Destroying the Village: Epilogue

Destroying the Village: Eisenhower and Thermonuclear War

Campbell Craig

1998

Epilogue: McNamara’s Dialectic

The Kennedy administration might well have decided after Berlin simply to adopt Eisenhower’s strategy of evading war and be done with it. Whether he realized it or not, the strategy had worked well for Kennedy during the fall of 1961, by making the plausible option of war unavailable when the East Germans erected the Wall and when Clay instigated the Friedrichstrasse standoff. Moreover, staying with the all-or-nothing policy was the course of least resistance, always a major factor in Kennedy’s decisionmaking. To adopt an alternative policy would mean having to get more money from Congress, alarming NATO allies, provoking Khrushchev, confronting Norstad, and disrupting a stable element of Cold War rivalry. Certainly it would be easier to let things remain as they were.

The man primarily responsible for fighting this potential inertia and maintaining the cause of flexible response was Secretary of Defense McNamara. For McNamara, the Berlin crisis had demonstrated above all else why the United States needed to obtain a flexible military policy as soon as possible. The administration had been handcuffed by Eisenhower’s policy—forced to tremble during the summer of 1961, accept the Wall passively in August, and then face the prospect of spasmodic thermonuclear war over Clay’s game at Friedrichstrasse. To be sure, McNamara had done well to evade war in October, but like Dulles during the Quemoy–Matsu crisis he did this only because for the moment he had no other choice. Like Dulles, McNamara viewed that experience not as an example of what to do in future crises, but as a regrettable necessity not to be permitted again.

With no one else in the administration determined or powerful enough to stop him, McNamara succeeded during late 1961 and early 1962 in making limited war against the communist bloc a genuine option of American military policy for the first time since 1955. This heralded a triumph for the idea that a major war, and particularly a nuclear war against the Soviet Union, could be limited and won—an idea promoted by people like Taylor, Kissinger, and now above all the brilliant theorist Thomas Schelling. But McNamara’s triumph was short-lived, killed in its infancy by the events in Cuba during October 1962.

Schelling’s Thesis

As we have seen, during the late 1950s strategists in think tanks and universities across America had been working intensively to develop an alternative to Eisenhower’s all-or-nothing strategy. Kissinger, Taylor, and many others besides had argued in favor of a strategy that would dictate reacting to Soviet aggression with a limited form of retaliation: by developing flexible military strategies, the United States could keep a general war with the Soviet Union from progressing all the way to thermonuclear annihilation.

There were two problems with these original strategies of flexible response. First, as has been pointed out already, they danced around the central deficiency of all limited nuclear war strategies: the question of how a nation stops fighting when it still has immensely destructive weapons at its disposal. Eisenhower had foreseen this problem from the beginning, and had made it the conceptual basis of his all-or-nothing military policy. A second problem, and one perhaps of greater interest to McNamara and his colleagues, was these strategies’ rather bleak message. Both Taylor and Kissinger argued that a general war with the Soviet Union could be limited, and won, at a level somewhere below all-out thermonuclear destruction. But how attractive really was a strategy that promised fifty million dead rather than 100 million, ten years of national recovery rather than thirty? Was there not something better than that? The brilliant theorist and economist Thomas Schelling of Harvard believed that there was. ₁

In 1960 Schelling published Strategy of Conflict, a pioneering study that applied game theory to human conflict, especially warfare. ₂ In a chapter devoted particularly to the question of nuclear war, entitled "The Threat That Leaves Something to Chance," Schelling unveiled an argument that departed substantially from the early ideas of flexible response—both from the published strategies of intellectuals like Kissinger and from the policy recommendations of John Foster Dulles. Schelling believed that the United States could use nuclear weapons in an aggressive manner to intimidate the Soviet Union, raise the specter of all-out war, and prevail in a Cold War crisis. Such a result would be better than Eisenhower’s choice of conciliation or all-out-war; it would also be superior to the limited wars of Kissinger and Taylor.

The threat of nuclear retaliation, Schelling argued, rests upon the ability of one side to persuade the other that in the event of a given transgression punishment will, not may, happen. The adversary must fear that if it makes a false move retaliation, and maybe holocaust, will follow. Eisenhower’s all-or-nothing strategy, by placing before the Soviet Union the unreal threat of blowing up the world over incremental forms of aggression, lacked this credibility.

A more credible nuclear policy, Schelling continued, should seek to convey the message to the other side that things will get out of hand, should aggression occur, even if the deterrer would rather avoid it. ₃ By provoking a crisis or starting a local war the aggressor would be playing with fire, pushing the deterrer toward a reaction that would inadvertently spiral into spasmodic war. It is to the deterrer’s advantage to cultivate this fear, to demonstrate by action that even modest aggression is going to lead to uncontrollable escalation. While Eisenhower’s strategy could not be faulted for failing to emphasize the specter of uncontrollable war, it lacked the mechanism to "demonstrate by action" to the Soviets what was soon to follow.

An effective way to show the Soviet Union that holocaust was to follow would be to "introduce new weapons" during the early stages of a conflict ₄ somewhere on enemy territory. The purpose of doing this is not to attain a tactical end, like blowing up a railway or a tank division, but to illustrate vividly what will happen unless the aggressor backs away at once. This escalation signifies that armageddon is nigh. That neither side wants this to happen is obvious; the terror exists for everyone. But the only way it can be stopped is for the aggressor to retreat. By doing so, it saves not only itself, but civilization as well. ₅

A new fact, Schelling wrote, exists in international politics: the world’s great powers were all terrified by the idea of general war. The United States could exploit this new fact by deliberately, actively, raising the specter of general war in order to force the Soviet Union to back down or unleash armageddon. Schelling called this kind of activism compellence. ₆ One does not wait, terrified, for something to happen: one takes the initiative. During times of crisis it is better to administer "punishment until the other acts, rather than if he acts." ₇ To prevail in conflict, one must neither shy away from facts, nor throw up one’s hands and say that nothing can be done, but use them boldly to obtain advantage. ₈ Instead of giving in to nuclear fear, use it.

Immediately following the publication of Strategy of Conflict Schelling elaborated upon his argument on nuclear war, both in a public scholarly article and in a private memorandum delivered to the president. In these two writings, Schelling made his ideas known to a wider audience and applied them more specifically to potential international crises.

In a 1962 article (written in 1961) published in the influential journal World Politics, Schelling made his recommended course of action for conflicts like the one happening in Berlin very clear. Eisenhower’s military policy was wholly unsuited to crises like these, because neither local resistance nor all-out retaliation were acceptable options in a war over an intermediate Cold War stake. In between these two extremes, Schelling wrote, "there is the strategy of risky behavior, of deliberately creating a risk that is credible precisely because its consequences are not entirely within our own and the Soviets’ control." ₉

What kind of risky behavior should this be? It is worth recalling Norstad’s statement in the fall of 1961 that if the communists shut off access to West Berlin the United States could respond at the probe level; once a battalion was defeated, the U.S. would have to choose between nuclear war and defeat. This was America’s military option in 1961. In this light, Schelling’s elaboration deserves to be quoted at length:

If we wish to convey that the war is getting out of hand, that it will shortly become locally very destructive in spite of efforts to confine it to military targets, we should pick military targets that cause destruction commensurate with the notion we want to convey.. . . In a nuclear exchange, even if it nominally involves only "tactical" weapons against tactically important targets, there will be a conscious negotiating process between two very threatening enemies who are worried that the war will get out of hand.. . .

In the desperate circumstances in which we were about to lose Western Europe, if there were no prospect of militarily winning an all-out war against the Soviet Union, and if there were little reason to suppose that the Soviets could expect to win militarily an all-out war against us, some kind of limited punitive warfare would almost certainly suggest itself. ₁₀

Here was an argument that clearly (despite the obfuscative passive voice at the end) made the case for an activist, aggressive policy of nuclear intimidation in crises like Berlin. The next time the United States gets into a showdown with the Soviet Union, Schelling was saying, it should "deliberately" create a risk designed to warn the Soviet Union that major war was about to erupt if it did not back off. Should the Soviets not surrender at this point, Schelling’s next step is clear (is it not?): the United States should "pick military targets" with the aim of persuading the Soviets that the war is about to become locally very destructive; tactical nuclear weapons, he writes, are well suited for this purpose. If at that point the Soviets did not give in, it would be time for pre-emptive nuclear war. Such a strategy would make everyone’s fear of nuclear war an advantage for the United States, for only a Soviet leadership made of stone would not back down.

At the same time he was composing this article, Schelling was invited by the Kennedy administration to comment on the Berlin Crisis. His paper, "Nuclear Strategy in the Berlin Crisis," was included in Kennedy’s reading packet for the trip to Hyannis Port in July. Once again, Schelling’s laid out his strategy of compellence clearly: ₁₁

The important thing in limited nuclear war is to impress the Soviet leadership with the risk of general war—a war that may occur whether we or they intend it or not. If nuclear weapons are introduced the main consequence will not be on the battlefield; the main consequence will be the increased likelihood and expectation of general war.

. . . Limited and localized nuclear war is not, therefore, a "tactical" war. However few the nuclears used, and however selectively they are used, their purpose should not be "tactical" because their consequences will not be tactical. With nuclears, it has become a war of nuclear risks and threats at the highest strategic level. It is a war of nuclear bargaining.

This is the way nuclears should be used if they must be used; ₁₂ this is therefore the way our plans should be drawn. And our requirements for nuclear weapons in Europe—numbers of weapons, their location, state of readiness, and means of delivery—should be derived from this concept of their use. ₁₃

Following upon the logic of his World Politics article, Schelling was recommending, in a paper directed to the President of the United States, that the administration use nuclear weapons after the outbreak of local war around Berlin to compel the Soviet Union to back down. As we have seen, and as Schelling’s last comment implies, this recommendation was moot for the moment for the simple reason that the hardware in Europe did not exist. Schelling’s awareness of this fact suggests, however, that he was hoping in his paper not so much to affect the way that Kennedy dealt with Berlin, as to use the crisis as a means of persuading the administration to prepare for future crises his way.

McNamara’s "No-Cities" Doctrine

As we have seen, McNamara was one of the administration’s original proponents of flexible response. ₁₄ Though he had backed away from it out of necessity during the Berlin crisis, he had not stopped believing that it was necessary to continue the fight to eliminate Eisenhower’s all-or-nothing military policy and develop a serious strategy of limited nuclear warfare. A tremendous bureaucratic fighter, by early 1962 McNamara had developed and was beginning to implement a new American military policy, which he called the "no-cities" doctrine. This new policy stipulated that the United States, for the first time since 1956, would officially seek to limit and control a war with the Soviet Union. ₁₅

In February, McNamara introduced this strategy at a speech in Chicago, in which he argued specifically that the gradual escalation incorporated in his new strategy was not meant so much to limit destruction per se as to threaten further war as a means of getting the Soviets to back down. "We may seek to terminate a war on favorable terms by using our forces as a bargaining weapon—by threatening further attack," McNamara argued in February. "Our new policy gives us the flexibility to choose among several operational plans."He then formally proposed the new strategy at a secret NATO meeting in early May. "[B]asic military strategy in general nuclear war," McNamara stated, "should be approached in much the same way that more conventional military operations have been regarded in the past." Berlin, he told his European colleagues, had revealed the bankruptcy of previous NATO all-or-nothing doctrine: Eisenhower’s strategy "has serious limitations for the purpose of deterrence and for the conduct of general nuclear war."

To avoid the immediate holocaust that that previous strategy insured, the United States would urge that its allies accept a new nuclear strategy which provided NATO decision-makers with a "variety of strategic choices," and the ability to wage a "controlled and flexible nuclear response." A possible element of this strategy, McNamara added, would be a "small, demonstrative use of nuclear weapons." ₁₆

McNamara announced the policy publicly in his famous commencement address at the University of Michigan in June 1962. "The very strength and nature of the Alliance make it possible to retain, even in the face of a massive surprise attack," McNamara said, "sufficient reserve striking power to destroy an enemy society if driven to it." McNamara elaborated this point in a part of the speech not commonly quoted:

We are convinced that a general nuclear war target system is indivisible, and if, despite all our efforts, nuclear war should occur, our best hope lies in conducting a centrally controlled campaign against all of the enemy’s vital nuclear capabilities, while retaining reserve forces, all centrally controlled. ₁₇

We know that the same forces which are targeted on ourselves are also targeted on our allies. Our own strategic retaliatory forces are prepared to respond against these forces, wherever they are and whatever their targets. This mission is assigned not only in fulfillment of our treaty commitments but also because the character of nuclear war compels it. ₁₈

A basic national security policy draft, submitted to the NSC just after McNamara’s speech, reiterated this kind of thinking: in the event of general war, the draft’s authors argued, the United States should seek to reduce Soviet offensive capability, retain strategic forces for "possible selective use," and facilitate "the conduct of negotiations designed to bring the war to an end on terms which are consistent with U.S. interests.. . . " ₁₉

As Lawrence Freedman has noted, there were some differences between Schelling’s strategy of compellence and McNamara’s no-cities doctrine. ₂₀ Most signficant of these was McNamara’s argument that the limited warfare NATO would wage, to defeat the enemy locally and remind it of possible destruction to come, should mix conventional and nuclear forces. This departed from Schelling’s contention that only nuclear weapons would raise the specter of war and create the climate of fear needed to compel the communists to back down.

In the bigger picture this was really a minor distinction. Since 1957 (longer than that, unofficially) it had been the policy of the United States that in a confrontation with the Soviet Union and its allies over major Cold War stakes, the West would wage general nuclear war once the conflict moved beyond minor military hostilities. A general nuclear war, moreover, would be an inherently unlimitable and uncontrollable phenomenon, something that America and its European allies would unleash only in a retaliatory, spasmodic way to begin and end World War III. Almost single-handedly, and in the space of only a few months, McNamara had discontinued this policy. In its place he had put forth a doctrine based upon the idea that the United States could win a limited war with the communist bloc by using selective military attacks to escalate the war on American terms and to remind the Russians of the destruction to follow if they did not retreat. ₂₁ McNamara was imposing upon the military policy of the West Schelling’s thesis that a direct war with the Soviet Union, even a nuclear war, could be controlled, restrained, manipulated, and eventually won. This was the aspect of the no-cities doctrine that mattered most.

The Cuban Missile Crisis: Eisenhower Wins

As Desmond Ball argues, McNamara gave up on the no-cities strategy by late 1962 for several reasons: unfavorable American reaction to the idea of a first nuclear strike; the prospect of a diversifying Soviet nuclear capability; the uniform rejection of it by NATO allies; and the exploitation of it by the services to secure unnecessary hardware. ₂₂

But McNamara did not simply give up on the no-cities strategy; he abandoned the entire notion of winnable nuclear war. And this abandonment manifested itself not just in a subsequent revision of American nuclear strategy but in the rest of McNamara’s public life. While his reasons for shelving the no-cities doctrine itself may have been due to traditional strategic and bureaucratic reasons, his sudden rejection of Schelling’s essential thesis came from a more visceral experience he underwent in late October 1962.

During the last two weeks of October 1962 the United States and the Soviet Union confronted each other over nuclear missiles that Khrushchev had deployed in Cuba. ₂₃ The Cuban Missile Crisis was the last, most dramatic, and probably most dangerous episode of the entire 1958–1962 crisis period. Nothing remotely like it has happened since.

In the early stages of the crisis Kennedy seriously considered the option of launching a conventional air strike against the missiles. According to the Schelling vision of strategy, this would have been a proper move, and after the Soviet response the U.S. would commence its "war of nuclear bargaining and demonstration." The objective: humiliation and defeat of the Soviet Union, and the liberation of Cuba. Here was a textbook opportunity to take advantage of the new doctrine. The weapons were now available, the military ready to go.

What instead happened? In exchange for a Soviet promise to remove the missiles, simply to restore the status quo ante, Kennedy agreed publicly that the United States would not invade Cuba. Through back channels, Kennedy agreed as well to remove medium-range nuclear missiles from Turkey. Had the back channels failed to appease Khrushchev the American administration was ready to ask the Secretary-General of the United Nations to call for a Soviet&-;American agreement to remove the missiles from Turkey and Cuba. ₂₄ Kennedy apparently had decided as well that if the Soviet Union had attacked the missiles in Turkey that the United States would not respond.

Where was McNamara in all of this? Firmly in the camp of appeasement. McNamara argued against the proposed air strikes, encouraged compromise, and endorsed the plan to remove the missiles from Turkey. ₂₅ During a meeting early in the crisis, McNamara expressed a willingness to go considerably further:

President Kennedy then said that he thought at some point Khrushchev would say that if we made a move against Cuba, he would take Berlin. McNamara surmised perhaps that was the price we must pay and perhaps we’d lose Berlin anyway. ₂₆

McNamara had worked diligently in late 1961 and early 1962 to overhaul the basic military policy of the West. His objective was to put the United States and its allies into a position to act confidently during a Cold War crisis, using its new limited weaponry and strategies to raise the specter of imminent nuclear war and compel the Soviets to back down. "Our new policy," as he said earlier in 1962, "gives us the flexibility to choose among several operational plans." But in confronting the possibility of war with the Soviet Union over the missiles in Cuba, a crisis that one might have thought to be tailor-made for the new policy, McNamara seems not to have been particularly interested in deciding which of his "several operational plans" he might want to use. His primary concern was to find some way to avoid war. He feared that any military conflict with the Russians could conceivably lead to all-out nuclear war, a prospect he found so unacceptable that he was ready to abandon Berlin to prevent it. In the heat of the Cuban crisis the idea of escalating the conflict in a virile way, of using, for example, a demonstrative nuclear attack to force the Soviets to think about all-out war, must have struck McNamara as obscene. When the tangible prospect of nuclear war materialized, McNamara shrank away from the Schelling strategy as if it were a deadly snake. Instead, he embraced Eisenhower’s strategy of evading nuclear war in its pure form.

Indeed, after October 1962 McNamara himself embarked upon his own campaign of evading nuclear war. He and Kennedy got rid of the no-cities doctrine and replaced it with a policy of assured destruction that was simply a new version of NSC–5707/8. During the Vietnam War McNamara refused to consider using nuclear weapons, preferring instead to preside over a horrible and losing conventional war that destroyed his career and perhaps his life. Since the 1960s McNamara has publicly and resolutely advocated the policy of minimum deterrence, writing at length in a 1982 Foreign Affairs article (co-written with McGeorge Bundy, George Kennan, and Gerard Smith) and in a book that nothing good whatsoever can come from a nuclear war. ₂₇ He renounced the Schelling strategy in as thorough a manner as it is possible to do. ₂₈

There is more to McNamara’s repudiation of the Schelling thesis than his conciliatory role in the Cuban Missile Crisis, or his reversion to an all-or-nothing American military policy. By rejecting Schelling McNamara was changing sides in a struggle between two basic ideas about war and politics, a struggle that had been going on well before Hiroshima and the ICBM.

On one side in this long contest of ideas was an understanding of warfare and human relations deriving from the Enlightenment. According to this view, the dilemma of the thermonuclear age could be solved, as could all human dilemmas, by reason, seriousness of purpose, scientific mastery. Of course the new weapons were immensely destructive: an all-out war would be a catastrophe of unprecedented proportions. But this fact made it only more important to stress that these were still only weapons, inanimate objects, tools in the hands of their human masters, and there was therefore no essential reason why capable leaders could not wage a war with them in a rational way. To deny this was to let the specter of war win, to admit that the human race had reached its limits of mastery and control. Not only was this an admission of human limitations: it was also a recipe for all-out thermonuclear destruction when the day came that the Soviet Union and the United States finally went to war.

On the other side of this struggle stands a more pessimistic school of thought. It takes from the sorry history of modern politics and warfare the conclusion that fear and panic routinely overtake reason and equanimity during times of violent crisis. At the outset of a war between nations possessing many thermonuclear weapons, this sort of fear would rise to unimaginable levels, driving from every decisionmaker all conceivable courses of action other than the course of all-out, spasmodic attack. Tremendous, overwhelming—and probably accurate—fears that the other side would not respect any limits to the war and was about to launch everything it had would create a climate of frenzy and terror, even bloodlust. In such circumstances a lone call for a limited strike, or a nuclear "demonstration" would appear ridiculous, pathetic, entirely outside the pale. This climate of fear and panic had dictated how major wars had been started and waged in the past: the stakes of thermonuclear war would only intensify it.

The struggle between these two ideas is not a new one. Clausewitz identified it after the Napoleonic wars, and it has informed the works of theorists, historians, strategists, and poets trying to understand the ferocious and inhuman nature of modern war. ₂₉

The difference, of course, between the debate over the nature of thermonuclear war and previous such debates is that it remains hypothetical. And unless we want to bet everything on the optimists, that is what it will always be. For if we lost this bet, and the pessimists turned out to be right, a thermonuclear war will have destroyed the human race, and along with it things like discourse and memory. The debate would remain forever unresolved, because those pessimists proven right, along with those optimists proven wrong, would all be dead.

Eisenhower’s genius was in recognizing that his position was indeed unverifiable. To prevent a thermonuclear war he would never be able to point to one and demonstrate its absolute nature to everyone’s satisfaction. Instead, he would have to portray it as such with historical comparisons and imagery, in the hope of persuading his advisers, allies, and the wider Cold War world to accept his vision of what a war between the United States and the Soviet Union would really entail. He accomplished this by forcing his negative views about human nature and war onto United States security policy, quashing with bureaucratic and rhetorical mastery more optimistic beliefs and strategies about the waging of thermonuclear war. The irony, of course, was that his reason for doing this in the first place was not based upon a negative view of humanity at all.

Endnotes

Note 1: In making the following stab at understanding Schelling I have been helped by reading Trachtenberg, History and Strategy, chapter one; Freedman, The Evolution of Nuclear Strategy, chapter fourteen; and Robert Jervis, The Meaning of the Nuclear Revolution (Ithaca, 1989), chapters 1–3. Back.

Note 2: Thomas Schelling, Strategy of Conflict (Cambridge: Harvard University Press, 1960). Back.

Note 3: Schelling (pp. 189–90) sees Eisenhower’s cavalier references to nuclear war during his first term in this light. Back.

Note 4: Schelling is ambiguous about whether the initial nuclear strike should happen after a war has begun or while a crisis is still peaceful. Back.

Note 5: Ibid., pp. 190–93. Back.

Note 6: Ibid., pp. 195–99. Back.

Note 7: Ibid., p. 196, emphasis in the original. Back.

Note 8: Ibid., pp. 239–42. Back.

Note 9: Schelling, "Nuclear Strategy in Europe," World Politics 14 (April 1962), p. 424. Back.

Note 10: Ibid., pp. 425, 427–28. Back.

Note 11: Indeed, much of this paper is reprinted verbatim in the 1962 article. Back.

Note 12: Again, Schelling here obfuscates an otherwise clear argument by introducing the conditional "if they must be used." Back.

Note 13: July 5, 1961 Paper Prepared by Thomas C. Schelling, FRUS 1961–63: 14, pp. 170–71. Emphasis in original. Back.

Note 14: This brief overview of McNamara’s new strategy relies upon Desmond Ball, "The Development of the SIOP, 1960–63," in Ball and Jeffrey Richelson, eds., Strategic Nuclear Targeting (Ithaca: Cornell University Press, 1986); Sagan, Moving Targets, pp. 28–31; and William Kaufmann, The McNamara Strategy (New York: Harper and Row, 1964). Ball writes that as soon as he took office "McNamara was an immediate convert" to the no–cities strategy. See "Development of the SIOP," p. 62. Back.

Note 15: Ball, "Development of the SIOP," pp. 63–64. Back.

Note 16: Quoted in Kaufmann, The McNamara Strategy, pp. 74–75. See also May 5, 1962 Address by McNamara at Ministerial Meeting of North Atlantic Council, FRUS 8 (1961–63), pp. 275–79. Back.

Note 17: Cf. Schelling: "Control over nuclear weapons in Europe must be tight and centralized." (Schelling paper, op. cit., p. 172). Back.

Note 18: Quoted in Kaufmann, The McNamara Strategy, pp. 116–17, emphasis added. Back.

Note 19: June 22, 1962 Basic National Security Policy draft, FRUS 8 (1961–63): 314. Back.

Note 20: Freedman, The Evolution of Nuclear Strategy, p. 236. Back.

Note 21: See Ball, "Development of the SIOP," p. 57. Back.

Note 22: Ibid., pp. 67–68. Back.

Note 23: Obviously this is not the place for any kind of detailed account of the Cuban crisis. I owe my understanding of McNamara’s role in it to my friend and colleague Phil Nash. Back.

Note 24: For an authoritative account of the deal on the Turkish missiles, see Philip Nash, The Other Missiles of October: Eisenhower, Kennedy, and the Jupiters, 1957–1963 (Chapel Hill, 1997), chapter 5. Back.

Note 25: See Richard New Lebow and Janice Gross Stein, We All Lost the Cold War (Princeton, 1994), pp. 119–123. Back.

Note 26: October 18, 1962 memorandum for file of meeting with the President, in Mary S. McAuliffe, ed., CIA Documents on the Intelligence agency 1992, p. 185. Again I thank Phil Nash for sharing this document. Back.

Note 27: Bundy, Kennan, McNamara, and Smith, "Nuclear Weapons and the Atlantic Alliance," Foreign Affairs 60 (Spring 1982); McNamara, Blundering into Disaster: Surviving the First Century of the Nuclear Age (New York, 1986). Back.

Note 28: Schelling has also written more recently on the calamity of nuclear war, but he has failed to reconcile this position with his arguments of the early 1960s. For example, he ends his article on proliferation, in the inaugural issue of International Security, with a section entitled "And Now for the Good News," in which he notes how shocking it would have been for someone to predict in 1951 that weapons would spread to many nations, that the superpowers and other nuclear states would experience numerous showdowns, that the number of nuclear bombs produced by these states would reach the tens of thousands, and that yet "not a single one of these bombs will go off accidentally." But that is not the only shocking news: "And most important of all," Schelling concludes his article, "there will have been no nuclear weapons fired in warfare." Schelling, "Who will have the Bomb?" International Security 1 (Summer 1976), pp. 90–91. He reiterates this argument in Foreign Affairs, stating that "I like the notion that. . . civilization depends on the avoidance of military aggression that could escalate to nuclear war." Schelling, "What went wrong with arms control?" Foreign Affairs 64 (Winter 1985–86), p. 233. Suffice it to say that in 1960 he did not like that notion. Back.

Note 29: See Paul Fussell, editor’s introduction to The Norton Book of Modern War (New York: Norton, 1991), pp. 18–25. Back.