China’s nuclear program was not the only target of U-2 overflights. In mid-1963, Deputy DCI Marshall Carter, Jack Ledford, and Kelly Johnson were all at Edwards Air Force Base in California. One day they met at the bachelor officers’ quarters. Ledford later recounted that Johnson had a couple of White Horse scotches, the only scotch he would drink, and that the three of them began talking. The question arose of whether it would be possible to operate U-2s from aircraft carriers. Such an ability would alleviate the problem of having to find foreign bases for U-2 operations—since many countries were nervous about a U-2 base on their territory. That conversation was not the first time such a possibility had been raised.125
What followed was Project WHALE TALE, the modification of some U-2s to permit carrier operations. The project began in August 1963 with a successful U-2C takeoff from the USS Kitty Hawk operating off San Diego. The landing was much less successful: The plane bounced, hit hard on one wing, and barely managed to become airborne again before reaching the end of the deck. To make landings possible, the landing gear was strengthened, an arresting hook was attached to the rear of the fuselage, and “spoilers” were placed on the wings that, with the push of a button by the pilot, would kill the plane’s lift.126
In May 1964, a U-2G—a U-2C with the necessary modifications—was aboard the USS Ranger when it set sail for the mid-Pacific. The plane’s target was Mururoa atoll, a twenty-mile-by-ten-mile strip of territory thousands of miles from land. For a U-2 to overfly it would require a carrier as a launch site. Mururoa was part of French Polynesia and was a target because a year earlier France had selected the atoll as its new nuclear test site, to replace the site in what had become independent Algeria. For the next four years, the CIA maintained a contingent of carrier-qualified U-2 pilots as well as three U-2Gs. During that time, France conducted six atmospheric and three underground nuclear tests on Mururoa. But the May 1964 flight was the sole operational mission from a carrier. After the first, senior officials had second thoughts on the propriety (and presumably the public relations risk) of spying on an ally, even a troublesome one.127
One proposed modification of the U-2, aimed at a quite different target, never was made. ORD’s director, Robert Chapman, advocated building a long-focal-length camera to be put in the nose of a U-2. Rather than looking down, the camera would look up into space and photograph the Soviet satellites that FMSAC was so curious about. Putting the camera in a U-2 would get it up above the atmosphere and avoid the loss of clarity due to atmospheric interference. Chapman hoped the clearer images would allow more precise evaluation of the satellite’s mission. However, the proposal was, as Bud Wheelon recalled, “not a very good idea” and was “stomped on” by the NRO.128
The NRO also helped terminate the low-altitude overflights of China that began with the ST/POLLY program. In June 1963, a P-3A Orion maritime patrol aircraft arrived at the Naval Aviation Depot at Alameda, California, after having been diverted from Florida-based training flights. The plane was modified by widening the main cabin door and adding a duplicate next to the original door. As a result, both doors swung inward and back out of the way, creating an opening approximately fifty-three inches wide. That Orion was the first of three P-3As that were remodeled as part of ST/SPIN—the follow-on to ST/POLLY.129
Assigned to manage the conversion program, which was carried out by the secretive E-Systems of Greenville, Texas, was OEL’s Robert Singel. As was the case with its predecessor, the primary function of the aircraft was to fly low into China in order to stir up and monitor China’s air defenses, data required by the Strategic Air Command. The modification of its cargo doors reflected the plan to use the planes for the same types of covert-action operations as the P-2V Neptune conducted—dropping equipment, arms, ammunition, and agents on some occasions, and thousands of propaganda leaflets on others.130
Along with the ELINT receivers that were carried on the Neptunes, the Orions were equipped with a variety of additional sensors, including side-looking airborne radar for missions flown along the Chinese border, communications intercept equipment, and an infrared detector. Reportedly, an acoustic eavesdropping device was tested, one so sensitive that it could detect engine and machine-manufacturing noises. The planes were also equipped with cameras for slant-range or oblique photography. To help satisfy nuclear intelligence requirements, the planes also carried an air-sampling apparatus that was connected to the ram air scoops to the rear of each side of the cockpit.131
The first ST/SPIN aircraft, manned by Nationalist Chinese personnel, began operations in 1964. Some missions involved flying along the southern border of the People’s Republic of China (PRC), with occasional penetrations to detect air defense radars and determine the characteristics of their signals. Missions into the interior, in addition to conducting airdrops, also involved locating military installations, intercepting military communications, and air sampling. Missions were eventually conducted over Burma and up into Tibet to collect intelligence on Chinese activities against the civilian population.132
The ST/POLLY and ST/SPIN programs were threatened by the behavior of both PRC and Nationalist Chinese officials. The PRC was occasionally able to shoot down an aircraft. P-2Vs were shot down in November 1961, June 1963, and June 1964. Another disappeared in January 1962, possibly as a result of an operational accident. In addition, the PRC’s information about the Nationalist Chinese personnel involved was so extensive that the intruding P-3As often received calls from PRC radio operations requesting to speak, by name, to one or more of the Nationalist Chinese on board. Further, the planes carried far more personnel than necessary, often twenty-seven instead of the required fourteen to sixteen. Personnel received bonuses for participating in such missions, and the Nationalist Chinese general who ran the program was willing to assign extra crew members to the missions in exchange for a kickback (which he also required from the first fourteen crewmen). When OSA chief Ledford told Wheelon that the general had been sent to prison for his actions, Wheelon told him that not only should the United States do nothing to get him out, but that the Nationalist Chinese should “throw away the fucking key.”133
In 1965, before the third P-3A had been modified, Singel received word that the program was being terminated by the NRO, which had provided the funding.134 It is unclear what combination of PRC and Nationalist Chinese actions, the capabilities exhibited by the NRO’s new ferret satellites, and budgetary limitations resulted in the cancellation decision. The last flights in the program occurred in 1966.135
In September 1966, the OXCART had still not flown an operational mission. For Wheelon, it proved a “hell of a challenge,” a weekly “four-alarm fire” that threatened to destroy the CIA’s “reputation for doing things on the cheap [and] quickly.” Bureaucratic and technical issues consumed “so god damn much of my time,” Wheelon recalled. Along with Ledford, he worked to ensure continued Air Force support, which included allocation of tankers to refuel the planes, for the program.136
According to Wheelon, for a while it was unclear if OXCART would ever reach operational status. In contrast to the U-2, it was behind schedule and over budget. The fundamental technical problem was compressor stalls, which induced horrendous shaking and caused pilots to bail out. But more than a plane was lost. One pilot died when his parachute didn’t open, and all but one Lockheed test pilot quit. Wheelon threatened to cancel the program unless Kelly Johnson, “who didn’t trust electronics,” put in an electronic system to correct the problem. Wheelon also put people to work developing a jammer that could limit the threat from hostile radars.137
But the reason OXCART was not operational by September 1966 was not problems with the aircraft, which had been corrected. By the end of 1963, there had been 573 flights totaling 765 hours. On July 20, 1963, an OXCART flew at Mach 3 for the first time, and in November it reached Mach 3.2, flying at 78,000 feet. On February 3, 1964, it flew for ten minutes at Mach 3.2 and 83,000 feet. By the end of 1964, there had been 1,160 flights, although
the flight time was only 1,616 hours. Eleven aircraft were available, four of which were reserved for testing.138
On January 27, 1965, an OXCART took off on its first long-range, high-speed flight. It flew for 1 hour and 40 minutes, with all but 25 minutes of its flight conducted at above Mach 3.1. Its total range was 2,580 nautical miles, at altitudes between 75,600 and 80,000 feet.139
OXCART was delayed because higher authorities turned down repeated proposals to employ the plane on operational missions. The first suggestion that OXCART might be ready for operational deployment came in early 1964, when OSA began planning for its use over Cuba under a program designated SKYLARK. On August 5, Deputy DCI Marshall Carter directed that OSA achieve emergency operational readiness by November 5. The plane would operate at Mach 2.8 and 80,000 feet. But SKYLARK was never implemented.140
In 1965, Asia became a possible theater for OXCART’s first operations. On March 18, McCone, McNamara, and Deputy Secretary of Defense Cyrus Vance met to discuss the growing threat to aerial surveillance of China, a threat manifested in the loss of several Air Force drones. They agreed that the CIA should take all preparatory steps necessary to overfly China with OXCART aircraft, flying out of Kadena Air Base on Okinawa. Deploying the planes to Okinawa, however, would require presidential approval.141
Four days after McCone and Vance met, OSA director Ledford briefed Vance on details of the scheme, designated Operation BLACK SHIELD, which had been drawn up for Far East operations. It called for three aircraft to be deployed to Okinawa for sixty-day stints twice a year. About 225 personnel would be involved.142
During that period, North Vietnam began to deploy SAMs around Hanoi, threatening U.S. reconnaissance capability. In early June, McNamara inquired about the use of OXCARTs as substitutes for U-2s and was told BLACK SHIELD could operate over Vietnam as soon as the planes could be certified for use.143
With deployment expected to take place in the fall, the detachment went about demonstrating the reliability of the aircraft and its systems at Mach 3.05 and 2,300 nautical miles. But longer flights at higher speeds and temperatures resulted in new problems, the most serious being with the electrical wiring system. Such problems prompted CIA program manager John Parangosky to visit Kelly Johnson in early August. Johnson decided to spend full-time at the site in order to get the job finished quickly.144
Four primary BLACK SHIELD aircraft were selected, and final validation flights conducted. OXCART achieved a maximum speed of Mach 3.29, an altitude of 90,000 feet, and sustained flight time above Mach 3.2 for 1 hour and 14 minutes. As a result, Johnson wrote Ledford that “my considered opinion is that the aircraft can be successfully deployed for the BLACK SHIELD mission with what I would consider to be at least as low a degree of risk as in the early U-2 deployment days. . . . I think the time has come when the bird should leave its nest.”145
However, higher authorities did not agree. Their reservations were not technical but political. Two days after Johnson’s letter to Ledford, the 303 Committee, the NSC group responsible for reviewing sensitive intelligence operations, received a formal proposal to unleash the OXCART over North Vietnam. It was but one of several proposals for OXCART operations that the 303 Committee vetoed. Vice-Admiral William Raborn, who had replaced McCone as DCI in late April 1965, raised the prospect of deploying OXCART to Okinawa at five 303 meetings during the first half of 1966 but always failed to win sufficient support.146
The Joint Chiefs of Staff and the PFIAB supported the CIA proposal, but top officials at State and Defense concluded that the political risks of basing aircraft in Okinawa outweighed any intelligence the OXCART might gather. On August 12, both groups presented their views to President Lyndon Johnson, who sided with the majority of the 303 Committee against deployment.147
Even before OXCART was given a go-ahead, the CIA was examining two possible successors. Project ISINGLASS envisioned a plane capable of speeds between Mach 4 and Mach 5 flying at 100,000 feet. A feasibility study by General Dynamics was completed in fall 1964, but OSA took no further action because the proposed aircraft would still be vulnerable to Soviet countermeasures.148
An even more radical proposal came in 1965 from McDonnell Aircraft under the designation Project RHEINBERRY (although some of the work apparently also fell initially under the ISINGLASS designation). The RHEINBERRY aircraft would be rocket-powered, launched from a B-52, and ultimately reach speeds as high as Mach 20 and altitudes of up to 200,000 feet.149
Favoring the proposal was General Bernard Schriever, who wanted to see ramjet technology developed but who was unsure the NRO would approve such an effort. He suggested to Wheelon that OSA might begin work on it, and the Air Force Systems Command would support the work. Wheelon raised the issue with Raborn, who raised it with McNamara, who told the DCI to forget it. Wheelon was not convinced it was needed, and the plane would be quite inflexible—capable of only one turn around the earth.150
Nor would it have produced much in the way of intelligence. After much effort, designers concluded that it was impossible to eliminate the shock wave created when the plane skipped along the atmosphere and impossible to photograph targets through the shock wave. The plane might have provided an exciting ride for the pilot but would have done nothing for intelligence analysts on the ground.151
*As noted earlier, at the time each CIA directorate was known as the “Deputy Directorate for.” In July 1965, each became the “Directorate of.” That change occurred during Wheelon’s tenure, and therefore the new titles are used from this point forward.
*There was also a land-based component to Briar Patch. Gene Poteat concluded that given the size of the radar and its probable high power, it should be possible to pick up its signal out to several hundred miles, regardless of where the signal was pointed. The signal would be scattered forward and over the horizon via a phenomenon known as tropospheric scatter. Based on intelligence that a Hen House was under construction a couple of hundred miles inland from Riga, Poteat located an island in the Baltic that appeared to be the right distance from the Hen House to install a tropospheric-scatter receiver that could intercept and continuously monitor the radar.
After extensive negotiations to gain access, OEL installed dual antennae, about fifty wavelengths apart, to reduce the expected atmospheric fading, and the receiver was put on automatic pilot. The Briar Patch system finally picked up the transmission from the targeted Hen House and every subsequent transmission. From monitoring the radar, the CIA learned that it traced U.S. satellites from the first orbit. It appeared that the Soviets had an “incredibly effective espionage network to tip off the Hen House when a U.S. intelligence satellite was about to be launched.” According to Poteat, when there was a lengthy hold of an impending launch from Vandenberg Air Force Base, the Hen House would switch off and come back on the air the instant the satellite lifted off from Vandenberg. (Gene Poteat, “Stealth, Countermeasures, and ELINT, 1960–1975,” Studies in Intelligence 42, 1 [Spring 1998]: 51–59.)
4
SPACE RECONNAISSANCE WARS
Bud Wheelon’s greatest and most lasting impact on the revitalized Directorate of Science and Technology (DS&T) and the CIA was in the area of space reconnaissance. During the fourteen months Herbert Scoville served as Deputy Director for Research, the conflict between the CIA and NRO over space reconnaissance had escalated from border skirmishes to war. That war would intensify dramatically between August 1963, when Wheelon assumed the helm at DS&T, and October 1965, when Brockway McMillan would depart the scene.
A variety of factors would fuel the conflict, including the relationship between the men as well as their personalities. There were also disputes relating to CORONA. But the most significant problem was the differing conceptions Wheelon, McMillan, and their organizations had of the roles of the CIA and NRO in the development and operation of space reconnaissance systems. The incompatibility of those views helped to intensify the bitter conflict over the development of new photographic and signals intelligence sat
ellites as well as the authority of the NRO and its director.
The bureaucratic bloodshed would be hard on the psyches and careers of several of those involved—including Wheelon and McMillan. But out of the chaos, order emerged in Washington—in the form of a new, and durable, agreement governing the CIA-NRO relationship. In addition, plans to develop two revolutionary reconnaissance systems would also safely survive the CIA-NRO wars, and the groundwork would be laid for a third revolutionary system. Years after the battles were over, if not forgotten, those systems would be producing valuable intelligence.
THE GREAT DIVIDE
One former CIA official described Wheelon as “the most acerbic . . . son of a bitch” he had ever met.1 Wheelon recalled being “pretty young . . . pretty impatient,” “brash,” and “full of himself.” He was “not tactful” with the “committee sitters” at the CIA. As result of his self-imposed time limit at the agency and the “extraordinary pressure” emanating from the Pentagon with respect to the reconnaissance issue, “amenities fell by the wayside.”2
Nor was McMillan a diplomat—particularly since his reading of the NRO charter convinced him he had been given full authority to manage the National Reconnaissance Program, subject only to the supervision of the Secretary of Defense. To complicate matters, there was already bad blood between the two men. Several years before, McMillan served as referee for a paper Wheelon had submitted to a prestigious technical journal. By the time the process was finished, each questioned the other’s intellectual honesty.3
The differences between the organizations and their view of their roles that existed during Scoville’s tenure carried over to the Wheelon years. That the Air Force element of the NRO was not an intelligence-producing organization and had no direct connection to one, such as DIA, continued to be a problem in the CIA’s view. In a meeting with McMillan during the interval between Scoville’s resignation and Wheelon’s becoming Science and Technology Chief, Deputy DCI Marshall Carter suggested that McMillan authorize a symposium for all his program directors and their deputies “to make abundantly certain that the people running our programs know that their sole purpose is to develop intelligence and not just be shooting another rocket in the air.”4
The Wizards of Langley Page 14
