COINTELPRO: When the Paranoia Was Filed in Triplicate

On the night of 8 March 1971, while most of America was watching the heavyweight title fight between Muhammad Ali and Joe Frazier, eight members of a group calling itself the Citizens’ Commission to Investigate the FBI broke into a small Bureau field office in Media, Pennsylvania, and carried out every file they could fit into suitcases. They had chosen fight night on purpose, gambling that the guard would be distracted. Among the documents they took were memoranda stamped with a word that meant nothing to the public and everything inside the Bureau: COINTELPRO. Over the following weeks the burglars mailed copies to newspapers and members of Congress, and the activists who had spent years insisting the government was spying on them, and being told they were paranoid, watched their paranoia turn into a matter of public record.

The programme, as documented

COINTELPRO — a Bureau contraction of “Counter Intelligence Program” — was not a rumour, an inference, or a leak of ambiguous fragments. It was a formal, named, internally documented set of operations run by the Federal Bureau of Investigation from 1956 to 1971, authorised at the top and recorded in the Bureau’s own memoranda, which is how we know its aims in the FBI’s own chilling vocabulary: to “expose, disrupt, misdirect, discredit, or otherwise neutralise” the activities of domestic political groups the Bureau’s director had decided were dangerous.

That director was J. Edgar Hoover, who by the 1950s had run the Bureau for three decades and would run it until his death in 1972, accountable in practice to almost no one. Under his hand COINTELPRO began by targeting the Communist Party USA and expanded relentlessly outward until it covered a startling swathe of American dissent: the Black civil rights movement, including the Southern Christian Leadership Conference and Martin Luther King himself; the Black Panther Party; the American Indian Movement; the anti-war and New Left student organisations; and the women’s liberation movement, which the Bureau infiltrated and catalogued as though a consciousness-raising meeting were a cell of saboteurs.

The methods went far beyond watching. The Bureau planted informants and agents provocateurs to push groups toward crimes that could then be prosecuted, or simply to sow the suspicion that tore organisations apart from inside. It fed fabricated stories to friendly journalists to discredit activists, sent forged letters to provoke violence between rival groups, and worked to get people sacked from jobs and evicted from homes. In the case of the Black Panthers, documents later showed the Bureau deliberately stoking a lethal feud between the Panthers and a rival organisation; the FBI’s campaign against the Panthers is part of the context of the 1969 police killing of the Chicago organiser Fred Hampton, who had been under Bureau surveillance through an informant.

The single most notorious artefact of the programme was the letter the Bureau sent Martin Luther King in 1964. Accompanied by a tape of secretly recorded audio, an anonymous note — later established to have originated inside the FBI — told King he was a fraud and an “evil, abnormal beast”, and closed with an unmistakable invitation to suicide: “There is only one thing left for you to do. You know what it is… There is but one way out for you.” The most celebrated moral leader in the country received, from his own government’s police, a letter urging him to kill himself. That is not a summary of a conspiracy theory. It is a summary of the record.

How it came apart

A programme built on absolute secrecy was undone by two acts of exposure, one illegal and one official, and the sequence matters.

The Media burglary of March 1971 was the crack. The stolen files, dribbled out to the press over the following months, put the word COINTELPRO into public circulation for the first time and proved that the Bureau was running organised operations against lawful political groups. Under the pressure the disclosures created, and facing a Freedom of Information Act lawsuit from the journalist Carl Stern, the FBI formally terminated the programme in April 1971 — though the practices it named did not all stop.

The official reckoning came four years later. In 1975 the Senate established a Select Committee to Study Governmental Operations with Respect to Intelligence Activities, chaired by Senator Frank Church of Idaho and universally remembered as the Church Committee. Working through 1975 and 1976, it conducted the most thorough public examination of the American intelligence agencies ever undertaken, and its findings on COINTELPRO were unequivocal. The committee concluded that the Bureau had conducted a “sophisticated vigilante operation aimed squarely at preventing the exercise of First Amendment rights” — that the government had, for fifteen years, secretly and systematically worked to sabotage citizens for their political beliefs. The Church Committee’s report remains the primary public record of what was done, and its reforms — the Foreign Intelligence Surveillance Act, the permanent congressional intelligence oversight committees — were built to ensure the programme could not simply resume in silence.

The scale the committee documented is easy to understate. COINTELPRO was not one operation but a family of them, with separate programmes aimed at the Communist Party, the Socialist Workers Party, “white hate groups” such as the Ku Klux Klan, “black nationalist hate groups” — a Bureau category that swept in mainstream civil rights organisations alongside militant ones — and the New Left. The Bureau opened files on a vast number of Americans whose only offence was political activity the director disliked; it tracked the movements of a generation of organisers, catalogued their affairs and their debts, and looked for the private detail that could end a career or a marriage. The animating theory, visible throughout the internal memoranda, was that dissent itself was a species of subversion, and that a citizen who organised against the government’s policies had forfeited the presumption of innocence. That premise, once you accept it, licenses everything the programme did. The whole architecture of the abuse follows from a single decision to treat political opposition as an enemy to be neutralised.

The fork, and why it is subtle here

On this desk the usual task is to find the seam where a documented scandal forks into a mythology the evidence cannot support. With COINTELPRO the seam is unusually hard to see, and that difficulty is the whole point of the piece.

The real programme licenses an enormous and largely legitimate suspicion of state surveillance. That suspicion is well founded. The fork is not a lie sitting next to a truth; it is a matter of scope and calibration. COINTELPRO proves that the government secretly surveilled and sabotaged specific political movements it had decided to treat as threats. The generalised belief it feeds is subtly larger: that the government is watching everyone, that any given individual’s private life is under active, personalised observation, that behind every institution sits a coordinated apparatus tracking each citizen. The first claim is documented history. The second is a claim about the ubiquity and competence of surveillance that the record does not, on its own, establish.

The gap is easy to walk across without noticing, because the true claim already lives so far into territory that sounds paranoid. A person who says “the FBI ran a secret programme to destroy the reputation of civil rights leaders and drive one to suicide” sounds, to anyone who does not know the history, like a conspiracy theorist — and is describing a Senate-documented fact. Once you have crossed a line that far out and been proved right, the next line, and the one after, offer very little resistance. If they did that, why not everything? The believer who slides from “the state surveilled dissident movements” to “the state is surveilling me, personally, right now” is not making a wild leap. They are extending a proven pattern one reasonable-seeming step past where the evidence stops, and the step is nearly invisible because everything up to it was true.

This is the recurring architecture of earned mistrust, and COINTELPRO is one of its purest specimens. The same shape appears in the way the real horror of MKUltra makes the imagined mind-control of every lone gunman feel plausible, and in the way the forty documented years of the Tuskegee study legitimise a mistrust of medicine that then generalises well past its origin. A proven wrong sets a template, and every later fear that matches the template’s silhouette gets waved through, because the guard learned — correctly — to recognise that silhouette as danger.

Why being right once is so corrosive

There is a particular psychological weight to having been vindicated, and it is worth naming with care, because it explains why COINTELPRO’s legacy is so much stickier than a simple exaggeration would be.

Imagine you spent the 1960s telling people your meetings were infiltrated, your mail was opened, your organisation was being pulled apart by forces you could feel but not prove. Imagine being called paranoid by friends, by journalists, by the respectable centre — and then, in 1971, watching stolen government files confirm that you had been right the entire time, that the enemy in the room was real and had a filing cabinet. What that experience does is not merely inform you; it recalibrates your entire instrument for detecting threat. Vindication of that magnitude teaches you that your suspicious reading of the world was the accurate one and the reassuring reading was the naïve one. From that day forward, “you’re being paranoid” is not a warning you can hear; it is the exact phrase that was used to gaslight you before the files came out.

This is the corrosive gift the government hands out whenever it is caught doing the unthinkable to people who warned that it was being done. It does not only harm the immediate victims. It permanently damages the culture’s ability to distinguish warranted suspicion from unwarranted, because the institution that should help draw that line is the one that erased it. When the watchdog is exposed as the wolf, every subsequent bark sounds equally credible, and the person who can no longer tell a real threat from an imagined one was made that way by a betrayal they did not invent.

The honest close asks something harder than either trusting the intelligence services or endorsing the belief that every camera is trained on you. It asks us to recognise that the people most inclined toward that belief are frequently the descendants, literal or political, of those who were genuinely surveilled — and that their vigilance is the scar tissue of a real wound. The activists of the 1960s who said they were being watched were right, and the Bureau had the paperwork to prove it. That fact does not make every later fear correct. But it should make anyone reaching for the word “paranoid” pause, and remember that the last time it was deployed with this much confidence, it was the state’s word, and the state was lying.