a team of subsystem specialists working together.” The results of such analysis would be a clear picture of mission performance and the system’s technical features.77

  Thus, the CIA was able to report to the White House on June 1, 1964, that Zond 1, which had been launched two months earlier, would reach the vicinity of Venus on July 20. The report noted that “we cannot yet tell whether it will impact on Venus or fly by, perhaps ejecting an instrumented probe to explore the planet’s atmosphere as it goes by.” (Zond 1 came within 62,000 miles of Venus, but the failure of its radio prevented any data from being returned.)78

  FMSAC personnel also sought to unravel the failure of the unmanned Cosmos 57 mission, which was placed into orbit on February 12, 1965. Analysts examined the data from radars that tracked the spacecraft’s movements in space and scrutinized intercepted telemetry in their attempt to understand the purpose of the mission and why the spacecraft burned up not long after its first orbit.79

  The space walk of Alexei Leonov a few weeks later pointed FMSAC analysts in the right direction. By comparing the telemetry from Cosmos 57 and Voshkod 2, they were able to determine that Cosmos 57 was a test for the automated system that operated the airlock Leonov needed to pass through. They also determined, from intercepts of signals to and from Cosmos 57, the key channels on which commands were transmitted to the spacecraft from the Soviet control station, as well those from the spacecraft to ground controllers that told how it was responding.80

  The explanation analysts pried from the intercepted signals was that while the spacecraft was in range of signals from one transmitter, it received a duplicate set from a second transmitter that was intended to pick up communications with the spacecraft when it flew out of range of the first. The double signals were merged into a single signal that instructed the spacecraft to fire its retro-rockets in preparation for descent. Possibly because of the mass of deployed airlock, the spacecraft then began tumbling some seventy-eight times a minute. Only because the airlock operations could be carried out manually was the Leonov space walk able to proceed as planned.81

  The Luna 9 mission of early 1966 was also of great interest to FMSAC analysts because a successful soft landing would have returned the Soviet Union to leadership in the lunar race. U.S. collection sites, including the Army’s Sinop facility in Turkey and the NSA’s STONEHOUSE facility in Ethiopia, enabled U.S. analysts to monitor launch, orbit, and ejection and to determine that Luna 9’s trajectory was on target.82

  From the intercepted telemetry, analysts determined when the spacecraft’s engine had been fired to send it cruising toward the moon. On February 3, it was oriented for retromaneuver while STONEHOUSE, Jodrell Bank, and other sites listened in order to provide FMSAC, DEFSMAC, and other interested parties with data to analyze. Luna 9 landed softly that same day. Its first signals from the moon included telemetry as well as what was soon recognized to be a fax transmission. In the United States and England, fax machines were modified to convert the signal into pictures—with the result that some of the pictures obtained at Jodrell Bank were published before the Soviets officially released them.83

  Of greater concern than the Soviet space program was the Soviet missile program—particularly the ICBMs and submarine-launched ballistic missiles (SLBMs) that could hit U.S. territory. From 1964 through mid-1966, the Soviets began testing three third-generation ICBMs—the SS-9 Scarp, SS-11 Sego, and SS-13 Savage. They also conducted tests of their SS-7 Saddler and SS-8 Sasin missiles.84

  Some issues about those programs, such as the numbers produced and deployed, the locations of deployment sites, and targeting policy, were considered outside of FMSAC’s charter. The technical intelligence analysts at FMSAC worried about key characteristics of the missiles and their warheads—missile size, yield, accuracy, range, throw weight, vulnerability to defensive systems, whether the warhead could be airburst, and whether the missile was liquid- or solid-fueled.85

  Such technical details were not merely of academic interest to missile designers, for they could have significant strategic implications. They were a key element in determining whether the Soviet strategic forces could strike certain types of targets, overwhelm missile defenses, or destroy U.S. ICBMs in a preemptive strike.

  The information mined by FMSAC analysts, only some types of which were available for any given missile launch, included optical, radar, and telemetry data. To estimate the size and shape of a reentry vehicle, photographs were most helpful but were rarely obtained. Less directly indicative, but still useful, were radar cross-section data.86 From estimates of size, estimates of yield follow
ed.

  Extracting intelligence from telemetry data required not only a facility for technical analysis but also ingenuity. Soviet missile designers knew what aspect of the missile’s performance each channel of telemetry measured and how that performance was being measured, but FMSAC analysts did not. In addition, they usually were confronted with an incomplete set of telemetry, since during the 1960s the United States was rarely able to gather telemetry during the earliest launch stage because Tyuratam was over the radio horizon from U.S. eavesdropping antennae.

  Despite such handicaps, FMSAC analysts could rely on the fact that certain basic measurements, such as acceleration and fuel pressure, were required during any test, and that the numbers associated with particular aspects of the missile’s performance would behave in a certain manner. Thus, when the system feeding propellant to an engine shut off, measured pressure would drop to zero in considerably less than a second, and the turbine would take four to eight seconds to coast to a stop.87

  An analogy can be drawn to a car: A person riding in a car would expect to have a set of gauges on the dashboard measuring speed, oil pressure, and other aspects of the car’s performance. Even if those instruments were placed in unconventional positions in the car and no units of measurement were indicated, it would be possible to correlate specific instruments with the car’s behavior to determine both the function of an instrument as well as the units of measurement employed. Thus, “given a fair sample of powered flight telemetry, the analyst can usually say whether the vehicle is liquid or solid-fueled, whether it has a single burning stage or multiple stages, and what ratio of payload to total weight it probably has.”88