by the eventual long-term recovery of one system more rapidly than the other and the subsequent domination over whatever was left of the world. This position—echoed at times by some American politicians—remained the official Soviet line throughout the 1970s.23 In both private and public diplomacy, however, Moscow acted with considerably more circumspection than its rhetoric might have suggested. In 1981, Soviet leader Leonid Brezhnev finally referred to the idea of nuclear victory as “dangerous madness,” which was a hopeful sign. Later revelations, however, showed that Brezhnev’s comments were actually part of an ongoing civil-military struggle at the highest levels of the Soviet regime over the issue of nuclear war.24 Whether the Kremlin’s high command really believed the USSR could win a nuclear war can never truly be known. It might have been a bluff to intimidate the Americans (who had more than a few generals and strategists with similar views themselves), or it may have represented an inability to think beyond the parallels with World War II that dominated Soviet military writings.25
Most American analysts were far less sanguine than the Soviet brass about the outcome of World War III, and believed that both sides would be destroyed in any sizable nuclear exchange. Initially, this new doctrine was called “assured destruction,” and finally, by the wry acronym that its founders believed best described it: “MAD,” or mutual assured destruction. Although MAD seemed like a simple idea in which the victim and the attacker both perish, it was actually far more complicated. Even its various proponents did not fully agree on what it meant, and there were competing notions of MAD during the Cold War that were not fully consistent with each other. Some accepted the possibility of limited nuclear use, while the most pristine version assumed that nuclear war inevitably meant total war and subsequent annihilation. In the end, however, even the Soviets had to bow to the reality that MAD itself was a fact rather than a policy, in the sense that the ability of both sides to destroy each other was empirically undeniable.26
More arguable is what that fact meant for actual policy. Proponents of MAD argued that to ignore the reality of mutual destruction was like trying to ignore the weather, which is always present and always the same for everyone. Opponents of MAD—many of them among the civilian and military national security elites both in the United States and USSR—rejected the helplessness and passivity implied in a doctrine of mutual destruction. They argued that even if MAD was a fact, it was self-defeating to admit it, and they rejected the idea of explicitly accepting a dangerous and delicately balanced nuclear standoff as at best counterintuitive, if not flatly ridiculous.27
As far as McNamara and the MAD advocates were concerned, Soviet and American arsenals had reached such staggering levels of quantity and destructive quality that mutual destruction was inescapable. The American public, however, was understandably skeptical about a strategy whose most prominent feature was to leave the United States open to attack. MAD, at least to some critics, sounded like a retreat in the face of the new and improved Soviet nuclear arsenal. Some of those critics were members of Congress, and McNamara consequently found himself having to reassure concerned legislators on Capitol Hill that the United States could still kill most of the Soviet population if necessary. “I think we could all agree,” he said in testimony before the Senate in early 1967, “that if [the Soviets] struck first we are going to target our weapons against their society and destroy 120 million of them.”28
The emergence of MAD as an explicit concept did not end the nuclear debate in either the United States or the Soviet Union. Rather, strategists on both sides intensified their search for ways out of the mutual destruction cage by trying to find more discriminating and thus supposedly more credible military uses for nuclear weapons. Opponents of a pure deterrence strategy, especially among the senior U.S. military leadership, argued that the only way to deter a nuclear war was to appear ready to fight one. Thus, as Kaplan notes, McNamara “talked Assured Destruction,” but “the actual targeting strategy” created by the U.S. military remained counterforce, and the Secretary’s attempts to use a doctrine of assured destruction as a means of “beating back the excessive demands of the military and [Joint Chiefs]” failed in the face of the Pentagon’s insistence on being able to fight a global nuclear war.29 (Two critics of this approach later sarcastically referred to these warfighting requirements as a strategy of “Nuclear Utilization Target Selection”—that is, “NUTS” as opposed to “MAD.”)30
This inherent tension between targeting and doctrine continues in U.S. nuclear strategy to the present day. Although America’s nuclear doctrine during the Cold War was termed “assured destruction,” its nuclear forces were designed and postured for both a counterforce war as well as retaliatory strikes. Military logic demanded that the most pressing targets for U.S. strategic forces should be the same type of enemy nuclear forces, followed by other military targets, energy and industrial infrastructure, and finally, population. This ordering of strikes was meant to communicate to the enemy that in the event of nuclear war, the United States would act like a rational military power and seek first and foremost to protect itself (while sparing innocent civilians as much as possible), and then proceed to strip the enemy of its ability to fight, its ability to govern, its ability to recover, and finally, its ability to exist at all. The paradox, of course, was that a readiness to proceed along so orderly a path of escalation also ran the risk of signaling that that the United States did not fear “assured” destruction and was willing to fight a nuclear conflict—thus theoretically giving the enemy an incentive to abandon further efforts at deterrence, and instead to strike first and hope for the best.
The advocates of MAD had the better argument about nuclear targeting, not only because of the destructive power of nuclear weapons, but because of the continual problem of “co-location,” meaning the proximity of military targets to nonmilitary areas. This is not a dilemma specific to nuclear war, of course: imprecise conventional bombers in World War II could not avoid devastating the areas around military targets placed in or near cities. Likewise, there was no way the nuclear counterforce scenario could really discriminate between military and civilian targets in any meaningful way—or at least not in a way that would allow the enemy to distinguish a limited strike from an all-out attack.
While there might be some hypothetical chance of limiting the use of nuclear arms on the battlefield, or even in the theater of war, those limits would almost certainly evaporate minutes into a strategic exchange. An attack meant only to cripple U.S. land-based missiles, for example, would require several hundred, if not thousands of warheads. Such a strike would be so large that the destruction would be indistinguishable from an attempt at national obliteration, especially in the midst of the kind of chaos which would surround a U.S. president in the few moments left for deliberation over how to respond. No matter what either side wanted or intended, the terrible imperative of MAD would quickly assert itself once such an exchange was under way
The search for a silver bullet that could solve this dilemma and allow a longer interval of nuclear warfighting (and thus ostensibly limit the conflict early) inevitably led both sides to think about missile defenses. By the mid-1960s, land- and sea-based intercontinental ballistic missiles (ICBMs) constituted the main Soviet and American deterrent forces, and once the superpowers could destroy each other in a matter of minutes, the urge to find a way out of that paralyzing reality was sure to follow. The idea of a national missile defense capable of destroying nuclear-armed, long-range ballistic missiles in flight was a natural extension of the Cold War arms race.31
The discussion about missile defenses in the 1960s was more important as a conceptual exercise rather than as practical effort. Missile defenses of any kind are complicated and difficult to construct even in the twenty-first century, but national defenses were virtually impossible in the era before the development of more advanced technologies. Supercomputers with only a fraction of the speed and computing power of a modern notebook or laptop computer cost millions of dollars each in the 1960s and 1970s. (Consider, for example, that the recently retired U.S. space shuttle at first carried less computing power aboard than people take for granted in their mobile phones today.) Moreover, “defenses” in the 1960s were not precise weapons that could cleanly eliminate incoming strikes and thus save
