When Deitchman arrived at ARPA, he took one look at Gouré’s work and realized the Rand analyst was simply repeating what Pentagon officials wanted to hear: that dropping bombs on insurgents was working, even though it was not. “I got hold of McNamara’s military assistant and turned off the prisoner of war interviews because that clearly was being slanted,” Deitchman recalled. “McNamara was getting a view of the war that was distorted from it.”
Deitchman believed the Rand work, and much of the social science research being conducted in Vietnam, lacked scientific rigor. His view was that social science, if treated more like the “hard” physical sciences, could be used for prediction. He wanted data, not entertaining lectures or self-reinforcing studies. Yet he found that growing opposition to American involvement in Vietnam, particularly on college campuses, was making it almost impossible for ARPA to find good academics to work for the Defense Department on Southeast Asia issues.
The year before, the debate over military support for universities exploded in the midst of campus unrest when it was revealed that the army, through the Special Operations Research Office at American University, was funding civilian researchers to study insurgency in Chile. Their connection to a military-funded project had not been disclosed. With student radicalism and opposition to the Vietnam War at a near fever pitch, Project Camelot, as the ill-fated work was called, triggered an avalanche of criticism in Latin America and the United States. Professors who took Defense Department funding were accused of promoting an imperialist agenda.
A tailor-made solution was offered by a company called Simulmatics Corporation, which promised to hire academics in their time off to work under contract to the Pentagon. Its co-founder Ithiel de Sola Pool enjoyed a reputation as a brilliant and politically savvy professor of political science at the Massachusetts Institute of Technology. He had close ties to senior national security figures, including Robert Komer, the former CIA official running the pacification program in Vietnam. Deitchman believed that Pool “could easily attract other well known scholars; among them were many who were experts on Vietnam, had been there before, spoke the language, and knew many of the key Vietnamese figures who could grant ‘access’ for research.” Deitchman figured that Simulmatics would allow ARPA to tap academic expertise, without actually having to work directly with universities. “As far as we in ARPA were concerned, a group like this had impeccable credentials and helped avoid many problems,” Deitchman later recounted in his memoir, The Best-Laid Schemes.
In 1966, ARPA gave Simulmatics a wide-ranging contract for social science work in Vietnam, and soon the first research teams started showing up in Saigon. It was to be one of the most disastrous contracts ever let under AGILE.
—
Simulmatics was created as a way of making money off human capriciousness, be it in elections or war. The company’s origins dated back to 1958, when William McPhee, a professor at Columbia University, developed a novel theory to predict television-viewing habits. McPhee pitched his work to Edward Greenfield, a New York businessman, who in turn introduced McPhee to Ithiel Pool. Greenfield and Pool liked the idea but thought elections were more promising as a business model, and they were right. Simulmatics shot to fame during the 1960 election of John F. Kennedy, when, in a series of reports to the Democratic National Committee, it correctly predicted voter habits. “This is the A-bomb of the social sciences,” Harold Lasswell, a Yale professor, said at the time, comparing Simulmatics’ work to a demonstration of the first nuclear chain reaction. Buoyed by that much-hyped success, Simulmatics began selling the services of what Harper’s Magazine labeled a “people machine” to government and private clients.
Simulmatics was selling exactly what Deitchman and proponents of military-funded social science seemed to want. Simulmatics had a “people machine” it wanted to test, and ARPA had people it wanted to test it on. Pool originally suggested that ARPA hire the company to run experiments covering such areas as “intelligence and population control” in a “laboratory province” in Thailand. “The basic idea as I see it is that a limited area of Thailand be used as a site for a major field test of security programs that the government of that country can undertake with American assistance,” Pool wrote. “Thailand is a country in which in certain areas the security problems are sufficiently real so that a field test is possible, as at the same time the government is sufficiently alert and active in meeting the threat so that sensible programs can be tried.”
While ARPA did not end up hiring Simulmatics in Thailand, it did give the company a wide-ranging contract for Vietnam, where the agency wanted a quick-reaction force that could deploy social scientists to answer questions or provide analysis on specific problems.
The Walter Slote study on the Vietnamese psyche, which had drawn inferences on an entire nation based on the dreams and sex lives of four men, was only the start. One of Simulmatics’ key employees was Joseph Hoc, a Vietnam-born Catholic priest who had been teaching at Boston College. Hoc worked on a report for “testing psychological warfare weapons” under contract to ARPA. “My research work has been carried out with the purpose of providing to the advanced research projects agency a possible means to predict and even to control human events in Viet Nam,” he wrote.
Hoc’s ideas were almost as nutty as Slote’s, though potentially more harmful. “It is possible to control the Vietnamese hamlet people and manipulate them through informal means of communications and to persuade them to react to a given situation in a desired manner,” Hoc wrote. Standard interview techniques used by social scientists are “not adequate,” he wrote. Hoc proposed to have Simulmatics pay villagers to spread false rumors in a hamlet, and then secretly record people’s reaction, or what he called “techniques of human manipulation.”
Several of those techniques were tried out with ARPA funding. Under Hoc’s guidance, Simulmatics ran a study testing “psychological weapons” in hamlets, some controlled by Vietcong forces, and others loyal to the South Vietnamese government. The “weapons” included an American-style chain letter that would be spread in hamlets to trick the Vietcong into rallying. Villagers balked at distributing the letter, believing it to be a Vietcong trick.
Simulmatics also sought to use Vietnamese faith in “prophecy and the power of holy men” by publishing and distributing five thousand copies of a booklet that prophesied the Vietcong defeat. By unfortunate coincidence the distribution took place just as the Tet Offensive was starting, an event the prophecy did not foresee. Several of the projects were, even Simulmatics admitted, a total failure, such as using folksingers to spread pro-government messages and creating cartoons with political messages. Most dismal, perhaps, was a “sorcerer’s project,” which had enlisted Vietnamese sorcerers—essentially local magicians—to sway villagers against the Vietcong. It failed because, as Hoc put it without a hint of irony, “the sorcerers did not say what they were supposed to say.”
Not surprisingly, Garry Quinn, an ARPA official, criticized Hoc’s report: the “variables were contaminated,” “error sources were not systematically investigated,” and “rules of inference were violated.” Hoc did not seem concerned by the criticism, reporting that Quinn, who kept a picture in his office of Snoopy’s famous admonition, “Curse You, Red Baron,” was just in a bad mood. “I don’t think that we should take it seriously,” Hoc said, proposing to continue the psychological warfare study. He was shocked when ARPA refused.
To head the Simulmatics office in Saigon, Pool hired Alfred de Grazia, a political scientist. During World War II, de Grazia had worked on propaganda and psychological operations; his academic career took a more unusual turn. In the early 1960s, de Grazia had sided with Immanuel Velikovsky, a best-selling author whose revisions of world history based on ancient myths had sent the scientific community into an uproar (among other ideas, Velikovsky claimed that Mars had left its orbit around 750 B.C.E. and almost collided with the earth). Regardless of the merits of Velikovsky’s theories, the selection of de Grazia, who had alienated
much of the academic community, was an odd choice for a program designed to recruit academics. The highly qualified social scientists Pool promised to send to Vietnam never materialized. Deitchman and other officials complained repeatedly that Simulmatics was sending to Vietnam unqualified, inexperienced people who seemed more intent on flouting military contracting regulations than conducting rigorous research. When told that spouses could not accompany researchers to Vietnam, one Simulmatics employee promptly put his wife on payroll as a researcher. In a letter, Deitchman railed against the “rank amateurishness” of Simulmatics employees.
Many projects failed from sheer incompetence. Military experts dismissed as amateurish the results of a study on the Vietcong amnesty program, known as Chiêu Hồi. A study on Vietnamese TV-viewing habits, run by a nurse, was conducted as if “someone had taken a book of rules about scientific methodology, then systematically violated each one.” Simulmatics sent what one ARPA official in Vietnam called “Briefcase Directors,” who traveled on short stints to the region while on break from university, bringing students with no regional or subject expertise to conduct fieldwork. Simulmatics let go the entire ARPA-trained Vietnamese interviewing team. Outraged, the fired employees staged a “Hate the U.S.” banquet and protested to the American ambassador.
Other complaints were more serious; at one point, an ARPA program manager back in Washington expressed concerns over reports that Simulmatics employees were “running around Vietnam with pistols, rifles and even automatic weapons.” The director of ARPA’s field unit in Vietnam confirmed that Simulmatics employees had asked for M16s and .38-caliber guns, but wrote that the request was denied. Perhaps the Simulmatics employees were just bragging about carrying weapons, the field director suggested. “Wonder what brand of pot they’re smoking?” he wrote.
De Grazia’s relationship with ARPA was soon no better than with academia. He began to send a series of increasingly irate letters back to Washington. The laundry list of complaints included everything from the military’s not providing transportation for Simulmatics employees (ARPA pointed out that Vietnam was, after all, a war zone) to broken copy machines. “If ARPA is to perform its vital job well, it has to watch the higher priorities of the struggle and avoid the chicken shit,” de Grazia wrote in one rant. When ARPA complained that Simulmatics’ reports were of poor quality and submitted without basic copyediting, de Grazia responded by pointing out the bad grammar used on a sign in the ARPA urinal: “Time after time I must stand facing it and read ‘If you want them to work, do not put cigarette butts in the urinals.’ ” Pool later conceded that de Grazia did not “pan out.”
By 1967, relations between Simulmatics and ARPA had reached a breaking point. Someone at ARPA drafted an entire list of Simulmatics’ failures, ranging from unqualified personnel to general incompetence. Pool called it “a malicious set of baseless stories.” In December, W. G. McMillan, the science adviser to the U.S. Military Assistance Command, Vietnam, wrote to Deitchman with a simple message: terminate the Simulmatics contract. “This Corporation has been operating in the Republic of Vietnam for approximately eighteen months attempting to perform social science research under ARPA contracts,” he wrote. “Review of these efforts makes it clear that Simulmatics has failed to meet contractual requirements.”
In interviews and in his book, Deitchman said simply that the Simulmatics contract did not work out for administrative reasons. His correspondence in the archives tells a much franker story. “One aspect of it may be that their severest critics are looking for sound scientific effort backed up by numbers and rigorous methodology,” he wrote. “Frankly, I don’t think we’ll ever get that from Simulmatics.”
A succession of ARPA and Pentagon officials demanded repeatedly that the contract be canceled for gross incompetence. The Simulmatics “scientific approach” to counterinsurgency had not succeeded, and though he knew there would be blowback, Deitchman finally ended the contract in early 1968. Edward Greenfield, the head of the company, showed up in Deitchman’s office, saying he would complain to McNamara; it was no idle threat given the company’s close ties to senior administration officials. Deitchman’s reaction was identical to what he told Herman Kahn: go ahead.
Almost bankrupt, Simulmatics made one last desperate attempt to get Pentagon money: the company in 1968 turned to William Godel, who was out of prison. Godel, whose post-ARPA career had included gunrunning in Southeast Asia, still had contacts with top Asian officials, particularly in Thailand. He worked out a deal with the head of the Thai air force to request that ARPA fund a program to study regional police and security forces. Simulmatics would run the contract. “Greenfield told me Godel would likely be associated with Simulmatics through a third-company arrangement,” Deitchman wrote to a colleague.
Deitchman knew he was in a bind. If he refused a senior Thai official, it would put ARPA’s work in Thailand in jeopardy. If ARPA agreed, it would mean putting Godel back on the agency’s payroll. In a memoir covering his years at ARPA, Deitchman recalled the Simulmatics episode only briefly, as one of generally good work but administrative missteps. ARPA, he wrote, was forced to terminate the contract for “bureaucratic reasons.”
His official correspondence, much of it classified at the time, told a different story. The files include dozens of lengthy memos chronicling the Simulmatics disaster. The return of Godel was apparently the final straw. “I ain’t having any,” Deitchman wrote. “I view it as an affront to the integrity of the ARPA program.” Deitchman swore that Simulmatics would not see another penny from ARPA, regardless of the political repercussions. He ended with a request that was never fulfilled, because his original letter was lying in the National Archives more than forty years later: “Burn this after reading.”
—
From start to finish, the Simulmatics experiment in Vietnam lasted barely eighteen months. Its work was a disaster, leaving a trail of incompetence, shoddy research, and politically disastrous missteps. If there was a lesson to be learned from Simulmatics, or ARPA’s attempt to solve the problem of insurgency using social science, it was that trying to study and influence human behavior is infinitely more difficult than collecting data on the flight of a ballistic missile. Deitchman, in acknowledging the failure, called it a form of the Heisenberg principle: “The fact and means of measurement and observation affect and change the phenomena being observed and all the participants.” In other words, humans will realize they are being observed, and they will adapt their behavior.
Nothing illustrated this truism better than Deitchman’s proposal to lighten the ninety-pound load that soldiers carried into the field. Much of that weight was due to inefficiency, Deitchman decided, because soldiers were carrying redundant items. If combat patrols operated as a “system,” rather than a collection of individual soldiers, then they could divide up the equipment more efficiently. One soldier, for example, could carry communications equipment, and another, extra ammunition. At first, the idea seemed to work, but then an interesting thing happened: soldiers began to use the extra room to load up on cans of Coke, and their packs got right back up to ninety pounds. “As we thought about that, we said, ‘Well, you know, whatever satisfies them,’ ” Deitchman recalled. “ ‘If they are getting nourishment out of that, maybe that’s okay.’ ”
For all of the criticism that Deitchman directed at Godel’s approach to counterinsurgency—and much of the criticism was valid—Deitchman’s attempt to transform social science into a hard science was also a failure. ARPA’s counterinsurgency work, though largely unsuccessful, rolled forward out of classic bureaucratic inertia. It had become central to ARPA—its third-largest program after missile defense and nuclear test detection—so to admit failure would have been to resign from a core mission. But with the Vietnam War ramping up, Congress was growing weary of ARPA’s meddling in world affairs and its sponsorship of social science research, which seemed to be a strange preoccupation for a military agency. In addition to its Vietnam work, ARPA was
studying childhood nutrition in the Middle East, and it was measuring the shape and size of Iranian soldiers’ heads and feet to design better uniforms. For critics of the Vietnam War, the ARPA programs were more grist for the mill.
“How does this get to be your responsibility?” demanded Glenard Lipscomb, a Republican representative from California, in one hearing reviewing AGILE’s expansion. “Why is this not the responsibility of the State Department, or the regular military departments?”
Because no one else was doing it, argued Herzfeld, an answer that did little to assuage the congressional doubters. In fact, ARPA was expanding AGILE still further around the world. “We are planning to concentrate more on the non-southeast Asia part of AGILE,” Herzfeld told Congress, when asked for the agency’s plans in 1967.
“How many countries would you expect to go into?” Lipscomb demanded. “How large an operation is Project AGILE to become? There seems to be no end of it.”
Herzfeld argued that ARPA’s expansion was justified, because its work was going well. “I think to some extent, Mr. Chairman, we are breaking ground here for a new way of looking at insurgency, how to stop insurgency while it is small,” Herzfeld told lawmakers. “This is absolutely a major military problem for the United States and it is largely unsolved. We were not able to stop insurgency while it was small in Vietnam. It got very big. It is not insurgency now but a war.”
That counterinsurgency strategies in Vietnam had utterly failed to stop the escalation into a full-scale war did not enter into the discussion. When another lawmaker asked Herzfeld if he saw an end to the war in Vietnam, and if so, when and how, Herzfeld responded with enthusiasm. “I am quite convinced, if one compares what is going on now with what was going on three or four years ago, that on the military side we are now really winning,” he said. “On the civilian side I think we have stopped the decay and we are gradually pulling up. This coupled with the winning on the military side, makes me quite convinced that we will win.”
The Imagineers of War Page 21