The Doomsday Machine

by Daniel Ellsberg

  This took courage. Even after the Soviet collapse, discussion of such topics remained guarded in Russia. In the early 1990s … Yarynich harbored a dream that someday both the United States and Russia might share the secrets of command-and-control. He was certain it could lead to deterrence with far fewer nuclear warheads. He also favored taking missiles off launch-ready alert. He tirelessly expounded his logic, yet governments were not interested. The high priests of nuclear command-and-control could not envision opening up to each other, not here, nor in Russia.

  The bottom line is that arrangements made in Russia and the United States have long made it highly likely, if not virtually certain, that a single Hiroshima-type fission weapon exploding on either Washington or Moscow—whether deliberate or the result of a mistaken attack (as in Fail Safe or Dr. Strangelove) or as a result of an independent terrorist action—would lead to the end of human civilization (and most other species). That has been, and remains, the inevitable result of maintaining forces on both sides that are capable of causing nuclear winter, and at the same time are poised to attack each other’s capital and control system, in response to fallible warnings, in the delusion that such an attack will limit damage to the homeland, compared with the consequences of waiting for actual explosions to occur on more than one target.

  Here, then, is the actual situation that has prevailed for more than half a century. Each side prepares and actually intends to attack the other’s “military nervous system,” command and control, especially its head and brain, the national command headquarters, in the first wave of a general war, however it originates. This has become the only hope of preempting and paralyzing the other’s retaliatory capability in such a way as to avoid total devastation; it is what must above all be deterred by the opponent. But in fact it, too, is thoroughly suicidal unless the other side has failed to delegate authority well below the highest levels. Because each side does in fact delegate, hopes for decapitation are totally unfounded. But for the duration of the Cold War, for fear of frightening their own publics, their allies, and the world, neither side discouraged these hopes in the other by acknowledging its own delegation.

  The only change in this situation has been that in the first weeks of the Trump administration, Russian news reports have begun acknowledging that the Perimeter system persists. In a February 2, 2017, article, Pravda revealed that the commander of Strategic Missile Forces Lieutenant-General Sergey Karakayev said five years ago in an interview in a Russian publication, “Yes, the ‘Perimeter’ system exists.250 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.” The Pravda report explained, “Nuclear-capable missile will thus be launched from silos, mobile launches, strategic aircraft, submarines to strike pre-entered targets, unless there is no signal from the command center to cancel the attack. In general … one thing is known for sure: the doomsday machine is not a myth at all—it does exist.”

  Ten days after President Trump’s inauguration251 in 2017, Pravda quoted his statements that “the United States should strengthen and expand the nation’s nuclear capacity” and “Let it be an arms race,” and then reported that “Not so long ago, the Russian Federation conducted exercises to repel a nuclear attack on Moscow and strike a retaliatory thermonuclear attack on the enemy. In the course of the operations, Russia tested the Perimeter System, known as the ‘doomsday weapon’ or the ‘dead hand.’ The system assesses the situation in the country and gives a command to strike a retaliatory blow on the enemy automatically. Thus, the enemy will not be able to attack Russia and stay alive.”

  What has not changed is American preoccupation with threatening Russian command and control: as if all the above revelations, including those of Blair and Yarynich, had not occurred or were thoroughly disbelieved. The National Defense Authorization Act for Fiscal Year 2017,252 passed with bipartisan support and signed by President Obama on December 23, 2016, included a provision which mandated a report by the Office of the Director of National Intelligence and the Strategic command on “Russian and Chinese Political and Military Leadership Survivability, Command and Control, and Continuity of Governmental Programs and Activities.” This provision of the law called for the U.S. Strategic Command to “submit to the appropriate congressional committees the views of the Commander on the report … including a detailed description of how the command, control, and communications systems” for the leadership of Russia and China, respectively, are factored into the U.S. nuclear war plan. The Pravda news stories quoted above, both appearing in the second week of the Trump administration, were explicitly responding to these provisions of this law signed a few weeks earlier in their explanation of the continuing need for Perimeter.

  Such plans and capabilities for decapitation encourage—almost compel—not only the Perimeter system but Russian launch on (possibly false) warning: either by high command (in expectation of being hit themselves imminently, and in hopes of decapitating the enemy commanders before they have launched all their weapons) or by subordinates who are out of communication with high command and have been delegated launch authority.

  As General Holloway expressed it in 1980, he had confidence that with such a decapitating strategy, a U.S. first strike would come out much better for the United States than a second strike, to the point of surviving and even prevailing. He was right about the hopelessness of the alternative forms of preemption. But in reality, the hope of successfully avoiding mutual annihilation by a decapitating attack has always been as ill-founded as any other. The realistic conclusion would be that a nuclear exchange between the United States and the Soviets was—and is—virtually certain to be an unmitigated catastrophe, not only for the two parties but for the world. But being unwilling to change the whole framework of our foreign and defense policy by abandoning reliance on the threat of nuclear first use or escalation, policy makers (probably on both sides) have chosen to act as if they believed (and perhaps actually do believe) that such a threat is not what it is: a readiness to trigger global omnicide.

  There is every likelihood that, for comparable reasons, similar secret delegation or Dead Hand systems or arrangements exist in every other nuclear weapons state—China, Britain, France, Israel, India, Pakistan, and North Korea—meaning that a Hiroshima-size explosion on any one of their capitals and/or central military headquarters is likely to lead to full-scale launching of their ready forces. The only difference is that none of these states could, at present, cause a full-scale nuclear winter, though an exchange between any two of them (except North Korea) could trigger enough global reduction in sunlight and loss of harvests for a decade to cause nuclear famine and the starvation of one to two billion people or more.253

  The Strangelove paradox afflicts not only the United States and Russia. Every new state that acquires nuclear weapons and comes face-to-face with the vulnerability both of the weapons systems and of the command and control apparatus confronts the same incentives, the same pressures from its military to delegate and sub-delegate authority to use them, and the same motives to keep that delegation secret from the rest of the world.

  Deployment of nuclear weapons by a new state doesn’t add just one new finger to a trigger on nuclear war. The world worries about the finger of an irresponsible or reckless third world leader, when the finger can just as easily be that of one of many functionaries working in a far-flung outpost for one of these leaders, new or old.

  The bottom line, once again: This is not a species to be trusted with nuclear weapons. Above all, not to be trusted with a full or partial Doomsday Machine. And that doesn’t just apply to “crazy” third world leaders.

  CHAPTER 20

  First-Use Threats

  Using Our Nuclear Weapons

  From an April 25, 1972, tape of Oval Office conversations, we have the following exchange between Richard Nixon and national security advisor Henry Kissinger, regarding possible American responses to an ongoing North Vietnamese offensive:


  PRESIDENT: I still think we ought to take the dikes out now. Will that drown people?

  HENRY KISSINGER: About two hundred thousand people.

  PRESIDENT [reflective, matter-of-fact]: No, no, no … I’d rather use the nuclear bomb. Have you got that, Henry?

  KISSINGER [like the president, low-key]: That, I think would just be too much.

  PRESIDENT [in a tone of surprise]: The nuclear bomb, does that bother you? I just want you to think big, Henry, for Christsakes.

  It was not the first time that Nixon had entertained such big thoughts. As his former chief of staff, H. R. Haldeman, reported in a memoir written as he awaited a prison term for his role in the Watergate scandal, Nixon had shared elements of his plan to end the Vietnam War during his presidential campaign in 1968.

  Nixon not only wanted to end the Vietnam War,254 he was absolutely convinced he would end it in his first year.

  … The threat was the key, and Nixon coined a phrase for his theory which I’m sure will bring smiles of delight to Nixon-haters everywhere. We were walking along a foggy beach after a long day of speechwriting [during Nixon’s presidential campaign in 1968]. He said, “I call it the Madman Theory, Bob. I want the North Vietnamese to believe I’ve reached the point where I might do anything to stop the war. We’ll just slip the word to them that, for God’s sake, you know Nixon is obsessed about Communism. We can’t restrain him when he’s angry—and he has his hand on the nuclear button—and Ho Chi Minh himself will be in Paris in two days begging for peace.”

  When I read this in 1978, I had a very uneasy feeling that I might have been the source of this crazy scheme, or at least for the phrase he had coined. I had given two lectures to Henry Kissinger’s seminar at Harvard in 1959, one titled “The Political Uses of Madness.” It was from a series of lectures I had given that spring on bargaining theory, with the overall title “The Art of Coercion: A Study of Threats in Economic Conflict and War.”

  To illustrate a counterintuitive proposition in bargaining theory I had arrived at, I pointed to the difficult challenge of making a credible threat to initiate nuclear attacks on a nuclear-armed state or one of its allies. After all, this amounted to a threat of massive suicide-murder. The consequences of carrying out this threat were so fearsome that it didn’t have to be very credible to be effective in achieving compliance. But at the same time, the consequences for the threateners themselves were such that it was a challenge to make their threat credible at all.

  I had described, as an example of one possible, though dangerous, solution to this problem, Hitler’s deliberate use of his reputation for madness255 and unpredictability—impulsiveness, recklessness, rage—to intimidate his adversaries and make his threats and ultimatums effective in the period prior to his actual invasions in World War II. Contrary to the expectations of his own generals, his blackmail had actually succeeded spectacularly in his bloodless occupation of the Rhineland, Austria, the Sudetenland, and Czechoslovakia. So the image of mad unpredictability could work. It actually had, for Hitler, but that was in considerable part because he actually was crazy, madly reckless and aggressive. I had never thought of it as an approach that would appeal to an American leader, nor be remotely advisable under any circumstances.

  When I read Haldeman’s memoir, I momentarily worried that Nixon had gotten at least the nickname for his “madman theory,”256 and perhaps even the concept, from Kissinger—that is, indirectly, from me. To my relief, as I read the account again closely, Haldeman placed the date of this conversation with Nixon in 1968—before Nixon had met Henry Kissinger for the first time in the fall of 1969. For good or ill—and there was nothing good about it—Richard Nixon had adopted this reckless policy without inspiration from Kissinger or me.

  Rather, the idea of achieving his secretly ambitious aims in Vietnam by nuclear threats had come from a more authoritative source: Dwight Eisenhower, under whom Nixon had served for eight years as vice president. As Haldeman recounted in this same passage about his madman theory, Nixon “saw a parallel in the action President Eisenhower had taken to end another war. When Eisenhower arrived in the White House, the Korean War was stalemated. Eisenhower ended the impasse in a hurry. He secretly got word to the Chinese that he would drop nuclear bombs on North Korea unless a truce was signed immediately. In a few weeks, the Chinese called for a truce and the Korean War ended.” (Nixon felt that Eisenhower, as former supreme commander in Europe in World War II, did not need the hint of madness that Nixon himself did for his nuclear threats to be credible, but, according to Haldeman, Nixon “believed his hard-line anti-Communist rhetoric257 of 20 years would serve to convince the North Vietnamese equally as well that he really meant to do what he said.”)

  It was not only Nixon who believed that nuclear threats were critical to achieving the armistice in Korea that has held—uneasily, at this moment—for the last sixty-four years. Eisenhower did himself. His former White House chief of staff Sherman Adams reported asking Eisenhower later how an armistice had at last been reached in Korea. “Danger of an atomic war,” he said without hesitation. “We told them we could not hold to a limited war any longer if the Communists welched on a treaty of truce. They didn’t want a full-scale war or an atomic attack. That kept them under some control.”258 His secretary of state John Foster Dulles gave this same explanation.

  Whether such threats actually affected the Chinese decision makers or whether they even received them remains uncertain and controversial. What is neither uncertain nor inconsequential is that the Eisenhower administration, including Richard Nixon, regarded them as successful. In line with this belief, Eisenhower and Dulles relied on such threats repeatedly, in a series of crises. Dulles’s self-congratulatory account in 1956 of the risk-taking strategy underlying the first few of these threats gave rise to the term “brinkmanship.” In words that echo throughout the Cold War—in fact, words that virtually define “cold war” in a sense that is returning in the last few years—Dulles said:

  Some say that we were brought to the verge of war.259 Of course we were brought to the verge of war. The ability to get to the verge without getting into the war is the necessary art. If you cannot master it, you inevitably get into war. If you try to run away from it, if you are scared to go the brink, you are lost.

  And as I was to discover soon after Nixon left office, this strategy did not end with Dulles and Eisenhower.

  In September 1974, just after Nixon’s resignation, Roger Morris, a former aide to Henry Kissinger, revealed for the first time in the magazine Washington Monthly that Nixon had directed plans for nuclear attacks on North Vietnam in October–November 1969. Morris had participated in an “October group” in the White House planning what his boss, Kissinger, had asked to be a “savage, brutal blow” that would bring the “little fourth-rate country” North Vietnam to its “breaking point.” When I asked Morris for specifics after reading that article, he told me that he had read detailed planning folders, with satellite photographs, for several nuclear targets in North Vietnam. One of these, a trans-shipment site for materiel coming in from China, was a mile and a half from the Chinese border. A nuclear attack on this was meant to send a strong “signal” to China. A low-yield airburst nuclear weapon above this railroad spur in the jungle, the planning folder estimated, would cause only “three civilian” casualties. Another prospective nuclear target was the Mu Gia pass from North Vietnam into the Ho Chi Minh trail in Laos.

  I had not known at the time about this planning going on in October 1969, just as I was beginning to copy the Pentagon Papers. I would have been astonished to learn of it, so soon in Nixon’s first term, even though I was already concerned that another, eventual North Vietnamese offensive, perhaps three or four years off, might trigger use of nuclear weapons. What I had just learned from my friend Morton Halperin, a deputy to Henry Kissinger who had left the National Security Council in September, was that Nixon—contrary to all public expectations—was not planning to withdraw from Vietnam unconditio
nally but was threatening to escalate dramatically to achieve a quasi-victory.

  Mort told me of the then-secret bombing of Cambodia that was already under way. That was precisely to demonstrate to the North Vietnamese, he said, that despite what the electorate had been led to believe, Nixon was ready to go beyond what LBJ had ever been willing to do. Other measures threatened—not in bluff—included invasion of the “sanctuaries” of Cambodia and Laos, mining of Haiphong, unrestricted bombing of cities and towns in North Vietnam up to the Chinese border, and possible invasion of North Vietnam. Warnings to the Soviet ambassador Anatoly Dobrynin as early as May, Halperin told me, had implied a readiness to use nuclear weapons if Nixon’s terms were not met. But neither of us imagined then that Nixon was prepared to do this in the fall of his first year.

  Nevertheless, Halperin’s revelations to me of the president’s secretly ambitious aims and his reliance on threats of escalation to achieve them were enough to prompt my decision to copy the Top Secret Pentagon Papers. I was sure his threats would not succeed, and would instead prolong the ground war and enlarge the air war, with heavy further casualties on both sides. If I had known then about Nixon’s imminent nuclear threats and plans and had any documents on these, I would have revealed them immediately, instead of the history in the Pentagon Papers, which ended in 1968 before Nixon came to office.

  Later, when the papers were published in 1971, Henry Kissinger’s fear that I did know about Nixon’s nuclear threats and plans, and might have documents to back it up, was sufficient reason for him to regard me as “the most dangerous man in America,”260 who “must be stopped at all costs.” As I mentioned in the introduction, it was the unlikely exposure of White House crimes against me—actions precisely intended to avert my revealing documents from the Nixon administration, beyond the period of the Pentagon Papers—that led to Nixon’s resignation facing impeachment, making the war endable nine months later.