Kill Chain: The Rise of the High-Tech Assassins

by Andrew Cockburn

  With the war’s end, the Army Security Agency (ASA), the vast organization that intercepted and decoded signals from around the world, remained largely in place. But the men of “broad education” with a “fair share of astuteness, skepticism and desire to solve puzzling problems” who had given the system its brains, largely went back to their law firms and other civilian careers. These included McCormack himself. Following an effort to re-create Special Branch inside the State Department that was crushed by bureaucratic opposition from within the department, he too returned to his practice. Clarke himself was head of the Army Security Agency, and still a powerful figure, but his chief patron, General Marshall, had left the army and could no longer protect him.

  Meanwhile, William Donovan, chief of the OSS, had nurtured ambitions to turn his organization into a permanent peacetime civilian intelligence agency and had penned a lengthy detailed memo to President Roosevelt suggesting just that. But the memo was rapidly leaked to Walter Trohan, Washington bureau chief of the right-wing Chicago Tribune, who duly reported that Donovan planned to establish a “gestapo,” which would threaten civil liberties. Donovan and others assumed that the leaker was J. Edgar Hoover, protecting the FBI’s bureaucratic prerogatives. However, Trohan assured me in 1985 that the source was in fact Steve Early, Roosevelt’s press secretary, who handed him the memo with the words “the boss wants this out.” The OSS was disbanded shortly afterward. Nevertheless a civilian intelligence agency did eventually emerge, the Central Intelligence Group, which in 1947 became the CIA.

  The peacetime agency retained its unequal status vis-à-vis the military. Initially it was manned not only with veterans of the OSS but also with officers detached from military intelligence. As can be the case with any new organization staffed from existing departments, transferees were not necessarily the best or the brightest. As a former military intelligence officer once snidely recalled early CIA staffing to me, “They walked the halls of the Pentagon, calling ‘bring out your dead wood!’” Like the OSS, the successor organization lacked any control over communications intelligence and the ever-growing budget that went with it, which remained in the firm grasp of the military, reorganized into the National Security Agency (NSA) in 1952. Only gradually and grudgingly was the CIA permitted access to the actual intercepts.

  With the expanding role of ever more costly technical intelligence systems such as surveillance satellites, the CIA’s share of the overall cold war intelligence budget, though vast in itself, was a small portion of the amount allotted to the Pentagon. Even when the agency did take a successful technical intelligence initiative with the brilliantly conceived U-2 reconnaissance plane, developed for a total cost of $19 million, it fell under the effective control of the Pentagon, which assigned the targets—bomber bases, missile silos, etc.—useful for military estimates but not for the CIA’s allotted task of assessing Soviet intentions. Even more secretly, and dangerously, the commander of the Strategic Air Command, General Curtis LeMay, had his own fleet of reconnaissance planes scanning Russia in the 1950s. Should he conclude that the Soviets were planning an attack, he blithely informed a shocked White House emissary in 1958, he would immediately launch a pre-emptive nuclear strike without reference to higher authority.

  So, while a perennial junior partner in the overall U.S. intelligence apparatus, the agency found function and purpose in the field of undercover clandestine operations—espionage, subversion, dirty tricks, and manipulation, among others—justified as necessary to maintain America’s political and economic influence in the non-Communist world. The Communist world was another matter. Despite constant effort, until late in the cold war the agency had little success in penetrating the iron curtain. Repeating the precedent of Britain’s wartime SOE, the agency recruited hundreds of unfortunate Koreans and Chinese as agents and dropped them into North Korea and China during the Korean War, never to be heard of again. “We were following in the footsteps of the OSS,” Don Gregg, a young CIA officer engaged in the exercise, later bitterly recalled. “We didn’t know what we were doing. I asked our superiors what the mission was and they wouldn’t tell us. They didn’t know what the mission was. It was swashbuckling of the worst kind.”

  In parts of the world not under the iron control of the Communists, the agency found a role as the undercover instrument of U.S. control, mainly thanks to its access to ready cash and its willingness to dispense it in large quantities. Thus the CIA’s first significant success was to inject money (siphoned off from the vaunted Marshall Plan) into the Italian elections of 1948 in sufficient amounts to defeat the very popular Communist Party. Cash was similarly the crucial factor in the 1953 coup that overthrew Mohamed Mossadeq, the democratically elected prime minister of Iran. CIA officer Kermit Roosevelt financed a network originally assembled by British intelligence that in turn recruited mobs and army units that forced Mossadeq to resign, and ushered in a quarter century of U.S.-supported rule by the shah.

  Iran fostered the CIA’s legend among Washington’s power elite as a silver bullet for fighting the underground cold war. The coup in Guatemala in 1954 solidified it. Yet in both cases, as one historian of the agency summarized, it was “bribery, coercion, brute force, not secrecy, stealth, and cunning” that won the day. The Guatemalan operation, in particular, was marked by bumbling and misadventure, beginning when plans for the coup mislaid by a CIA agent in a Guatemala City hotel room were splashed across every newspaper in the Western world. Not included among those documents, however, were CIA plans for “low methods” in Guatemala.

  Starting in 1952, according to internal agency documents, senior officials in the euphemistically named Directorate of Plans were compiling lists of “top flight communists whom the new government would desire to eliminate in event of successful anti-communist coup.” In a later initiative, headquarters ordered the coup-plotters to train two “assassination specialists,” a move encouraged by the State Department, while also demanding that a “list of names be compiled for study by staff officers to determine if they meet the latest criteria for inclusion on the Junta’s disposal list … it is requested that a final list of disposees be approved promptly to permit planning to proceed on schedule.”

  To aid in training the specialists, someone at headquarters helpfully came up with “a study of assassination” complete with scholarly references to the derivation of the operative word: “thought to be derived from ‘hashish,’ a drug similar to marijuana.” The guide provided useful justifications for the deed: “Killing a political leader whose burgeoning career is a clear and present danger to the cause of freedom may be held necessary.” With regard to practical techniques, the guide advised against a manual approach—“it is possible to kill a man with the bare hands, but very few are skillful enough to do it well”—while recommending the “contrived accident … a fall of 75 feet or more onto a hard surface” as the most effective technique. Other methods discussed in detail included drugs, “edge” [sharp] weapons, blunt weapons, though these “require[d] some anatomical knowledge for effective use,” and firearms of various types.

  In the event, according at least to the agency’s own review of the episode, none of the listed “disposees” actually were eliminated. On the other hand, human rights groups estimate that the military regimes ushered in by the 1954 coup killed at least one hundred thousand civilians over the next thirty-five years. As the unknown author of the little manual observed: “Assassination can seldom be employed with a clear conscience. Persons who are morally squeamish should not attempt it.”

  Whereas the quest for victims in Guatemala had been merely a matter of prudent housekeeping, the obsessive targeting of Fidel Castro, leader of the Cuban revolutionaries who overthrew a U.S.-allied dictator in 1959, was something entirely different. The eagerness with which the Kennedy brothers urged the CIA to find a way to kill the Cuban leader, who had turned to the Soviet Union for aid and protection, would not be matched until the hunt for Osama bin Laden a generation later. According
to Castro’s longtime bodyguard, Fabien Escalante, there were no less than 638 separate plots on the Cuban leader’s life, most of them ultimately directed from the CIA. They ranged from the juvenile, an exploding cigar, to the criminal, as in the infamous liaison with mafia hoodlums such as “Handsome Johnny” Roselli and Santo Trafficante, supposedly eager and ready to organize the “hit.” There can be little doubt as to the direct personal involvement of the president of the United States. As Richard Bissell, the CIA’s deputy director for plans later testified, the orders to set up an assassination squad, known as the Special Group (Augmented), came directly from National Security Adviser McGeorge Bundy and his deputy, Walt Rostow, and “the President’s men would not have given such encouragement unless they were confident it would meet with the President’s approval.”

  The Cuban and Guatemalan initiatives were by no means the only assassination plots concocted in the early cold war years. A partial list includes Indonesian President Sukarno, Kim Il-Sung of North Korea, Iranian Prime Minister Mohammed Mossadeq, Philippines opposition leader Claro Recto, Indian Prime Minister Jawaharlal Nehru, General Rafael Trujillo of the Dominican Republic, Francois “Papa Doc” Duvalier of Haiti, and Egyptian leader Gamal Abdel Nasser. Such plans were usually obscured with euphemisms such as “executive action,” an evasion disdained by British Prime Minister Anthony Eden in commissioning his own plot against Nasser during the 1956 Anglo-Egyptian Suez crisis. “I want him destroyed, don’t you understand?” Eden bellowed down the phone to his intelligence chief. “I want him murdered.”

  Until the 1960s, the agency’s lethal and paramilitary activities had been largely restricted to variations of the hit-and-run attacks outlined above. (An attempt at something more ambitious, sponsoring the ill-conceived 1961 invasion of Cuba with a force of anti-Castro exiles—the Bay of Pigs—had ended in disaster.) However, expanding wars in Southeast Asia offered fresh scope for such missions. In Laos, for example, a supposedly neutral country where the United States was officially not involved, the CIA ran a full-scale “secret war” using foot soldiers drafted from the Hmong and other indigenous tribes, including many children, some as young as eight years old. This was certainly not a war for the morally squeamish. Anthony Posephny, more commonly known as “Tony Po,” who commanded the Hmong effort, once cheerily reminisced to me about the time the U.S. embassy in Vientiane, the Laotian capital, had queried his “body count” of eliminated enemies. “So I sent them a head-count,” he chortled, “heads in a bag. But the bag arrived on a Friday and sat in an un-air-conditioned office all weekend until a secretary opened it up on Monday morning. I got into trouble over that!”

  The CIA’s secret war in Laos was certainly a big operation, costing possibly as much as $100 million a year in 1969 dollars. (At its hub was a base, Long Tien, that grew to a city with a population of 40,000 before an enterprising reporter, T. D. Allman, revealed its existence.) Even so it paled in scope when compared to the U.S. Air Force’s campaign in another part of Laos, the automated battlefield of Igloo White and Task Force Alpha, which had cost $2 billion simply to set in motion. However, the CIA was by no means excluded from the trend toward automated warfare, complete with powerful computers. Beginning in January 1967, U.S. officials in Saigon began feeding a list of 3,000 suspected VCI (Viet Cong Infrastructure), into an IBM-1401 computer installed in the “political order of battle” section of the Combined Intelligence Center in Saigon. Officially called the Viet Cong Infrastructure Information System, it has been referred to as the first computerized blacklist. By no coincidence, the initiative came to life at the same time as Task Force Alpha was throwing its invisible net over the Ho Chi Minh Trail.

  The ill-fated electronic fence had been born out of Secretary McNamara’s frustration at the failure to win the war by bombing the “critical node” of North Vietnamese oil tanks early in 1966. At that same time, the technocratic defense secretary had commissioned a study of means to systematize the counter-guerrilla effort in South Vietnam, eventually spawning the Intelligence Coordination and Exploitation for Attack Against the Infrastructure Program (ICEX).

  Just as sensors dropped into the jungle canopy over the Ho Chi Minh Trail would relay signals to the omnivorous IBM-360 machine in Thailand, teams of American and Vietnamese intelligence officials fed dossiers into the IBM-1401 computer in the vast, air-conditioned Saigon intelligence center. Before long the list had grown to 6,000, expanding at a rate of 1,000 a month. In August of the following year, Robert “Blowtorch” Komer, the White House official who had been promoted to supervise the overall pacification effort, mandated a quota of eighteen hundred VCI “neutralizations” per month as a “management tool.” So while Igloo White measured success by the number of Vietnamese supply trucks destroyed, the sister program inside South Vietnam counted bodies.

  The CIA’s neutralizers were initially known as counterterror teams. Rebranded soon after as the functionally equivalent provincial reconnaissance (PRU), they were made up of native Vietnamese who were overseen by American advisers. “Sure we got involved in assassinations,” one CIA official, Charlie Yothers, later told a reporter. “That’s what PRU was set up for—assassination. I’m sure the word never appeared in any outlines or policy directives, but what else do you call a targeted kill?” News of the program inevitably was leaked to the press, prompting eventual congressional hearings. Just as inevitable, the revelations triggered a concerted effort to portray Phoenix (only one of the successive names for the program) as something much more benign. Thus in 1969 the New York Times described it as an initiative to “sideline” members of the Viet Cong infrastructure.

  By 1971 euphemism had been cast aside. That year, after William Colby, later head of the CIA, confirmed that 20,587 people had already been killed under the auspices of Phoenix, members of Congress were openly deriding it as “a program for the assassination of civilian leaders.” Alternatively, the program met with wholehearted approval at the other end of Pennsylvania Avenue, where President Richard Nixon reacted angrily to proposed cuts in the PRU program with the straightforward command, “We’ve got to have more of this. Assassinations, that’s what they, the Vietnamese communists, are doing.”

  The statistics of eliminated Viet Cong “infrastructure” regurgitated by the computers, summarized in the totals solemnly recounted by Colby and other senior officials, gave the appearance of precision and progress, just as the even more powerful computers of Task Force Alpha revealed equally exact numbers of trucks accounted for on the Ho Chi Minh Trail. As with the truck war, the impression of precision cast by the Phoenix computers was misleading. Though large numbers of people were indeed being killed, they were not necessarily the bona fide Communist officials posthumously listed in the account books. Even the program’s own statistics indicate that enemy “civilian leaders” weren’t in much danger from Phoenix: officially only 150 “Senior VCI” were “neutralized” in 1969. When Komer asked a friend, Colonel Robert Gard, to be the military deputy to the Phoenix program, Gard flatly refused. As he told author Nick Turse years later, “I didn’t know a lot about it, except that it was an assassination program, subject to killing innocents.” Unsurprisingly, it was well penetrated by Viet Cong agents who could thereby use it to work through their own blacklists. One CIA officer later recalled that Saigon regime officials and even Americans used the PRU for “shaking down the Vietnamese, arresting them if they didn’t pay protection money, even taking bribes to free suspects even if they’d already been arrested.”

  Vincent Okamoto, later a distinguished Los Angeles Superior Court Judge, who in 1969 was a highly decorated combat veteran assigned to a Special Forces unit, has a pungent reminiscence vividly illustrating the chasm between the computerized high-value targeting concept and the murderously haphazard reality of Phoenix in action. “I had never heard of it until they told me I was part of it. I did it for two months and didn’t like it at all … the CIA and the boys in Saigon would feed information into computers and would come u
p with a blacklist of Vietnamese who were aiding the enemy. The problem was, how do you find the people on the blacklist? It’s not like you had their address and telephone number. The normal procedure would be to go into a village and just grab someone and say, ‘Where’s Nguyen so-and-so?’ Half the time, the people were so afraid they would say anything. Then a Phoenix team would take the informant, put a sandbag over his head, poke out two holes so he could see, put commo wire around his neck like a long leash, and walk him through the village and say, ‘When we go by Nguyen’s house, scratch your head.’ Then that night Phoenix would come back, knock on the door and say, ‘April Fool, motherfucker.’ Whoever answered the door would get wasted.”

  As America withdrew from Vietnam, the assassination program was widely deemed an embarrassment, if not a war crime, and best forgotten. But as the years passed and moral outrage receded along with guilty consciences, attitudes toward Phoenix gradually became more positive. In 2004, for example, David Kilcullen, a highly lauded and influential theorist of counterinsurgency, called for a “disaggregation strategy” targeting insurgent networks on a global scale “that would resemble the unfairly maligned (but highly effective) Vietnam-era Phoenix program.”