to multiple-warhead ballistic missiles based on land and under the sea.
At these levels of numbers, what use were nuclear threats? Ironically, since every scenario for a major exchange led down the same path of annihilation, the U.S. and Soviet heartlands were now safer from direct assault, since neither side could chance a first strike. The sneak attack scenario feared in the 1950s—the so-called “BOOB,” or “bolt out of the blue,” attack—was no longer possible: a first-strike could neither disarm the victim nor save the attacker. The much greater complication, with North America itself now vulnerable to Soviet weapons, was not whether the United States could defend itself, but whether the Americans would risk nuclear war for their allies in NATO.
The attempt to protect one group of nations with the nuclear weapons of another is the problem of “extended deterrence.” To kill in self-defense, or in defense of the family or group, is a common human instinct. Dying for others requires overcoming the instinct for self-preservation with altruism and a notion of a greater good. It may well be, as the New Testament teaches, that there is no greater love than that “a man lay down his life for his friends,” but U.S. leaders realized early on in the Cold War that to risk the lives of entire nations and the peace of the world itself on behalf of others was a more complicated proposition, especially after the two world wars had taught humanity a bloody lesson in what the defense of alliances could mean. The Soviets might not doubt that the Americans would make a brave last stand and use nuclear weapons to protect the North American homeland. But could the United States make an equally credible nuclear threat on behalf of their friends in Europe?
The United States was in a painful bind. Washington could hardly back away from a nuclear guarantee for NATO without seeing the Alliance crumble, first politically and then militarily, under the Soviet conventional threat. There was no alternative to defending Europe, but in the face of Soviet conventional superiority, the defense of Europe was impossible without keeping nuclear weapons in play.
When the United States held a nuclear monopoly, a strategy such as Massive Retaliation could more credibly threaten to punish the Kremlin for aggression against NATO. Holding the nuclear trump card lowered the cost to the Americans of making threats on Europe’s behalf, not least because the consequences of any test of that resolve would fall almost entirely on the Soviet Union. But once the USSR obtained a secure retaliatory capability, that guarantee would now have to extend to the point of being willing to place the continental United States squarely in the Soviet nuclear crosshairs. Thus, extended deterrence was an immense gamble: it rested not on the intuitive understanding of self-defense, but increasingly on the imponderable question of whether a U.S. president would really risk trading Chicago for Bonn or New York for Paris.
Accordingly, the first order of business for U.S. strategists in the early 1960s was to scrap Massive Retaliation, ending its short tenure within only a few years of its announcement. Future U.S. presidents and their advisors would need more options to deal with a full spectrum of Soviet aggression other than the single choice of incinerating the USSR. A strategy such as Massive Retaliation could not serve the goals of extended deterrence, because it would require the Soviets to believe that an American president would be so steely—or so unhinged—that he would escalate directly from conventional hostilities in Europe to central nuclear war between the United States and the USSR. The key to strengthening the Western deterrent thus relied on showing the Soviet leaders that their American counterparts would, in effect, be left with no other way out but nuclear conflict. A new strategy would have to chart a path, one the Soviets could understand, that would credibly link a crisis to a catastrophe, and therefore induce caution in Moscow.
In order to establish the requirements for a more credible deterrent, Western thinkers began to explore the actual steps to nuclear war. In 1957, for example, Henry Kissinger published his seminal work, Nuclear Weapons and Foreign Policy, and the generation of strategists McNamara would later call upon were already working through these problems at research centers such as the RAND Corporation in California. It was at RAND that Herman Kahn first conceptualized the escalatory “ladder” described in his 1960 opus, On Thermonuclear War. Kahn’s ladder was not a guide to nuclear war itself; rather, it was a detailed examination of various steps each side might take before finding themselves, willingly or not, propelled from a cold peace to complete destruction. Each rung of the ladder, such as “limited evacuation of cities” or “barely nuclear war,” was a stop along the way toward the final step of “spasm” or “insensate” war.19 Massive Retaliation was not credible because it was a threat to make an intentional leap from the lowest rungs of the ladder of escalation to the very highest. The challenge for U.S. policymakers trying to craft a new deterrent in its place, then, was to reduce the distance between each of these rungs by inserting realistic options that did not require momentous or even irrational decisions.
Shortly after arriving in Washington, the Kennedy administration set about trying to construct a more durable U.S. deterrent, both conceptually and materially. The first trial balloon from the Kennedy White House, however, was a hopelessly optimistic “no-cities” strategy, in which Washington and Moscow somehow would agree to refrain from targeting each other’s civilian population centers. This proposal represented an American attempt to draw a distinction between “countervalue” attacks, which would strike a full range of social and political targets such as cities and government institutions, and “counterforce” attacks, which would be aimed only at military assets. The goal, at least in theory, was to enhance crisis stability by showing a willingness to discriminate between military targets and cities; presumably, an enemy would be more willing to tough it out and not attack during a tense period if assured that the other side had not targeted civilians and urban centers. At the least, an offer to limit targets might spur a similar pledge from the opponent and keep a nuclear exchange from raging out of control. Ideally, it would help avert war itself not only by sending a message of restraint to the Soviets, but by showing that the United States had come up with a real purpose for nuclear weapons besides the mindless killing of Soviet citizens.20
“No cities” was one of many attempts to place a rung on Kahn’s ladder somewhere between the outbreak of conventional hostilities and total nuclear war. Like Massive Retaliation, however, it was inherently flawed. Indeed, it was a strategy that only a pure theorist could love, and it could not survive first contact with the real world, where its success would have to rely on the goodwill of a cooperative adversary in the midst of a possible holocaust. Even if both belligerents could reach some sort of prior agreement about conducting a nuclear war, a strategy of “city-avoidance” was doomed from the start by the fact that so many Soviet targets, and no small number of American assets, were located close to population centers. Were Moscow and Washington, the military nerve centers of their respective nations, to be spared? (And if they were slated to be destroyed, who would be left to negotiate a ceasefire or surrender on either side?) The sanctity of cities could never be guaranteed, and a promise not to hit them, or at least not to hit them in the first thirty minutes of the war, was not a promise worth making.
After a period of vigorous debate in the 1960s (during which the shock of the 1962 Cuban missile crisis nearly rendered the whole Soviet-American conflict moot), the United States and its allies in 1967 codified a strategy of “Flexible Response,” which NATO described as “a flexible and balanced range of appropriate responses, conventional and nuclear, to all levels of aggression or threats of aggression” (emphasis added).21 Here, the Americans and their allies tried to overcome the credibility problems inherent in “extended deterrence,” and to link the defense of North America to the defense of the entire North Atlantic community. Rather than threaten discretionary retaliation at “times and places of our choosing” with the consequent killing of millions of civilians on both sides, U.S. and NATO strategists instead crafted a strategy of deliberate escalation backed by a wider menu of both conventional and nuclear military choices. Now, NATO would fight its way up the escalatory ladder, instead of jumping from the first rung to the last.
As before, the conventional defense of NATO in Europe had virtually no chance of succeeding against a Soviet invasion. That, of course, was the point: the United States and its allies might not resort to strategic nuclear attacks at the outset of the war, but if faced with a Soviet onslaught, they would have no choice but to escalate to tactical nuclear