Operation Northwoods: The Plan to Manufacture a War

In March of 1962, the Chairman of the Joint Chiefs of Staff put his name on a document proposing that the United States military stage terrorist attacks on American soil, hijack civilian airliners, sink boats full of refugees, and kill American citizens, all so the government could blame Cuba and manufacture public support for an invasion.

The document, addressed to Secretary of Defense Robert McNamara and bearing the code word "Northwoods," proposed pretexts the United States could manufacture to justify invading Cuba, including the shootdown of a civilian airliner or the bombing of a U.S. ship at Guantanamo Bay.[1] This was not a theory. This was not a leak from an anonymous source with an axe to grind. The plans reportedly included the possible assassination of Cuban emigres, sinking boats of Cuban refugees on the high seas, hijacking planes, blowing up a U.S. ship, and orchestrating violent terrorism in U.S. cities, all developed as ways to trick the American public and the international community into supporting a war against Fidel Castro's Cuba.[2]

This is a signed government document, sitting in the National Archives under its own catalog number, that most Americans have never heard of. It stayed buried for thirty five years. And when it finally surfaced, it barely made a ripple in the mainstream press.

I am not writing this piece to tell you that the government did this to the American people. It did not. President Kennedy said no. But I am writing this piece because the fact that senior military leadership drafted it, approved it, and formally submitted it for consideration tells you something true and permanent about how the machinery of war gets built, regardless of which party happens to be sitting in the Oval Office when it happens.

To understand how a room full of the country's most senior military officers arrived at a plan involving American casualties, you have to understand what came before it in the room.

In April of 1961, the CIA backed a paramilitary invasion of Cuba by roughly 1,400 Cuban exiles at the Bay of Pigs. It was a catastrophic failure. Kennedy took the public blame, but privately, the failure humiliated him in front of his own military leadership, and it convinced the Joint Chiefs of Staff that the civilian side of the government, meaning Kennedy himself, was too cautious to finish what had been started against Castro's government.

What followed was Operation Mongoose, an extensive campaign of covert operations against Cuba carried out by the CIA, officially authorized on November 30, 1961 by President Kennedy.[3] Mongoose was run through a body called the Special Group (Augmented), which included the President's Special Assistant for National Security Affairs McGeorge Bundy, the Deputy Under Secretary of State, the Deputy Secretary of Defense, the Director of Central Intelligence, and the Chairman of the Joint Chiefs of Staff, General Lyman Lemnitzer.[4] This was not a rogue cell operating outside the chain of command. This was the highest tier of American national security leadership, sitting in the same room, reviewing and approving covert action against a foreign government on a rolling basis.

Declassified documents from the Digital National Security Archive describe the intense focus of the Kennedy administration on Cuba, along with the disbursement of millions of CIA dollars on raids conducted by Cuban exiles.[5] By the time Lemnitzer's Joint Chiefs sat down to draft Northwoods in early 1962, the appetite for direct, decisive action against Cuba was not fringe thinking. It was the operating consensus of the men running point on Cuba policy. The only friction was that Kennedy kept refusing to authorize the kind of overt military commitment his generals wanted.

That friction is the room Northwoods came out of. It was not an isolated document written by a lone extremist in a back office. It was the next logical step, in the minds of the men proposing it, from an already-running covert war that had failed to produce results fast enough for the people running it. When the option of provoking Cuba into giving a pretext ran out of patience, the alternative that landed on Secretary McNamara's desk was for the United States to manufacture its own.

This is the part I want to sit with for a second, because it is easy to read about a 1962 memo and file it away as a historical curiosity. It should not be filed away. The men in that room were not cartoonish villains. They were career military officers and senior policymakers operating inside the normal bureaucratic structure of the United States government, and they concluded, as a serious operational option, that American civilian deaths were an acceptable price to manufacture support for a war their elected commander in chief had not agreed to fight. That is not a conspiracy theory. That is what the document says. And it is worth remembering that this willingness, to treat the American public as an input to be managed rather than a population to be protected, does not require a specific party or a specific decade. It requires a bureaucracy convinced that its own judgment supersedes the people it answers to.

The document itself is fifteen pages, dated March 13, 1962, addressed directly to Secretary of Defense Robert McNamara, and signed by the Chairman of the Joint Chiefs of Staff on behalf of the entire Joint Chiefs. It was not a brainstorm memo from a mid-level staffer. It carried the institutional weight of the country's top military leadership.

The proposals, developed as part of the broader Operation Mongoose program, included staging the assassinations of Cuban exiles living in the United States, fabricating a Communist Cuban terror campaign to be carried out in Miami, other Florida cities, and Washington D.C., sinking a boatload of Cuban refugees on the open water, faking a Cuban air force attack on a civilian airliner, and manufacturing a "Remember the Maine" style incident by blowing up an American ship and blaming Cuban sabotage.[6] The recommendations section of the memo made clear the Joint Chiefs' willingness to stage actual terrorist events with real casualties in order to justify military intervention, with the stated cumulative goal of provoking a Cuban response that would give the United States its pretext.[7]

This was not one isolated scenario. It was a menu. Multiple options, each one built around the same premise: if Cuba would not attack the United States, the United States would manufacture the appearance that it had, and pay for that appearance in American blood if necessary.

Bureaucratically, the document was careful. The Joint Chiefs recommended the memo be forwarded only as a preliminary submission suitable for planning purposes, assumed that other agencies would submit similar proposals, and recommended that responsibility for both the overt and covert military aspects of the operation be assigned to the Joint Chiefs of Staff itself.[6] The document carried explicit instructions that it not be forwarded to unified or specified combatant commanders, not be forwarded to U.S. officers assigned to NATO, and not be forwarded to the American delegation at the United Nations Military Staff Committee.[8] In other words, the men proposing this understood exactly how radioactive it was, and structured its distribution to keep it contained to the smallest possible circle of the national security apparatus.

The document was declassified in 1997 through the Kennedy Assassination Records Review Board process, and entered wider public awareness in 2001 after author James Bamford obtained and publicized a copy, with the National Security Archive posting it online alongside contemporaneous news coverage.[1][9] Every signature on that document belongs to a man who, at the time, had access to the machinery of the United States military and used it to draft a plan involving the deaths of American civilians as an acceptable cost of doing business.

From McNamara's desk, the proposal went to the Oval Office. That is where it stopped.

The popular version of this story ends with a dramatic presidential veto. The documented version is quieter, and more useful for understanding how institutional accountability actually works, or doesn't.

Kennedy summoned Lemnitzer to the Oval Office on March 16, 1962, three days after the memo reached McNamara's desk, to discuss what the Joint Chiefs had proposed.[10] Lemnitzer did not walk in and lay out the plan's most graphic scenarios. According to Lansdale's notes on the meeting, Lemnitzer instead told Kennedy that the Joint Chiefs had plans for creating plausible pretexts to use force against Cuba, framed around either attacks on U.S. aircraft or a Cuban action elsewhere in Latin America that could be used to justify retaliation.[10] Kennedy responded bluntly that they were not discussing the use of military force, and pointedly noted that Lemnitzer might find American forces tied up in Berlin or elsewhere rather than free to use the divisions he wanted for Cuba.[11]

That is the extent of the documented confrontation. There is no signed rejection letter with the same dramatic weight as the approval memo. What exists is a meeting, a blunt verbal no, and a paper trail that goes quiet on further pretext development after that point. Circumstantial evidence is strong that the proposal was simply not acted on, though it remains unclear whether Kennedy was ever briefed on the full and more graphic details contained in the plan's annexes.[12] That distinction matters. It is possible, based on the documented record, that Kennedy rejected a sanitized verbal summary rather than the complete written scenarios involving Miami and Washington terror campaigns. The historical record supports "Kennedy said no," full stop. It does not clearly support "Kennedy read the worst parts of this document and personally killed it," which is the more dramatic version that tends to circulate.

The aftermath is similarly murkier than the popular telling. Within a few months of the refusal to endorse Northwoods, Lemnitzer was denied another term as Chairman of the Joint Chiefs, and in November 1962 was appointed commander of U.S. European Command and NATO's Supreme Allied Commander Europe, a move some accounts characterize as a positional demotion from the chairmanship.[13] That characterization is disputed. At least one competing account frames the reassignment as part of a broader military leadership overhaul following the Bay of Pigs failure and ahead of the Cuban Missile Crisis, rather than direct fallout from Northwoods specifically.[14]

Complicating the timeline further, at least one biographical account places European interest in Lemnitzer for the SACEUR post as early as July 1962, with the formal appointment reportedly delayed only by the onset of the Cuban Missile Crisis that October, a sequence that sits uneasily with a straightforwardly punitive reading of the November reassignment, and that neither confirms nor forecloses a causal link to Northwoods.[15]

I want to be straight with you about what that means for the argument. It is fair, and well documented, to say Kennedy rejected a plan to manufacture American casualties for a war he did not want to fight, and that the man who signed that plan was out of his job within the year. It is not fair to assert, as settled fact, that the reassignment was punishment specifically for Northwoods, since the timeline overlaps with other, larger institutional pressures on Kennedy's relationship with the Joint Chiefs. What is not in dispute is the outcome: nobody was investigated, nobody was charged, and Lemnitzer finished his career as one of the most powerful military officers in the Western alliance. Whatever this was, it was not accountability in any meaningful sense.

This tension between the Kennedy White House and the Joint Chiefs, over the use of covert pretexts and unchecked military escalation, is not an internet-era invention. It is documented in mainstream, peer-reviewed presidential history, including Robert Dallek's biography of Kennedy and H.R. McMaster's study of civil-military relations inside the same Pentagon leadership structure a few years later, during the run-up to Vietnam.[16][17] The pattern here was not a single rogue general. It was a documented rift over how much latitude military leadership believed it had to manufacture the conditions for war, a rift that outlasted the Kennedy administration entirely.

Nobody involved in drafting Operation Northwoods was ever investigated, charged, or publicly disciplined for it. Lemnitzer's chairmanship ended on its normal cycle and he moved on to one of the most prestigious commands in the Western alliance. The paper trail on further pretext development for Cuba goes quiet after McNamara's rejection, and the circumstantial evidence is strong that the proposal was simply never acted on rather than formally repudiated through any disciplinary process.[12] There was no hearing. There was no public accounting. The document that proposed killing American civilians to manufacture a war was filed away, classified, and left there.

It stayed classified for thirty five years, and only became public because James Bamford obtained a copy through the declassification process created by the JFK Assassination Records Review Board in the 1990s, a legal mechanism built to investigate Kennedy's assassination, not to audit Cold War military planning.[18] Northwoods did not surface because some oversight system caught it. It surfaced as a side effect of an entirely unrelated declassification effort, decades after everyone who might have been held accountable for it had already retired, in several cases already died. If John F. Kennedy had not been assassinated the following year, and if that assassination had not eventually forced open millions of pages of unrelated Cold War records, there is a real argument that this document might still be sitting in a classified file today.

That is not an accident of one document. It is close to how the system worked generally. The Church Committee's mid-1970s investigation into U.S. intelligence activities found that only around fourteen percent of covert action projects were ever individually reviewed by the oversight structure responsible for approving them, with the remaining majority treated as low-risk and low-cost operations that were effectively rubber-stamped without case-by-case scrutiny.[19] Northwoods was drafted and submitted through exactly this kind of structure, the Special Group (Augmented) covering Operation Mongoose, before the Church Committee ever existed to examine how thin that oversight really was. The Committee's findings came a decade after Northwoods was written, applying to an entirely different set of programs, but the structural lesson lands on Northwoods anyway: a document proposing the deaths of American civilians as an acceptable pretext for war was not an aberration that slipped past a vigilant system. It moved through a system that, according to the government's own later investigation of itself, was not built to catch things like it in the first place.

Northwoods is the plan that got rejected. Two years later, the government used a nearly identical playbook, and this one worked.

On August 4, 1964, President Johnson told the country that North Vietnamese forces had carried out a second attack on U.S. Navy ships in the Gulf of Tonkin. Congress responded by passing the Gulf of Tonkin Resolution on August 7, unanimously in the House and 88 to 2 in the Senate, handing Johnson the legal authority to use all necessary measures in Southeast Asia.[20] That resolution became the legal foundation for the escalation of the Vietnam War.

More than forty years later, the release of nearly two hundred documents related to the Gulf of Tonkin incident, along with tapes from the Johnson Presidential Library, revealed what historians had long suspected but could never prove: there was no second attack on U.S. Navy ships in the Tonkin Gulf that August.[20] Among the most significant of those documents was an internal NSA historical study by historian Robert J. Hanyok, which concluded that the signals intelligence used to confirm the second incident had been misrepresented as it moved up the chain of command, implicating Secretary of Defense McNamara in distorting the picture presented to Congress.[20][21] That conclusion is not universally accepted within the specialized literature; a response circulated within the NSA itself disputed Hanyok's inference of deliberate distortion, arguing the agency had performed capably during the crisis and that the available signals intelligence, standing alone, could not conclusively prove an attack had not occurred.[22] What is not in dispute, across every account, is that the August 4 engagement did not happen as it was officially described to Congress and the public at the time. Confidential testimony later given to the Senate Foreign Relations Committee, and declassified decades afterward, showed that naval personnel directly involved cast doubt on the official version of events, and an internal investigation ultimately accused McNamara of misleading the committee in his testimony.[23]

I want to be precise about what the comparison is and is not. Northwoods proposed manufacturing a pretext from nothing, up to and including staged attacks with real casualties on American soil. Tonkin was not that. Tonkin was a real, ambiguous naval engagement on August 2, followed by a second incident on August 4 that the government's own later internal history concluded had not actually happened as described, and that the Secretary of Defense presented to Congress with more certainty than the underlying intelligence supported. The severity is different. The mechanism is not. In both cases, senior Pentagon leadership was willing to shape or fabricate the factual basis for war rather than let Congress and the public make that decision on accurate information. Northwoods shows you what the Joint Chiefs were willing to propose when given the chance. Tonkin shows you what happened when a similar impulse, distort the pretext to get the war leadership already wanted, actually made it into execution.

This is also the point in the piece where I want to be honest about something: both of these episodes happened under Democratic administrations, Kennedy and Johnson. That is a fact of the timeline, not a partisan conclusion. The pattern this reveals, senior military and civilian defense leadership willing to manufacture or distort the pretext for war, is not a Democratic pathology or a Republican one. It shows up again four decades later in the run-up to the 2003 Iraq War, under a Republican administration, when intelligence about weapons of mass destruction was presented to the public with far more certainty than the underlying evidence supported. I am not going to relitigate Iraq in this piece, that is its own investigation for another day, but it belongs in the same sentence as Northwoods and Tonkin for one reason: the willingness to shape the factual basis for war is a feature of the national security apparatus itself, not of whichever party happens to hold the White House when it happens.

I did not write this piece to relitigate 1962. I wrote it because a room full of the most powerful military officers in the country once looked at the question of how to build public support for a war and concluded, as a serious operational proposal, that killing American civilians was an acceptable answer. President Kennedy said no. Two years later, under a different president, a different version of the same instinct, shading the truth about what actually happened in a foreign gulf, got a resolution through Congress with a single House vote against it. The instinct did not require a signed order to stage an atrocity. It only required officials willing to let Congress and the public believe something happened, or something happened for reasons other than what they were told.

OpinionI think the United States has been, for most of my lifetime and long before it, a country whose government is willing to manufacture or distort the conditions for war when it wants one badly enough and the public isn't there yet. Vietnam is the one that should worry every American the most, not because it is the most extreme example on paper, Northwoods is more extreme on paper, but because it is the one where the manufactured pretext actually became the legal basis for a war that killed over fifty eight thousand American service members and an estimated three million Vietnamese people. Northwoods tells you what they were willing to propose. Tonkin tells you what they were willing to do when the opportunity presented itself.

OpinionThat is why the current war with Iran worries me more than most people seem to be talking about. By every public poll I can find, this may be the most unpopular war in modern American history, arguably rivaling or exceeding the opposition Vietnam eventually generated, except this opposition showed up from the very beginning rather than building slowly over years.[24][25][26] I do not know what that means. I do not have evidence that anything resembling Northwoods or Tonkin is happening right now, and I want to be completely clear that I am not alleging that it is. What I am saying is that history has already shown us, twice, in the same decade, that when an administration wants a war badly enough and the public does not, the instinct to manufacture consent rather than earn it is not hypothetical. It has already happened. And I don't think it's unreasonable to ask, in real time, whether the same pressure that produced Northwoods and Tonkin exists today, and what it would look like if it were acted on again.