Surveillance Valley

by Yasha Levine

  Pool saw computers as more than just apparatuses that could speed up social research. His work was infused with a utopian belief in the power of cybernetic systems to manage societies. He was among a group of Cold War technocrats who envisioned computer technology and networked systems deployed in a way that directly intervened in people’s lives, creating a kind of safety net that spanned the world and helped run societies in a harmonious manner, managing strife and conflict out of existence. This system wouldn’t be messy or wishy-washy or open to interpretation; nor would it involve socialist economic theories. In fact, it wouldn’t involve politics at all but would be an applied science based on math, “a kind of engineering.”

  In 1964, at the same time his company was doing counterinsurgency work for ARPA in Vietnam, Pool became a vocal supporter of Project Camelot, a different counterinsurgency effort funded by the US Army and backed in part by ARPA.79 “Camelot” was just a code name. The project’s full official title was “Methods for Predicting and Influencing Social Change and Internal War Potential.” Its ultimate goal: to build a radar system for left-wing revolutions—a computerized early warning system that could predict and prevent political movements before they ever got off the ground.80 “One of the project’s anticipated end products was an automated ‘information collection and handling system’ into which social researchers could feed facts for quick analysis. Essentially, the computer system would check up-to-date intelligence information against a list of precipitants and preconditions,” writes historian Joy Rohde. “Revolution could be stopped before its initiators even knew they were headed down the path to political violence.”81

  Project Camelot was a big undertaking that involved dozens of leading American academics. It was very dear to Pool personally, but it never got very far.82 Chilean academics who were invited to participate in Project Camelot blew the whistle on its military intelligence ties and accused the United States of trying to build a computer-assisted coup machine. The affair blew up into a huge scandal. A special session of the Chilean Senate was convened to investigate the allegations, and politicians denounced the initiative as “a plan of Yankee espionage.”83 With all this international attention and negative publicity, Project Camelot was shut down in 1965.

  In 1968, Lick’s Cambridge Project at MIT picked up where Camelot left off.84

  To Lick, the Cambridge Project was the realization of the interactive computer technology he had been pushing for. Finally, after nearly a decade, computing technology had advanced to a point where it could help the military use data to fight insurgencies. The Cambridge Project included several components. It ran a common operating system and a suite of standard programs custom-tailored to the military’s “behavioral science mission” that could be accessed from any computer with an ARPANET connection. It was a kind of stripped-down 1960s version of Palantir, the powerful data mining, surveillance, and prediction software the military and intelligence planners use today. The project also funded various efforts to use these programs in ways that were beneficial to the military, including compiling various intelligence databases. As a bonus, the Cambridge Project served as a training ground for a new cadre of data scientists and military planners who learned to be proficient in data mining on it.

  The Cambridge Project had another, less menacing side. Financial analysts, psychologists, sociologists, CIA agents—the Cambridge Project was useful to anyone interested in working with large and complex data sets. The technology was universal and dual use. So, on one level, the goal of the Cambridge Project was generic. Still, the project was customized to the military’s needs, with particular focus on fighting insurgencies and rolling back communism. A big part of the proposal Lick submitted to ARPA in 1968 focused on the various types of “data banks” the Cambridge Project would compile and make available to military analysts and behavioral scientists connected through the ARPANET:85

  • Public opinion polls from all countries

  • Cultural patterns of all the tribes and peoples of the world

  • Archives on comparative communism… files on the contemporary world communist movements

  • Political participation of various countries.… This includes such variables as voting, membership in associations, activity of political parties, etc.

  • Youth movements

  • Mass unrest and political movements under conditions of rapid social change

  • Data on national integration, particularly in “plural” societies; the integration of ethnic, racial and religious minorities; the merging or splitting of present political units

  • International propaganda output

  • Peasant attitudes and behavior

  • International armament expenditures and trends

  It was clear that the Cambridge Project wasn’t just a tool for research, it was counterinsurgency technology.

  In the late 1960s and early 1970s, huge antiwar protests erupted on university campuses across the country. Activists occupied buildings, stole documents, published newsletters, staged sit-ins and marches, clashed with police, and became increasingly violent. At the University of Michigan, students attempted to block campus recruitment by Dow Chemical, which produced the napalm that rained down on Vietnam.86 Someone blew up the Army Mathematics Research Center at the University of Wisconsin.87 The Weather Underground set off a bomb inside Harvard’s Center for International Affairs.88 They wanted to stop the Vietnam War. They also wanted to halt the cooptation of academic research by the military-industrial complex.

  ARPA programs were a constant target. Students protested against the ILLIAC-IV, the massive ARPA supercomputer housed at the University of Illinois.89 They targeted the Stanford Research Institute, an important ARPA contractor involved in everything from chemical weapons research to counterinsurgency work and development of the ARPANET. Students occupied the building, shouting, “Get SRI out!” and “Down with SRI!” A few brave contractors stayed behind to protect ARPA’s computers from the angry mob,90 telling protesters that computers were “politically neutral.”91 But are they?

  The student demonstrations against the Cambridge Project were part of this wave of protests sweeping the country. The common belief among students at MIT and Harvard was that the Cambridge Project, and the bigger ARPA network it was tied to, was essentially a front for the CIA. Even some professors began turning on it.92 The language of Licklider’s proposal—talk about propaganda and monitoring political movements—was so direct and so obvious that it could not be ignored. It confirmed students’ and activists’ fears about computers and computer networks and gave them a glimpse into how military planners wanted to use these technologies as tools for surveillance and social control.

  A crew of activists from Students for a Democratic Society produced a small but informative booklet that laid out the group’s opposition to the initiative: The Cambridge Project: Social Science for Social Control. It sold for a quarter. The cover featured a series of punch cards being fed into a computer that transformed “Black Militancy,” “Student Protest,” “Strikes,” and “Welfare Struggles” into “Counterinsurgency,” “Ghetto Pacification,” and “Strike Breaking.”93 At one point, the pamphlet’s producers gathered on Technology Square at the edge of the MIT campus. They obtained a copy of Lick’s Cambridge Project proposal and set fire to it. Lick, ever enthusiastic and confident in his ability to sway people to his way of thinking, met the protesting students outside and attempted to reassure them that everything was okay—that this ARPA project wasn’t some nefarious initiative cooked up by spies and generals. But students would have none of it.

  “The group was hostile,” Douwe Yntema, the director of the Cambridge Project, told M. Waldrop.94 “But he [Licklider] was pretty cool about it. At one point, in fact, they had a copy of the proposal and tried to set fire to it—not very successfully. Well, after a few minutes Lick said, ‘Look, if you want to burn a stack of paper, don’t just try to light it. Spread the pages out first.’ So he showed them
how, and it really did burn much better!”

  But the students gathered there had a deep understanding of the political and economic dimensions of ARPA’s military research, and they were not going to be dismissed like petulant schoolchildren. They persisted. Lick tried to be a good sport about it, but he was disappointed.95 Not in the project. No, he was down on the kids. He believed the protesters did not understand the project and completely misread its intentions and military ties. Why couldn’t young people understand that this technology was completely neutral? Why did they have to politicize everything? Why did they think America was always the enemy and would use technology for political control? He saw the whole thing as a symptom of the degradation of American youth culture.

  The demonstrations against the Cambridge Project involved hundreds of people. They were ultimately a part of the larger antiwar movement at MIT and Harvard that attracted the leading lights of the antiwar movement, including Howard Zinn. Noam Chomsky showed up to lambast academics, accusing them of running cover for violent imperialism by “investing it with the aura of science.”96 But in the end, the protests didn’t have much of an effect. The Cambridge Project proceeded as planned. The only change: further proposals and internal discussions for funding omitted overt references to military applications and the study of communism and third world societies, and project contractors simply referred to what they were doing as “behavioral science.”

  But behind the scenes, the military and intelligence dimension of the project remained foremost. Indeed, a classified guide from 1973 commissioned by ARPA for the Central Intelligence Agency noted that, although the Cambridge Project was still experimental, it was nonetheless “one of the most flexible” tools available for complex data and statistical analysis in existence, and recommended that the CIA’s international security analysts learn how to use it.97

  The Cambridge Project ran for a total of five years. As time would prove, the kids were right to fear it.

  Chapter 3

  Spying on Americans

  Historical mythmaking is made possible only by forgetting.

  —Nancy Isenberg, White Trash

  On June 2, 1975, NBC correspondent Ford Rowan appeared on the evening news to report a stunning exposé. Baby-faced with light blue eyes, he spoke straight into the camera and told viewers that the military was building a sophisticated computer communications network and was using it to spy on Americans and share surveillance data with the CIA and NSA.1 He was talking about the ARPANET.

  “Our sources say, the Army’s information on thousands of American protesters has been given to the CIA, and some of it is in CIA computers now. We don’t know who gave the order to copy and keep the files. What we do know is that once the files are computerized, the Defense Department’s new technology makes it incredibly easy to move information from one computer to another,” Rowan reported. “This network links computers at the CIA, the Defense Intelligence Agency, the National Security Agency, more than 20 universities, and a dozen research centers, like the RAND Corporation.”

  Rowan had spent months piecing the story together from several “reluctant whistleblowers”—including ARPA contractors who were alarmed at how the technology they were building was being used. For three days after the initial story, he and his colleagues at NBC evening news aired several more reports looking more closely at this mysterious surveillance network and the shadowy agency that had built it.

  The key breakthrough in the new computer technology was made at a little known unit of the Defense Department—the Advanced Research Projects Agency, ARPA.

  ARPA scientists created something new in computer communications with this device, it’s known as the IMP, the interface message processor. Different computers communicate in different computer languages. Before the IMP it was enormously difficult, in many cases impossible, to link the various computers. The IMP, in effect, translates all computer messages into a common language. That makes it very very easy to tie them into a network.

  The government is now using this new technology in a secret computer network that gives the White House, the CIA, and the Defense Department access to FBI and Treasury Department computer files on 5 million Americans.

  The network, and it is referred to as “the network,” is now in operation.… This means that from computer terminals now in place at the White House, the CIA, or the Pentagon, an official can push a button and get whatever information there might be on you in the FBI’s vast computer files. Those files include records from local police agencies which are hooked to the FBI by computer.2

  Rowan’s exposé was phenomenal. It was based on solid sources from the Pentagon, the CIA, and the Secret Service, as well as key ARPANET insiders, some of whom were concerned about the creation of a network that could so seamlessly link multiple government surveillance systems. In the 1970s, the historical significance of the ARPANET was not yet apparent; what Rowan uncovered has become only more relevant in hindsight. It would take more than twenty years for the Internet to spread into most American homes, and four decades would pass before Edward Snowden’s leaks made the world aware of the massive amount of government surveillance happening over the Internet. Today, people still think that surveillance is something foreign to the Internet—something imposed on it from the outside by paranoid government agencies. Rowan’s reporting from forty years ago tells a different story. It shows how military and intelligence agencies used the network technology to spy on Americans in the first version of the Internet. Surveillance was baked in from the very beginning.

  This is an important fact in the history of the Internet. Yet it has vanished down the collective memory hole. Crack any popular history of the Internet and there is no mention of it. Even today’s foremost historians do not seem to know it occurred.3

  Counterinsurgency Comes Home

  In the late 1960s, as engineers at MIT, UCLA, and Stanford diligently worked to build a unified military computer network, the country convulsed with violence and radical politics—much of it directed at the militarization of American society, the very thing that the ARPANET represented. These were some of the most violent years in American history. Race riots, militant black activism, powerful left-wing student movements, and almost daily bombings in cities across the country.4 The United States was a pressure cooker, and the heat kept building. In 1968, Robert Kennedy and Martin Luther King Jr. were assassinated, the latter’s death triggering riots across the nation. Antiwar protests swept American university campuses. In November 1969, three hundred thousand people descended on Washington, DC, for the largest antiwar protest in the history of the United States.5 In May 1970, the Ohio National Guard fired on protesters at Kent State University, killing four students—called “Nixon’s Massacre” by Hunter S. Thompson.

  To many, it seemed that America was about to explode. In January 1970, a former military intelligence officer by the name of Christopher Pyle tossed more wood on the blaze.

  Pyle was political science PhD student at Columbia University. He wore glasses, had a mop of hair parted on the side, and carried himself with the thoughtful and meticulous manner of an academic. He had been an instructor at the US Army Intelligence School in Fort Holabird outside Baltimore and saw something there that concerned him enough that he had to blow the whistle.6

  In early 1970, he published an exposé in the Washington Monthly that revealed a massive domestic surveillance and counterinsurgency operation run by the US Army Intelligence Command. Known as “CONUS Intel”—Continental United States Intelligence—the program involved thousands of undercover agents. They infiltrated domestic antiwar political groups and movements, spied on left-wing activists, and filed reports in a centralized intelligence database on millions of Americans.7 “When this program began in the summer of 1965, its purpose was to provide early warning of civil disorders which the Army might be called upon to quell in the summer of 1967,” reported Pyle. “Today, the Army maintains files on the membership, ideology, programs, and prac
tices of virtually every activist political group in the country.”

  CONUS Intel was masterminded in part by General William P. Yarborough, the army’s top intelligence officer at the time. He had a long, distinguished career in counterinsurgency and psychological operations, from World War II to the Korea and Vietnam conflicts. In 1962, General Yarborough took part in the influential US Army “limited war” counterinsurgency symposium held in Washington, DC, which J. C. R. Licklider also attended.8 Fear of a domestic insurgency was swirling in military circles, and the general was not immune. He came to believe that there existed a growing communist conspiracy to foment unrest and to overthrow the United States government from within. His evidence? The burgeoning civil rights movement and the surging popularity of Martin Luther King Jr.

  Yarborough looked at the masses of people agitating for racial equality and didn’t see Americans getting politically involved because of legitimate grievances and concerns. He saw dupes and foreign agents who, whether they realized it or not, were part of a sophisticated insurgency operation financed and directed by the Soviet Union. This was not the view of a lone conspiracy nut but was shared by many of Yarborough’s peers in the army.9

  When race riots broke out in Detroit in 1967 a few months after Martin Luther King delivered a speech trying to unite the civil rights and antiwar movements, Yarborough told his subordinates at the US Army Intelligence Command: “Men, get out your counterinsurgency manuals. We have an insurgency on our hands.”10

  William Godel had set up ARPA’s Project Agile to fight insurgencies abroad. General Yarborough focused on an extension of that same mission: fighting what he saw as a foreign insurgency on American soil. Just as in Vietnam, his first order of business was to take out the insurgents’ base of local support. But before he could start clearing the weeds, his men needed information. Who were these insurgents? What motivated them? Who called the shots? Who were their domestic allies? Among what groups did they hide?