Chapitre 1
The Nuclear Precipice: A Whistleblower's Warning About Doomsday Machines
Picture this: You're in a White House office in 1961, holding a classified document that reveals how our world would end. The paper shows that a U.S. nuclear first strike would kill 275 million people immediately, rising to 325 million after six months. Add another 300 million deaths from fallout across Europe and neutral countries. Six hundred million deaths total-a hundred Holocausts in one strike. This wasn't fiction or a hypothetical scenario. This was the actual U.S. war plan, and Daniel Ellsberg, later famous for leaking the Pentagon Papers, was staring at it.
"The Doomsday Machine" has become essential reading for anyone concerned with existential threats to humanity. Former Secretary of Defense William Perry called it "a powerful and important book," while Noam Chomsky described it as "an absolutely imperative read in this hair-trigger nuclear age." The book draws on Ellsberg's unique insider experience as a nuclear war planner and his five decades of activism against nuclear weapons, offering a chilling glimpse into how close humanity has repeatedly come to annihilation.
Chapitre 2
Inside the Nuclear Planning Machine
When I joined RAND Corporation in 1958 as a thirty-year-old strategic analyst, I entered a world of apocalyptic calculations. My colleagues and I were engaged in what seemed the most important work imaginable-preventing nuclear war by ensuring Soviet leaders would face unacceptable consequences for attacking America. The irony wasn't lost on me: I was working to guarantee genocide-level retaliation capabilities despite my personal opposition to city bombing during World War II.
The logic of nuclear deterrence was brutally simple. According to classified intelligence, we faced a ruthless enemy seeking global dominance through potential nuclear first strikes. Albert Wohlstetter, RAND's brilliant strategist, emphasized that deterrence wasn't measured by pre-war forces but by what Soviets could foresee as our "second-strike capability" after their attack. The implication was chilling-adequate deterrence required a survivable capability to kill more Soviet citizens than the twenty million who died in World War II.
While most RAND colleagues studied weapons vulnerability, I specialized in the understudied but critical area of nuclear command and control-how senior military officers and especially the president would make and implement decisions about nuclear retaliation. This represented the ultimate application of decision theory under uncertainty-potentially the last decision a national leader would ever make.
What particularly concerned me was the problem of "ambiguity" in strategic warning. Technical systems designed to provide warning were inherently unreliable. In 1960, the new Ballistic Missile Early Warning System radar in Greenland generated a false alarm during a VIP visit to NORAD headquarters. As executives including Tom Watson of IBM watched, the threat level climbed to 5-indicating 99.9% certainty of attack. The visitors were evacuated in panic, believing nuclear war had begun. The cause? Radar signals bouncing off the rising moon.
My first RAND memo questioned whether pilots would actually follow fail-safe protocols in crisis situations. Drawing on my Marine Corps experience, I argued that officers' behavior would reflect not just standing orders but their sense of mission and assessment of the current situation. The absence of an "Execute" order after launch during apparent attack would itself be ambiguous-it might mean return was desired, or that communications had been destroyed by enemy action.
Chapitre 3
The Illusion of Presidential Control
Throughout the Pacific Command, I discovered the two-man rule for nuclear control was systematically circumvented. Though elaborate procedures existed requiring two officers to authenticate nuclear orders using split codes and separate safes, duty officers routinely arranged ways to bypass these safeguards. Personnel shortages meant posts were frequently manned by only one officer, who had secretly arranged access to all codes and combinations "if necessary." Every command post I visited had undermined these protections.
Even the Minuteman missile system's more sophisticated controls were compromised. John Rubel discovered operators had set timing intervals to zero, allowing single control centers to launch without required confirmation. Later, Bruce Blair revealed to a shocked Robert McNamara that the electronic locks McNamara had mandated for Minuteman missiles were permanently set to "00000000" by Strategic Air Command.
Most disturbing was the deliberate omission of any authentication procedure to recall aircraft once launched. Unlike Go codes, no Stop or Return codes existed in authentication envelopes. This wasn't accidental-military commanders feared civilian leaders might have second thoughts after ordering a nuclear strike. As one officer explained, the concern was that "a civilian president might have second thoughts about the attack under way," potentially leaving forces "disorganized and vulnerable" to enemy attack. The system was designed to make presidential recall of nuclear bombers impossible once launched.
In 1959, I learned President Eisenhower had secretly delegated nuclear launch authority to Admiral Harry Felt, CINCPAC commander, for use whenever communications between Washington and Hawaii were disrupted-which happened daily due to atmospheric interference. This delegation had led to sub-delegation far below the theater commander level, creating numerous points of potential unauthorized nuclear initiation.
The military's determination to maintain operational freedom extended to violating international agreements. At the Marine air base at Iwakuni, Japan, I discovered a secret arrangement violating our treaty with Japan: an LST ship called USS San Joaquin County, disguised as an electronics repair vessel, was permanently anchored just offshore with nuclear weapons aboard. This ship, stationed within Japanese territorial waters mere hundreds of yards from shore, allowed Iwakuni's planes to access nuclear weapons within minutes rather than hours-a clear treaty violation that risked catastrophic diplomatic consequences if discovered.
Chapitre 4
The Single Integrated Operational Plan: Blueprint for Omnicide
From the beginning of my study in Hawaii, I sought access to the actual war plans an Execute order would activate. I was granted unprecedented civilian access to the Top Secret "cage" at CINCPAC headquarters-a wire-netted, guarded room containing all current nuclear plans. Studying these documents revealed a startling omission: in every plan for war with the Soviet Union, Chinese targets (including all major cities) were also struck. This reflected CINCPAC's strong institutional bias-since few Soviet targets were within their range, Pacific forces would be "sidelined" in a Russia-only conflict.
When I raised the possibility of war against only the USSR with Vice Admirals Kivette and Ekstrom, they reacted with visceral shock. "You have to assume some modicum of rationality in higher authority that they would not do something so insane as to go to war against one Communist power while letting the other one off scot-free," Admiral Ekstrom gasped.
The technical challenges of nuclear war planning were overwhelming. CINCPAC planners worked around the clock just to produce one war plan against the entire Sino-Soviet bloc. They faced the critical problem of "interference" where nuclear explosions could destroy nearby aircraft. Their solution involved intricate timing calculations, but the plans were fundamentally flawed: Execute orders arrived hours apart rather than simultaneously; wind conditions (completely ignored in planning) would dramatically affect aircraft arrival times; and many bases never received test orders at all.
The complexity meant planners couldn't create multiple plans-it took them all year to produce a single yearly-updated plan. Operations centers throughout the Pacific displayed large secret maps showing nuclear targets, but these maps didn't clearly demarcate between China and Russia. The Sino-Soviet bloc appeared as one giant landmass with arrows and pins indicating targets. Computer programs listed coordinates but not countries, making it impossible to quickly distinguish Russian from Chinese targets.
By 1960, all nuclear war planning was consolidated into the Single Integrated Operational Plan (SIOP) at SAC headquarters in Omaha. This plan welded all U.S. warheads into one coordinated attack, with targets including every city in the Soviet Union and China with populations over 25,000. The arsenal included about 1,700 SAC bombers carrying thermonuclear bombs up to 25 megatons-each with more firepower than all explosives detonated in human history.
The SIOP briefing I witnessed revealed the plan's horrifying scale. A briefer demonstrated successive attack waves from B-52s, fighter-bombers, B-47s and missiles, showing Moscow being obliterated beneath plastic overlays. The plan called for forty megatons on Moscow alone-four thousand times more than Hiroshima and far exceeding all non-nuclear bombs dropped in WWII.
Most shocking were the casualty projections: the fallout alone (not counting blast or direct radiation) would kill 100 million Soviets and 300 million Chinese-half the population of each country. When someone questioned whether China should be targeted if they weren't involved in the war, General Power dismissively replied that changing the plan would "really screw it up."
Chapitre 5
Drafting a New Nuclear Strategy
In spring 1961, Harry Rowen tasked me with drafting the general war section of the new Basic National Security Policy for the Department of Defense. This crucial document would guide all Pentagon war planning, replacing Eisenhower's "New Look" and "massive retaliation" doctrines with Kennedy's promised "flexible response" strategy.
The Air Force concept of "war fighting" or "damage limiting" involved prolonged, controlled attacks on Soviet military targets including hardened missile sites and command centers. This strategy conveniently supported Air Force priorities like the B-70 bomber program (later the B-1) and called for thousands of land-based ICBMs with improved accuracy.
While I doubted counterforce's effectiveness, I still questioned what we should do if deterrence failed. I accepted that land-based missiles should target military sites, but I believed more promising was a "coercive," "no cities" strategy aimed at terminating war quickly before all weapons were employed against urban targets.
This approach required three key elements: avoiding enemy cities in initial strikes (announcing this intention beforehand), maintaining protected reserve forces to preserve threat capability, and preserving command and control systems on both sides capable of terminating operations.
On my thirtieth birthday, April 7, 1961, I completed my first draft of the general war section of the Basic National Security Policy. My twelve-page proposal challenged virtually every aspect of existing nuclear war plans, proposing strategic reserves, reliable command systems, physical safeguards against unauthorized actions, flexible contingency planning, and options to avoid attacking cities.
The final version, redrafted for Deputy Secretary Gilpatric's signature, was sent to the JCS chairman on May 5 as "Policy Guidance on Plans for Central War." My revised guidance became the foundation for operational war plans under Kennedy-reviewed by me for Gilpatric in 1962, 1963, and again under Johnson in 1964. It has reportedly influenced U.S. strategic war planning ever since.
Chapitre 6
The Cuban Missile Crisis: Closer to Doomsday Than We Knew
On October 22, 1962, I watched President Kennedy announce on television that the Soviets were installing "offensive" ballistic missiles in Cuba. His statement that "a single missile" launch from Cuba would trigger "a full retaliatory response upon the Soviet Union" caught my attention-having written guidance for the SIOP nuclear war plan, I recognized this meant general nuclear war over a single missile.
Upon arrival in Washington, Harry asked me to analyze what thirty-eight Soviet missiles in Cuba could do to America's strike-back capability. Both Washington and Omaha (SAC headquarters) were within range of the medium-range ballistic missiles (MRBMs), some already operational. This meant Soviet forces could hit our command posts with virtually no warning-essentially enabling decapitation strikes.
However, I knew what most didn't: thanks to our delegation systems, this wouldn't prevent full, quick retaliation from our massive surviving forces. We had already designed alternate command posts at sea, airborne, and underground for this very scenario. SAC bombers had already dispersed to over thirty civilian airfields.
While Harry Rowen and I had confidently assessed the risk of nuclear war during the crisis as extremely low-perhaps one in a thousand-I was stunned to learn afterward that Paul Nitze had estimated it at one in ten, and believed his was the lowest estimate among ExComm members.
What Kennedy never knew was that 42,000 Soviet troops (not 7,000-17,000 as estimated) were in Cuba with over 100 tactical nuclear weapons. Even more alarming, the Presidium had previously authorized local commanders to use these weapons against an invasion force without Moscow's direct orders. Though this authorization had been withdrawn after Kennedy's October 22 speech, Khrushchev couldn't be confident the new orders would be followed in combat conditions.
While Kennedy and Khrushchev were still negotiating on October 27, an even more dangerous situation was developing underwater. Soviet submarine captains, suffering in 140-degree heat with failing ventilation systems and rising carbon dioxide levels, were being harassed by American depth charges they interpreted as attacks.
On submarine B-59, conditions were dire. As crew members fainted from heat and carbon dioxide poisoning, Captain Savitsky, furious and exhausted after hours of depth charge attacks, ordered the nuclear torpedo prepared: "Maybe the war has already started up there while we are doing somersaults here! We're going to blast them now! We will die, but we will sink them all-we will not disgrace our Navy!"
Unlike other submarines where only the captain and political officer needed to agree to launch, B-59 required a third officer's consent because brigade chief of staff Vasili Arkhipov was aboard. Arkhipov refused to authorize the launch, likely saving the USS Randolph carrier group from nuclear destruction. Had this occurred, American leaders would have assumed it came from an undetected Cuban missile, triggering Kennedy's promised full nuclear response against the Soviet Union.
The submarine crisis wasn't the only near-catastrophe that Saturday. A SAC U-2 spy plane accidentally strayed into Soviet airspace over the Chukot Peninsula when its pilot became disoriented by the northern lights. As Soviet MiGs scrambled to intercept it, the Alaskan Air Command launched F-102A fighters armed only with nuclear air-to-air missiles to protect the U-2.
The true danger of nuclear weapons isn't just proliferation to "rogue" states, but fundamentally their very existence in the arsenals of superpowers, even when led by responsible leaders. Kennedy and Khrushchev-both relatively cautious and humane-came horrifyingly close to launching forces neither contemplated using at the crisis's start. Had their bargaining continued one more day, nearly all humans then living might have perished.
Chapitre 7
The Path to City Bombing: Crossing the Moral Threshold
Where did the road to doomsday begin? The fission-bomb mechanism was first envisioned by physicists, tested at Alamogordo, and revealed to the world at Hiroshima and Nagasaki. But before that came a crucial transformation in military thinking: when did setting fire to cities and burning civilians from the sky become not only acceptable but necessary? This change preceded the nuclear era but relied on two developments converging in World War II: the belief that airpower was key to victory, and the increasing willingness to regard civilian populations as legitimate military targets.
On September 1, 1939, as Hitler invaded Poland, President Roosevelt appealed to all belligerents to refrain from "the ruthless bombing from the air of civilians in unfortified centers of population" which had "sickened the hearts of civilized men and women and profoundly shocked the conscience of humanity." Britain and Germany both agreed to this standard.
The 1939 agreement didn't hold. Hitler's bombing of London in 1940 (the Blitz) was an obvious breach. But a year later, British civilian and military leaders officially and deliberately, though secretly from their public, adopted and expanded Hitler's tactics as their primary basis for attacks on Germany from early 1942. The United States eventually joined in, primarily for operational reasons rather than effectiveness.
By early 1945, particularly in Japan under General Curtis LeMay, targeting cities to cause maximum civilian casualties-what FDR had termed "human barbarism"-became the primary form of attack. This policy was kept secret throughout the war because Churchill, Roosevelt, and Truman publicly claimed they were observing the principles of civilian immunity. That was a lie.
The RAF discovered that incendiary bombs were far more effective than high explosives for destroying cities. They deliberately targeted densely packed workers' housing where fire would spread rapidly, and developed tactics using both high explosives and incendiaries-the former to block streets and disrupt firefighting efforts, the latter to create uncontrollable blazes.
By 1943, British bombing strategy evolved to create "firestorms"-massive conflagrations that altered local weather patterns. The technique involved simultaneous mass bombing with incendiaries across large areas, creating fires that would merge into an unstoppable inferno. The updraft from super-heated air created hurricane-force winds (up to 150 mph) that fed the flames.
Curtis LeMay emerged as the architect of America's firebombing strategy against Japan. When he took command in January 1945, he quickly abandoned precision bombing in favor of firebombing. For the March 9-10 Tokyo raid, he ordered B-29s to fly individually at low altitude, stripped of guns to carry more bombs, using a mixture of magnesium and napalm incendiaries.
The Tokyo fire created what the Japanese called a "red wind" (akakaze) reaching 28 mph, producing a sweep conflagration-a tidal wave of flame hundreds of feet high. The moving wall of flame projected radiant heat that would burn people before flames reached them, with temperatures reaching 1800F. People fleeing to canals boiled to death by the thousands. Between 80,000-120,000 people died that night.
This context explains why Truman never hesitated about using atomic bombs-the moral threshold of deliberately killing Japanese civilians had been crossed months earlier. For five months before August 1945, the U.S. had been systematically killing Japanese civilians. The atomic bomb was simply more efficient, doing with one bomb what previously required 300 bombers.
Chapitre 8
The Terrifying Gamble: Scientists Risking Earth's Atmosphere
In July 1942, as Manhattan Project scientists contemplated the theoretical possibility of atmospheric ignition, Arthur Compton made the remarkable judgment that it would be "better to accept the slavery of the Nazis" than risk ending all life on Earth. This was despite Hitler's regime having already begun murdering millions of Polish and Jewish civilians.
Initially, Compton and Oppenheimer agreed these bombs must never be made unless calculations proved they couldn't ignite the atmosphere. Yet the Manhattan Project continued at full speed, not because this possibility was disproven, but because the risk was deemed "very unlikely" though not impossible. Compton eventually decided that a "three in a million" chance was acceptable-a number essentially pulled from thin air.
As the Trinity test approached in July 1945, anxiety remained high. Enrico Fermi, far from joking, offered to take bets on whether the explosion would "burn New Mexico" or "ignite the whole world." He privately estimated the chance of atmospheric ignition at "about ten percent." Scientists like Fermi and Teller continued calculating and worrying until the final moments.
When the bomb finally detonated, many observers initially feared the worst. James Conant thought "the whole world has gone up in flames," while Sam Allison felt profound relief at still being alive. Remarkably, this existential gamble was taken solely by scientists-there's no evidence President Truman or other Washington officials outside the Manhattan Project were ever informed of the atmospheric ignition risk.
Three years earlier, Enrico Fermi had inspired Edward Teller to pursue an explosion a thousand times closer to actual doomsday-what would become the hydrogen bomb. Scientists understood that the fission bombs they were creating would be triggers for these vastly more powerful thermonuclear weapons, potentially destroying all human civilization.
After the Soviet Union tested a fission bomb in September 1949, Teller advocated for a crash program to develop the H-bomb. The General Advisory Committee of the Atomic Energy Commission unanimously rejected this proposal in October 1949, with members like Oppenheimer, Conant and others describing it in unprecedented moral terms as an "extreme danger to mankind" that "wholly outweigh any military advantage." Fermi and Rabi went further, calling it "almost one of genocide" and "an evil thing considered in any light," urging the President to renounce its development on "fundamental ethical principles."
Despite this opposition, President Truman announced on January 31, 1950 that he had directed the AEC "to continue with its work on all forms of atomic energy weapons, including the so-called hydrogen or super-bomb."
Chapitre 9
Dismantling the Doomsday Machine
Ellsberg argues that the risk of Doomsday could be eliminated without requiring total nuclear disarmament. While the risk of a single nuclear weapon destroying one city cannot be reduced to zero, the near-extinction threat to humanity can be eliminated through dismantling most existing weapons in both the United States and Russia.
The greatest remaining risk of nuclear winter comes from preemptive attacks triggered by false alarms or accidental detonations. Ellsberg proposes dismantling the Minuteman III missile force and reducing Trident submarine capabilities that target Russian land-based missiles. This would eliminate the strategic rationale for "launch on warning" policies on both sides.
Ellsberg reveals that U.S. nuclear strategy has not been about deterrence but about improving first-strike capability-a fundamentally illusory goal. He calls for Congress and the National Academy of Sciences to investigate nuclear war plans against the scientific findings on nuclear winter, but acknowledges this will require public pressure and likely whistleblowers revealing classified information.
The Soviets responded to U.S. decapitation threats with the "Perimeter" system, known informally as "Dead Hand." This automated retaliation system would allow low-level officers in deep underground centers to detect nuclear explosions in Moscow through multiple sensors. If Moscow was destroyed and communications severed, they could launch ICBMs that would send "Go" signals to missile sites, automatically launching Soviet missiles without human intervention at the sites.
Recent Russian news reports have confirmed Perimeter still exists, with Strategic Missile Forces Lieutenant-General Sergey Karakayev stating, "The system is on alert. If there's a need for a retaliatory strike, the command for an attack may come from the system, not people." Similar delegation or Dead Hand arrangements likely exist in every nuclear weapons state, meaning a Hiroshima-size explosion on any capital could trigger full-scale nuclear launches.
Ellsberg concludes that omnicide-the near extinction of life on Earth-is flatly illegitimate as an instrument of national policy and must be recognized as criminal and evil. He criticizes the dispassionate academic tone used when discussing nuclear policy, arguing that what's missing is recognition that these policies are "dizzyingly insane and immoral."
In his final thoughts, Ellsberg addresses the common denial that nuclear dangers "can't be that bad." He emphasizes that monstrous policies are not created by monsters, but by normal politicians, analysts, and military strategists-exemplifying Arendt's concept of the "banality of evil." He finds hope in historical precedents like the fall of the Berlin Wall and South African majority rule, concluding with Martin Luther King Jr.'s warning about "the fierce urgency of now" and the choice between "nonviolent coexistence or violent coannihilation."
Chapitre 10
The Whistleblower's Regret
Ellsberg harbored a secret during his two-year indictment for copying the Pentagon Papers, through the Watergate investigations, and for over forty years since. Beyond the Pentagon Papers, from fall 1969 until leaving RAND Corporation in August 1970, he copied everything in his Top Secret safe and more from his Secret and Confidential files-approximately 15,000 pages in total. Most of these "other Pentagon Papers" contained his notes and studies on classified nuclear war planning, command and control of nuclear weapons, and nuclear crises.
He intended to disclose it all, not just the Vietnam material, believing the nuclear revelations were even more important. He separated the nuclear documents and gave them to his brother Harry for safekeeping, planning to release them after the Vietnam material had run its course. Tragically, these nuclear documents were lost when tropical storm Doria hit Hastings-on-Hudson, washing away the buried box from the town dump where Harry had hidden it.
Ellsberg reflects that while he eventually obtained security clearances himself in 1958, they became "his undoing." At RAND Corporation, he and colleagues became preoccupied with preventing nuclear war by deterring a supposed Soviet surprise attack during the "missile gap"-a threat as unfounded as the earlier fear of a Nazi crash bomb program or later concerns about Saddam Hussein's WMDs.
"Working conscientiously, obsessively, on a wrong problem, countering an illusory threat," Ellsberg writes, "I and my colleagues at RAND had distracted ourselves and helped distract others from dealing with real dangers posed by the mutual superpower pursuit of nuclear weapons." He concludes that official secrecy and deceptions about nuclear weapons policies threaten human survival, emphasizing the need for radical changes toward eliminating Doomsday Machines and ultimately abolishing nuclear weapons.
