Military history

Chapter 13

The Rationality of Irrationality

This is a moral tract on mass murder: How to plan it, how to commit it, how to get away with it, how to justify it.

—James Newman, review of Herman Kahn, On Thermonuclear War

Despite Brodie’s nomenclature, the first atomic weapons were not “absolute.” They were in the range of other munitions (the bomb that destroyed Hiroshima was equivalent to the load of some two hundred B-29 bombers). Also, at least initially, the weapons were scarce. The key development introduced by atomic bombs was less in the scale of their destructive power than in their efficiency. By the start of the 1950s, this situation had been transformed by two related developments. The first was the breaking of the U.S. monopoly by the Soviet Union, which conducted its first atomic test in August 1949. Once two could play the nuclear game, the rules had to be changed. Thought of initiating nuclear war would henceforth be qualified by the possibility of retaliation.

The second development followed from the first. In an effort to extend its effective nuclear superiority, the United States developed thermonuclear bombs, based on the principles of nuclear fusion rather than fission. This made possible weapons with no obvious limits to their destructive potential. In 1950 the American government assumed that the introduction of thermonuclear weapons would allow the United States and its allies time to build up conventional forces to match those of the Soviet Union and its satellites.

When Dwight D. Eisenhower became president in January 1953, he saw things differently. He wanted to take advantage of American nuclear superiority while it lasted, and also reduce the burden of spending on conventional rearmament. By this time, the nuclear arsenal was becoming more plentiful and more powerful. The strategy that emerged from these considerations became known as “massive retaliation,” following a speech made by Secretary of State John Foster Dulles in January 1954, when he declared that in the future a U.S. response to aggression would be “at places and with means of our own choosing.”1

This doctrine was interpreted as threatening nuclear attack against targets in the Soviet Union and China in response to conventional aggression anywhere in the world. Massive retaliation was widely criticized for placing excessive reliance on nuclear threats, which would become less credible as Soviet nuclear strength grew. If a limited challenge developed and the United States had neglected its own conventional forces, then the choice would be between “suicide or surrender.” Dependence upon nuclear threats in the face of an opponent able to make threats of its own sparked a surge of intellectual creativity—later described as a “golden age” of strategic studies.2 At its core was the key concept deterrence, to be explored with a range of new methodologies designed to cope with the special demands of the nuclear age.


The idea that palpable strength might cause an opponent to stay his hand was hardly new. The word deterrence is based on the Latin deterre—to frighten from or away. In its contemporary use it came to reflect an instrumental sense of seeking to induce caution by threats of pain. It was possible to be deterred without being threatened, for example, one might be cautious in anticipation of how another might respond to a provocative act. As a strategy, however, deterrence involved deliberate, purposive threats. This concept developed prior to the Second World War in contemplation of strategic air raids. The presumption of civilian panic that had animated the first airpower theorists retained a powerful hold on official imaginations. The fear of the crowd led to musings on the likely anarchy that would follow sustained attacks. Although the British lacked capabilities for mass long-range attacks prior to the war, they doubted the possibility of defense and believed that only the threat of punitive attacks could hold Germany back. Ultimately, Britain had to rely on defense, which it did with unexpected success thanks to radar. The raids against Britain, and those mounted in return against Germany with even greater ferocity, resulted in terrible civilian pain but had limited political effects. Their main effect was on the ability to prosecute the war by disrupting production and fuel supplies. The surveys undertaken after the war demonstrated the modest impact of strategic bombing compared with the pre-war claims. But this did not really matter because the atom bomb pushed the dread to a new level. As Richard Overy put it, with air power the “theory had run ahead of the technology. After 1945 the two reached a fresh alignment.”3

Deterrence answered the stark exam question posed by the arrival of nuclear weapons: What role can there be for a capability that has no tactical role in stopping armies or navies but can destroy whole cities? Answers in terms of war-fighting, though explored by the Eisenhower administration, appeared distasteful; answers in terms of deterrence promised the prevention of future war. It sounded robust without being reckless. It anticipated aggression and guarded against surprise but could still be presented as essentially reactive. The difficulty was whether deterrence could be expected to hold if it was self-evidently based on a bluff. Credibility appeared to depend on a readiness to convey recklessness, illustrated by another of John Foster Dulles’s comments about the need to be ready “to go to the brink” during a crisis. Thus the residual possibility of use left a formidable imprint, precisely because it would be so catastrophic.

This reinforced the view that the main benefit of force lay in what was held in reserve. The military capacity of the West must never be used to its full extent, though for the sake of deterrence the possibility must exist. As decades passed, and the Cold War still did not turn hot, deterrence appeared to be working. At times of crisis there was a welcome caution and prudence all around. War was avoided because politicians were all too aware of the consequences of failure and the dangers of preparing to crush enemies with overwhelming force. The dread of total war influenced all considerations of the use of force, and not just those directly involving nuclear weapons. It was never possible to be sure where the first military step, however tentative, might lead.

The impossibility of a fight to the finish affected all relations between the American and Soviet blocs. There developed “a predominance of the latent over the manifest, of the oblique over the direct, of the limited over the general.”4 If, as it seemed, there was no way of getting out of the nuclear age, then deterrence made the best of a bad job. While it was often difficult to explain exactly how deterrence had worked its magic—and historians can point to some terrifying moments when catastrophe was round the corner— yet a third world war did not happen. The fact that the superpowers were alarmed by the prospect of such a war surely had something to do with its failure to materialize.

The importance of deterrence meant that considerable efforts were devoted to exploring the concept and examining its policy implications. Deterrence succeeded if nothing happened, which led to a problem when working out cause and effect. Inaction might reflect a lack of intention or an intention once present that had lapsed. Deterrence of an intended action could be due to a range of factors, including some unrelated to the deterrer’s threats and some related in ways the deterrer did not nececessarily intend. According to the most straightforward definition, that deterrence depended on convincing the target that prospective costs would outweigh prospective gains, it could be achieved by limiting gains as much as imposing costs. Preventing gain by means of a credible ability to stop aggression in its tracks became known as deterrence by denial,5 while imposing costs became deterrence by punishment. Denial was essentially another word for an effective defense, which if recognized in advance would provide a convincing argument against aggression. Thus the main conceptual challenges concerned punishment, especially the most brutal punishment of all: nuclear retaliation.

As deterrence became wedded to a foreign policy of containment, interpreted as preventing any Soviet advances, both major war and minor provocations had to be deterred, not just those directed against the United States but also those directed against allies, and even the enemy’s enemies. Herman Kahn, an early popularizer of some of the more abstruse theories of deterrence, distinguished three types: Type I involved superpower nuclear exchanges; Type II limited conventional or tactical nuclear attacks involving allies; and Type III addressed most other types of challenges.6 At each stage, the requirements in terms of political will became more demanding, especially once both sides had acquired nuclear arsenals. It was one thing to threaten nuclear retaliation to deter nuclear attack, quite another to threaten nuclear use to deter a non-nuclear event. Because it was always unlikely that the United States would be directly attacked by a major power with anything other than nuclear weapons, the most likely non-nuclear event to be deterred would be an attack on an ally. This requirement came to be known as “extended deterrence.” Because of the development of Soviet capabilities, U.S. methods of deterrence became less confident, moving from disproportionate to proportionate retaliation, from setting definite obstacles to aggression to warnings that should aggression occur the consequences could be beyond calculation, from assured and unconstrained threats of overwhelming force to a shared risk of mutual destruction.


The theorist who did more than any other to explore the conundrums of deterrence and nuclear strategy was Thomas Schelling. He was one of a number of figures in and around RAND during the 1950s—including Bernard Brodie, Albert Wohlstetter, and Herman Kahn—who despite their differences contributed to a developing framework for thinking about these weapons that acknowledged their horrific novelty yet tried to describe their strategic possibilities. At the time, Kahn—ebullient and provocative, and one possible model for Stanley Kubrik’s Dr. Strangelove—was the best known. His book On Thermonuclear War forged a link with Clausewitz, at least in its title, although his biographer doubted whether he had ever even done more than skimmed Clausewitz: “He never showed a smidgen of interest in any strategic theorist.”7 Wohlstetter described his prose as “dictated through a public address system.”8

He was nuclear strategy’s “first celebrity,” with his “physical mass and somewhat geeky cast” confirming myths that the ultimate war would be the product of the imaginings of “mad geniuses.” A mass of statistics on the likely character of nuclear war would be qualified by breezy and hardly comforting statements, such as “barring bad luck and bad management,” and led to policy options that were evaluated in terms of possible losses of units of humanity measured in the millions.9 Kahn’s fellow nuclear strategists objected as much to his showmanship and the bad name he gave their new profession as to his claims about emerging victorious from the apocalypse. An enthusiastic advocate of civil defense, Kahn was convinced that control was possible in all types of conflict, even nuclear war.

Schelling was a more substantial theorist, developing ways of thinking about conflict that illuminated nuclear issues while remaining relevant to broader strategic questions. After the mid-1960s, when he felt he had said much of what he wanted to say about nuclear matters, he turned his attention to other issues, ranging from crime to cigarette smoking, but still applied the same essential approach. His achievement was underlined by the award of a Nobel Prize in economics in 2005 for “having enhanced our understanding of conflict and cooperation through game-theory analysis.”10 Yet Schelling’s relationship to game theory was equivocal. He did not describe himself as a game theorist but rather a social scientist who used game theory on occasion. He hit upon his big idea before he came across game theory as a means through which it could be expressed. He preferred to reason through analogy in ways that purists found maddening. Schelling’s reputation depended on his gifts as a brilliant expositor who wrote with elegance and lucidity, traits for which this particular field of endeavor were not well known.11

Schelling did not claim to have achieved the “science” that had long been sought in strategy or that formal logic could in principle lead to a mathematical solution. He shared the view, growing among the operational research community, that advanced mathematics and abstract models were making their work less accessible to potential users, 12 and he always opposed the suggestion that strategy was or should be “a branch of mathematics.”13 He confessed to having learned more “from reading ancient Greek history and by looking at salesmanship than studying game theory.” The greatest achievement of game theory, as far as he was concerned, was the payoff matrix. It was extraordinarily useful to be able to put together in a matrix a “simple situation involving as few as two people and two choices.”14

His equivocation on game theory was not unique. Other nuclear strategists who worked at RAND during the 1950s tended to talk of following the “spirit” of game theory rather than its rules. In a 1949 article, Brodie referred to game theory in a footnote as a source of “mathematical systematization,” adding that “for various reasons” he did not share the authors’ “conviction that their theory could be directly and profitably applied to problems of military strategy.”15 Later, while finding its “refinements” of little use, he acknowledged the value of the “constant reminder that in war we shall be dealing with an opponent who will react to our moves and to whom we must react.”16 Few of the books on nuclear strategy made much, if any, mention of game theory. This absence was notable in a book by one of the founders of game theory, Oskar Morgenstern.17 Bruce-Briggs suggests that the close association between nuclear strategy and game theory was a consequence of the reception of Kahn’s On Thermonuclear War. Although Kahn had used neither game theory nor mathematics, he was accused of being the most extreme example of a game-theory-wielding militarist, a moniker implying great technical capacity but no moral sensibility. Schelling was also included in this category.18 Schelling observed at the time, “I don’t see that game theory is any more involved than Latin grammar or geophysics; but its quaint name makes mysterious and patronizing references to it an effective ploy.”19

Schelling had little background in military issues. Trained as an economist, he worked on the implementation of the postwar Marshall Plan for economic reconstruction in Europe. This led to his general interest in negotiations of all types, particularly the process of finding points that could support an agreed solution, possibly through tacit as much as explicit bargaining. After publishing an article demonstrating the possibility of arriving at common solutions without direct communication,20 he read Luce and Raiffa’s Games and Decisions and saw the potential of game theory.21 His interest in how “nations, people, or organizations go about committing themselves to threats and promises in bargaining positions” led him into contact with RAND in 1956 and he spent a productive year there from 1958 to 1959.22Schelling was able to test his developing theories in the company of other key thinkers from a variety of disciplines, all trying to make sense of the nuclear age. Although he was offered jobs in the Kennedy administration, he preferred to keep his independence. He did work as a consultant, however.

Many of the ideas and concepts Schelling developed along with his colleagues at RAND became familiar and entered the strategic vernacular, but it is important to note just how novel and radical they were. Critics complained, with some justice, that the methodology allowed talk about dreadful possibilities in dispassionate terms and contemplated moves that should never be countenanced by a civilized people. Their models did not offer a way of transcending the Cold War conflict and failed to accommodate the ideological and geopolitical issues. These were important limitations, but they should not hide the achievement of developing a way to think about conflict that could also accommodate cooperation.

Schelling started with the special features of a game of strategy, compared with those of chance or skill: “Each player’s best choice depends on the action he expects the other to take, which he knows depends, in turn, on the other’s expectation of his own.” Strategy was all about interdependence, “the conditioning of one’s own behavior on the behavior of others.” This could cover any social relationships in which there was a mixture of conflict and cooperation. All partnerships were to some degree precarious, just as all antagonisms were to some degree incomplete. The combination of conflict and cooperation was at the heart of the theory. It became irrelevant when there was an absence of either. Schelling noted that the theory “degenerates at one extreme if there is not scope for mutual accommodation, no common interest at all even in avoiding mutual disaster; it degenerates at the other extreme if there is no conflict at all and no problem in identifying and reaching common goals.”23

On this basis the role of force could be rethought. Traditionally it had been used by countries to take and hold what they wanted: “Forcibly a country can repel and expel, penetrate and occupy, seize, exterminate, disarm and disable, confine, deny access, and directly frustrate intrusion or attack. It can, that is, if it has enough strength. ‘Enough’ depends on how much an opponent has.”24 In setting up an alternative to brute force, Schelling made one of his most startling assertions: “In addition to weakening an enemy militarily it can cause an enemy plain suffering.” Contrary to prevailing views—and established international law, for that matter—that stressed the importance of avoiding unnecessary suffering, Schelling claimed that the ability to hurt was “among the most impressive attributes of military force.” Its value lay not in actually doing so, which would constitute a gross failure of strategy, but in what opponents might do to avoid it. So long as the violence could both be anticipated and avoided by accommodation, it had coercive value. “The power to hurt is bargaining power. To exploit it is diplomacy—vicious diplomacy, but diplomacy.” Under this proposition, strategy moved from considerations of conquest and resistance to deterrence, intimidation, blackmail, and threats.

Coercion was therefore at the heart of the theory. The hurt did not have to be nuclear. The same framework could work with less punitive forms, for example, economic sanctions. It could also take in the traditional distinction between offense and defense, although not in the sense of being able to be certain of either conquering territory or stopping any invasion at the borders. The point about coercion was that it involved influencing through threats rather than controlling the opponent’s behavior. The defensive equivalent was deterrence, persuading an enemy not to attack; the offensive equivalent was “compellence,” inducing withdrawal or acquiescence. Deterrence demanded an opponent’s inaction; compellence demanded action or ceasing adverse actions. Deterrence was about the status quo and had no obvious time limits; compellence projected forward to a new place and could be urgent. Deterrence was easier because all that was required was that an action be withheld. The target could deny that one was ever contemplated. Compliance was more conspicuous with compellence, more evidently “submission under duress,” less “capable of being rationalized as something that one was going to do anyhow.” The two could merge. Once an initial deterrent threat had failed and an opponent was acting in a hostile way, the next threat must be compellent. In a conflict in which both sides could hurt each other but neither could forcibly accomplish its purpose, and in which the balance of advantage kept changing hands, the requirement to deter and to compel could shift, depending on who was on top at any time.25

Nuclear threats had a special character. Executing them would be an unusually horrible thing to do, but a state with a nuclear monopoly might feel that it was not too difficult to gain strategic advantage by threatening others. What made the difference was that something equally horrible might come back in return. How could one benefit from threats that lacked credibility because of the risk of retaliation and could thus be exposed as bluff at the first challenge? Again, Schelling addressed this conundrum by turning traditional concepts upside down. The aim of strategy, it had been supposed, was to exert the maximum control over the course of an unfolding conflict.

Schelling asked a different question: could there be strategic advantages in accepting a loss of control? Coercive threats worked by influencing an opponent’s choices. Perhaps their choice could be made more difficult by limiting one’s own. To inject credibility into an apparently irrational stance, why not work to create an essentially irrational situation?

The idea was to shift the onus of decision onto the other side, so that it was the opponent who was obliged to choose between continued combat and backing down. Only “the enemy’s withdrawal can tranquilize the situation; otherwise it may turn out to be a contest of nerves.”26 There were precedents: the Greeks burning their bridges to show they would stand and fight against the Persians; the Spanish conqueror Cortez conspicuously burning his ships in front of the Aztecs. By removing retreat as an option, your men had no choice but to fight, while the enemy would be discouraged by an apparent display of confidence.

In the nuclear sphere, at one extreme, choice could be wholly conceded to the opponent by making the threatened action automatic, beyond recall unless stopped by an act of compliance. That was the notion of the “doomsday machine”: pass a line and nothing could be done to stop the detonation and the shared calamity. Removing all choice was unacceptable, so Schelling posed the problem in terms of progressive risk. The opponent would know that even if the threatener had second thoughts, the threat might still be implemented. This set up the possibility of a “competition in risk-taking” which could turn war into a contest of “endurance, nerve, obstinacy and pain.” This would be not quite a doomsday machine, but the threatened would know that the threat could not be wholly bluff because the threatener was not completely in control. Schelling called this “The Threat That Leaves Something to Chance.” The feature of such threats was that “though one may or may not carry them out, the final decision is not altogether under the threateneds control.”27 In his version of Clausewitz’s friction, Schelling emphasized the ubiquity of uncertainty that gave this type of threat credibility:

Violence, especially in war, is a confused and uncertain activity, highly unpredictable depending on decisions made by fallible human beings organized into imperfect governments depending on fallible communications and warning systems and on the untested performance of people and equipment. It is furthermore a hotheaded activity, in which commitments and reputations can develop a momentum of their own.28

Whereas Clausewitz saw friction as undermining all but the most dogged of strategies, Schelling saw how these uncertainties might be used creatively, if recklessly. The uncertainty would grow as a crisis turned into a limited conflict and then moved toward general war, getting out of hand “by degrees.”29 Skillful tactics would exploit this fact, not shrink from it. The assumption was that it was worth letting a “situation get somewhat out of hand” because the opponent would find such circumstances intolerable. Deterrence was possible because of a situation in which terrible things might happen (which was credible because of human irrationality) rather than a specific threat to do those things (which was incredible because of human rationality).

The potential rationality of irrationality was illustrated using the game of “chicken.” Two cars were driven toward each other by delinquent teenagers, Bill and Ben, anxious to prove their toughness. The first to swerve lost. If both swerved, nothing was gained; if neither swerved, everything was lost. If Bill swerved and Ben did not, then Bill suffered humiliation and Ben gained prestige. The matrix appeared as shown below.

Table 13. I The figures in the corners refer to values attached to alternative outcomes.





1 Swerve

2 Don’t Swerve

1 Swerve




+ 20



2 Don’t Swerve



+ 20




The minimax strategy dictated that they both swerved as the best of the worst outcomes. This represented the natural caution displayed by both sides during the Cold War. Timing, however, made a difference. Bill was prepared to swerve, but Ben swerved first. Bill won because he delayed his commitment. He kept his nerve longer. Perhaps he was confident that Ben would swerve because he knew him to be weak-willed. Suppose that Ben was aware of this impression and sought to correct it. He wanted Bill to think him reckless or even a bit crazy. A number of ruses might reinforce such an impression: swaggering, boasting, or feigning drunkenness. Irrationality became rational. If Ben could persuade Bill that he had taken leave of his senses, Ben might just prevail.

This illuminates the basic problem with this line of argument. Even if one was apparently committed to a patently irrational course to impress the opponent, a foot would still hover close to the brake pedal and hands would stay firmly on the steering wheel. What might work for two individuals was less likely to work for governments who needed to convince their own people that they knew what they were doing. Even if the internal audience was tolerant of ruses designed to suggest a loss of control, such stunts could not be a regular feature of crisis management. Whether the game involved individuals or states, it was difficult to pretend irrationality consistently, in one game after another. Like deceptive strategies, pretended irrationality would be difficult to repeat as it would affect perceptions of behavior next time round. Indeed it could be counterproductive if the other side overcompensated. The more often the game was played, the more dangerous it could become. The full importance of any strategic encounter lay not only in the implications for the matter at hand but also in the long-term impact on the relationship between the two adversaries. The results of strategies adopted in a particular game would affect their potential success if used in subsequent games. Game theory presented simultaneous decisions by players. Schelling understood that the moves often took place sequentially, so that the structure of the game changed each time.30

The mutual learning process was important to Schelling’s schema. It was almost a mission to reorient game theory to take account of the fact that “people can often concert their intentions or expectations with others if each knows that the other is trying to do the same.” Unlike theorists who argued that equilibrium points could be found using mathematics, Schelling insisted that points would suggest themselves as being obvious or natural. This required “some common language that permits them to hold discourse.” Communication of this sort between adversaries would not allow for great subtlety or sophistication, especially if the language did not emerge through formal negotiations or declarations. It could be tacit as much as explicit, dependent on prominent symbols and values in a shared culture, guided by tradition and precedence, with mutual understandings created and reinforced through deeds as much as words. It would draw “on imagination more than on logic; it could depend on analogy, precedent, accidental arrangement, symmetry, aesthetic or geometric configuration, casuistic reasoning, and who the parties are and what they know about each other.”31 Certain focal points would become salient. They would need to be simple, recognizable, and conspicuous. In Arms and Influence, Schelling gave examples of features that might suggest themselves to opposing forces who could not communicate directly:

National boundaries and rivers, shorelines, the battle line itself, even parallels of latitude, the distinction between air and ground, the distinction between nuclear fission and chemical combustion, the distinction between combat support and economic support, the distinction between combatants and non-combatants, the distinctions among nationalities.32

Once proper communication was possible, and the players could use direct speech and overt bargaining, the “pure-coordination game,” Schelling suggests, “not only ceases to be interesting but virtually ceases to be a ‘game.’ ”33 Yet for all the possibilities of indirect communication, the influence of norms and conventions of behavior, or the focal points thrown up by nature, it was hard to see how they could be more reliable than direct communications. In circumstances where the opportunities for direct communication were sparse, as was the case between the two ideological blocs for much of the Cold War, Schelling’s insight about the possibility of still finding shared focal points by indirect means was valuable. But it could not be taken too far. It did not necessarily mean that these points would be found when they were really needed. Moreover, when two sides were working with such different sets of values and beliefs, what was salient for one might not be so salient for the other. Without direct communication to verify that an agreed point had been found, it was possible to miscalulate by assuming that the other side attached the same salience to the same point or by assuming that agreement on such matters was impossible. It could seem, as Hedley Bull observed in a review of Arms and Influence, that the superpowers would be “sending and receiving messages and ironing out understandings” with “scarcely as much as a nod or a wink.”34

First and Second Strikes

Schelling argued that not only was it possible to think of nuclear strategy in terms of bargaining and coercion but it was unwise to think of it in any other way. This challenged the idea of a decisive victory directly by claiming that at least in the nuclear area it made no sense. That did not mean that there was no concept of what a decisive nuclear victory would look like. To ensure success it would have to take the form of a knockout blow that left the opponent no chance to retaliate. This was not a possibility that either side in the Cold War ever felt able to dismiss entirely. It provided part of the dynamic of the arms race between the two sides and governed calculations of risk. A “first-strike capability” came to refer to the potential ability to disarm the enemy in a surprise attack. No military operation ever conceived could be as fateful. It would be the first and only time it would be attempted and would be launched in secret, using untried weapons against a range of disparate targets in a wholly unique scenario, using equally untried defenses to catch any retaliatory weapons. Whether such a capability was in reach depended on evaluations of the developing capabilities of offensive and defensive weapons.

In a celebrated RAND study of the mid-1950s, a team led by Albert Wohlstetter demonstrated that the air bases of the U.S. Strategic Air Command could be vulnerable to a surprise attack. Retaliation from such an attack would be impossible, thereby exposing the United States and its allies to Soviet blackmail.35 This challenged the prevailing view that nuclear weapons could be used solely in “counter-value” strikes against easily targeted political and economic centers. A “counter-force” strike against military targets would potentially be strategically decisive because it would leave the opponent without any way of retaliating. If, however, the attacked nation was able to absorb an attempted first strike and retain sufficient forces to hit back then it would have a “second- strike capability.” Wohlstetter believed that this study, drawing on “the tradition of operations research and empirical systems analysis,” far more than Schelling’s musings, discovered the “vulnerabilities of strategic forces.”36

Suppose both sides had a first-strike capability. Brodie set out the alternative possibilities in a 1954 article in which he observed that in a world in which “either side can make a surprise attack on the other” it would make sense to be “trigger-happy.” As with the “American gunfighter duel, frontier style,” the “one who leads on the draw and the aim achieves a good clean win.” If neither side had the capability, however, trigger-happiness would be suicidal, and restraint only prudent.37 Depending on how the technology developed, there would either be powerful pressures to preempt at times of high political tension, which could lead to a dangerous dynamic, or there would be considerable stability, as no premium would be attached to unleashing nuclear hostilities. Thus, confidence in stability depended on expectations with regard to the opponent’s attitude and behavior. In a compelling example of his mode of analysis, Schelling described a “reciprocal fear of surprise attack” to show how an apparently stable system of deterrence could suddenly be destabilized even when there was no “fundamental” basis for either side to strike first: “A modest temptation on each side to sneak in a first blow—a temptation too small by itself to motivate an attack—might become compounded through a process of interacting expectations, with additional motive for attack being produced by successive cycles of ‘He thinks we think he thinks we think . . . he thinks we think he’ll attack; so he thinks we shall; so he will; so we must.’ ” To reduce any chance of such thoughts, nuclear systems should be unequivocally geared to a second strike: relatively invulnerable and relatively inaccurate. In practice this meant that cities would be threatened, not weapons.

The logic became even more uncomfortable and paradoxical. Nothing should be done to diminish the murderous consequences of a nuclear war because nothing should be done to encourage any thought that it was worth starting. “A weapon that can hurt only people, and cannot possibly damage the other side’s striking force,” Schelling explained, “is profoundly defensive; it provides its possessor no incentive to strike first.” The danger lay with weapons intended to “seek out the enemy’s missiles and bombers—that can exploit the advantage of striking first and consequently provide a temptation to do so.”38 The aim was to stabilize the U.S.-Soviet nuclear relationship. On this basis, Schelling noted, missile-carrying submarines were admirable for second-strike purposes. They were extremely hard to find and destroy at sea, but it was also hard (at the time he was writing) to use them for accurate strikes against the enemy forces. For this reason, Schelling argued that Americans should not want a monopoly on these submarines, for if they “have either no intention or no political capacity for a first strike it would usually be helpful if the enemy were confidently assured of this.”

If such reasoning led to conclusions that seemed quite bizarre to the military mind, the same was true for those on the other side of the argument, campaigning for radical measures of disarmament. The more weapons one had, the more difficult it was for the adversary to wipe them out in a surprise attack. An agreement designed to stabilize the nuclear relationship would be easier to maintain at higher rather than lower levels, for it would be much more difficult to prepare to cheat by hiding extra missiles if the starting numbers were high.39 Neither the military nor the disarmers were at all sure that their activities should be mutually reinforcing. The term “arms control” was in fact coined in the 1950s precisely to identify forms of mutual understanding that were compatible with the new imperatives of military strategy.40 It meant that the military had to get used to the idea that while opposing the enemy’s force they must also collaborate, implicitly if not explicitly, in avoiding the kinds of crises in which withdrawal is intolerable for both sides, in avoiding false alarms and mistaken intentions, and in providing—along with its deterrent threat of resistance or retaliation in the event of unacceptable challenges—reassurance that restraint on the part of potential enemies will be matched by restraint of our own.41

In line with Schelling’s general interest in how productive agreements could be reached without direct communication, arms control could involve “induced or reciprocated ‘self-control,’ whether the inducements include negotiated treaties or just informal understandings and reciprocated restraints.”42

In any event, technological developments supported the second strike. Attempts to develop effective defenses against nuclear attack proved futile. By the mid-1960s, fears eased of a technological arms race that might encourage either side to unleash a surprise attack. For the foreseeable future, each side could eliminate the other as a modern industrial state. Robert McNamara, as secretary of defense, argued that so long as the two superpowers had confidence in their capacity for mutual assured destruction—an ability to impose “unacceptable damage” defined as 25 percent of population and 50 percent of industry—the relationship between the two would be stable. These levels, it should be noted, reflected less a judgment about the tolerances of modern societies and more the point at which extra explosions would result in diminishing marginal returns measured by new damage and casualties, the point at which—to use Winston Churchill’s vivid phrase—“all you are going to do is make the rubble bounce.”

If serious fighting did begin, the incentives would shift. Assuming no rush into nuclear exchanges, it would still be possible to shape the development of the conflict by drawing on the potential of what could happen. So long as cities were spared there was some hope of establishing a new bargain, even in the midst of war. But once cities were destroyed there was nothing else to lose. Attacks on cities would be “a massive and modern version of an ancient institution: the exchange of hostages.” Keeping something of value vulnerable was a way of enforcing good behavior.43 Like Clausewitz, Schelling saw how raw and angry passions could also undermine restraint.

The process by which a conflict intensified and became more dangerous came to be described as “escalation.” This word (not one that Schelling favored) came into vogue to describe a tragic process as a limited war became total. It was based on the metaphor of the moving staircase that once started could not be stopped, however much the original decision might be regretted. The term—initially interchangeable with words like explosion, eruption, and 4rigger—was first used to challenge the idea of a limited nuclear war. Henry Kissinger, for example, defined escalation in 1960 as “the addition of increments of power until limited war insensibly merges into all-out war.”44 Schelling was aware of opportunities to use the process for bargaining purposes as well as how these would become fewer as control over events was progressively lost. To get the aggressor to stop and preferably go back to the starting point, relinquishing captured territory, the threat would have to be credible and serious, yet the circumstances would be those in which a previous deterrent threat had not been taken sufficiently seriously. The function of limited war therefore had to be understood less in terms of ensuring that war stayed limited and more in terms of posing “the deliberate risk of all-out war” and keeping the risk of escalation “within moderate limits above zero.”45 The role of the first nuclear exchanges would not be “solely or mainly to redress a balance on the battlefield” but primarily “to make the war too painful or too dangerous to continue.”46

Schelling developed his ideas before it became apparent that the superpower confrontation would become dominated by thoughts of mutual destruction. The possibilities that Schelling explored did not materialize because the consequences of any nuclear use would be so horrendous that they did not lend themselves to subtle and clever maneuvers. Crisis behavior became careful, cautious, and circumspect. So much of Schelling’s framework can be considered in retrospect as a mind-clearing exercise, exploring a range of possibilities that never moved beyond speculative hypotheses but at least demonstrated the inadequacy of conventional strategic thought. During the 1950s, with memories of past lurches to war still strong, few were confident that a third world war could be indefinitely postponed. The exploration of the logic of deterrence, and why it made sense to accept this logic rather than attempt to circumvent it, was important enough to justify the effort.

Existential Deterrence

It might have even been possible to imagine a major war between the two superpowers in which nuclear weapons were not used, although few would have been prepared to rely on continuing restraint. The core problem which niggled away at America’s strategists was that of extended deterrence, the commitment to bring nuclear means to the aid of non-nuclear allies. Once a stalemate was reached, it seemed reckless to consider nuclear war on behalf of allies. But the Europeans were assumed to have insufficient conventional strength to hold back a determined assault by the Soviet-led Warsaw Pact. If Europe was not to be overrun, there at least had to be a possibility that the United States would initiate nuclear war. Were it not for this basic political commitment, reflecting a vital interest, there would have been no need to worry about Schelling’s “threats that left something to chance.” This idea was best captured by so-called tactical nuclear weapons whose military value could never be properly explained but conveyed the risk that once entangled in a land war in Europe, they could trigger nuclear war in a way that was beyond rational consideration.

By the start of the 1960s, there was a developing view in the United States that the best way to ease this problem was to reduce dependence upon nuclear threats by increasing conventional forces—to create deterrence by denial. The difficulty was that a conventional build-up would be expensive and such obvious efforts to reduce their nuclear liabilities would suggest to European minds that the Americans might consider European security a less than vital interest. Behind this there was a disjunction between the formal strategic analysis emerging from the American think tanks and the politics of Europe, divided between two hostile ideological blocs, yet enjoying some stability. The Europeans did not view the continent as being on the edge of war. They understood that nuclear threats might not be credible, but deterrence could still work because of the residual possibility that in the irrational, intense circumstances of another European war, nuclear weapons might be used. The possibility did not have to be very large for political leaders to decide to stick with a manageable status quo. On this basis, the key to deterrence was alliance, the close link between American power—including its nuclear arsenal—and European security. The threat to deterrence was anything which undermined that link.

Here was a clash of strategic frameworks. One was top down, the classical, grand strategic perspective which focused on the formidable reasons not to risk war when it was possible to imagine catastrophe for all involved. The other was bottom up, an operational analysis which considered where the advantage might lie in a conflict, should the politicians decide the stakes were worth the fight. This pointed to an inability to match Soviet conventional strength. Only increasingly incredible nuclear threats were available should Moscow take advantage of this vulnerability, raising the possibility that it just might.

This issue came to a head in 1961 when newly elected President Kennedy was faced with a major challenge over the status of Berlin. The old German capital was firmly in communist East Germany, yet as part of the postwar settlement it had been divided into two. West Berlin, connected uneasily to West Germany, offered an easy route out for East Germans seeking to escape communism. This was a major irritant to Moscow. There were threats that summer of a Soviet move to cut off West Berlin and bring it into communist control. As the city was indefensible by conventional means, any effort to prevent this carried a risk of nuclear war. Ultimately this risk was sufficient for the communists to limit the provocation, and they built a wall which divided Berlin and so kept their people hemmed in.

During that summer’s crisis, a paper by Schelling setting out his ideas on limited nuclear conflict was passed to Kennedy. The paper stressed the importance of heightening the risk to the enemy rather than making a futile bid to win a decisive victory. “We should plan for a war of nerve, of demonstration and of bargaining, not of tactical target destruction.” This paper apparently made a “deep impression” on Kennedy. Schelling had been talking with McGeorge Bundy, Kennedy’s national security advisor. They shared a concern that the military seemed unable to think through the “hideous jump between conventional warfare and a single massive all-out blast.”47 His main contribution to policy at this time, however, was to help set up a “crisis game” that sought to simulate, as closely as possible, the confused and stressful conditions decision-makers might face and the questions to be addressed should tensions over Berlin escalate. Schelling’s game explored how the Berlin crisis might unfold. This had the advantage of being a contained scenario, in which the dimensions and core views of the protagonists were known. In September 1961, a number of rounds of this game took place in Washington designed to impress on the participants the “bargaining aspect of a military crisis.” The games forced senior policymakers—military and civilian—to work out responses to various contingencies. The conclusions, which had an effect on both official thought and Schelling’s future theorizing, emphasized the pressure of events. It was far harder to communicate efficiently than was often assumed, as the enemy saw only the actions and not the intent behind them, and there would be far less time for diplomacy to operate than had been hoped.

Yet it also became extraordinarily difficult in the games to trigger a large-scale conventional war, much less a nuclear conflict. According to Alan Ferguson, Schelling’s collaborator, “our inability to get a fight started” was the “single most striking result.”48 The games also highlighted a problem with Berlin: “Whoever it is who has to initiate the action that neither side wants is the side that is deterred. In a fragile situation, good strategy involves leaving the overt act up to the other side.”49 The game therefore gave little support to the idea that any nuclear use, even for signaling purposes, provided realistic options for NATO in the event that the Berlin crisis worsened but it did reinforce the size of the gap between conventional and nuclear war. Reporting back to Kennedy, an aide highlighted the difficulty of using “military power flexibly and effectively for tactical purposes in the conduct of the day to day political struggle with the Soviet Union.”50

The next year Kennedy faced an even greater crisis prompted by the discovery that the Soviet Union was building missile sites in Cuba. Many of the conversations of leading players on the American side were recorded as they debated potential moves and counter-moves. The president spent much of the crisis trying to determine the effect of a particular course of action on Moscow, and to do this he tried to put himself in Nikita Khrushchev’s shoes. In doing so, Kennedy supposed that the Soviet leader was cast in the same mould, responding to the same stimuli and facing the same sort of pressures from his own hard-liners, finding it as hard as Kennedy would to back down from public commitments. He was fearful that a missile strike against Cuba would lead to a Soviet attack against Turkey, where American missiles of equivalent range and targeting were based; a blockade of Cuba would revive the issue of a blockade of West Berlin.

Kennedy formed an in-group of key officials—known as ExComm— to debate the alternatives. One option was to launch air strikes against the offending bases in Cuba to take them out before they became operational. This option had to take into account whether it might be possible to get away with a small “surgical” strike or whether the risk could only be removed by continual and heavy strikes, possibly followed by an invasion. Another option involved a more gradualist approach, demonstrating resolve through a blockade to prevent military equipment getting to Cuba. ExComm’s decisions in part depended on the practicalities: the air force’s confidence in their ability to find and destroy the bases, the quality of the air defenses they would face, and the risk that some of the weapons were already operational. When confronting the possibility of air strikes, especially without warning, a number of members of ExComm felt uneasy. The United States had, after all, been the victim of a surprise air strike on December 7, 1941. The president’s speechwriter, Ted Sorenson, noted that he had no trouble writing the speech announcing the blockade but great difficulty writing one to report an air strike. The other advantage of the blockade was it did not preclude tougher action if it failed to produce immediate results. It kept options open and the opponent guessing.

There was still anxiety over whether the blockade could be enforced. Robert Kennedy wrote of his brother as he waited to see how Soviet ships would respond.

I think these few minutes were the time of greatest concern for the President. Was the world on the brink of a holocaust? Was it our error? A mistake? Was there something further that should have been done? Or not done? His hand went up to his face and covered his mouth. He opened and closed his fist. His face seemed drawn, his eyes pained, almost gray.

From the other side, consider a long, impassioned private letter to Kennedy that arrived two days later from the Soviet leader Nikita Khrushchev:

If people do not show wisdom, then in the final analysis they will come to a clash, like blind moles, and then reciprocal extermination will begin. . . . we and you ought not to pull on the ends of the rope in which you have tied the knot of war, because the more the two of us pull, the tighter that knot will be tied. And a moment may come when that knot will be tied so tight that even he who tied it will not have the strength to untie it, and then it will be necessary to cut it, and what that would mean is not for me to explain to you, because you yourself understand perfectly of what terrible forces our countries dispose.51

On Saturday, October 27, 1962, tensions were at their height after mixed messages from Moscow—one conciliatory, the other tough—and the added tension of an American spy plane getting shot down over Cuba. The presumed response was to retaliate against Soviet surface-to-air missile sites in Cuba. While this might be held back, at some point surveillance would need to start again, putting U.S. aircraft at risk and making a response unavoidable. Robert McNamara set out a possible script. If surveillance aircraft were fired upon, the United States would have to respond. There would be losses of aircraft, and “we’ll be shooting up Cuba quite a bit.” This was not a position that could be sustained for very long. “So we must be prepared to attack Cuba—quickly.” This was going to require an “all-out attack” involving air strikes, with “sorties every day thereafter, and I personally believe that this is almost certain to lead to an invasion, I won’t say certain to, but almost certain to lead to an invasion.”

The next stage assumed a tit-for-tat reprisal from Khrushchev: “If we do this, and leave those missiles in Turkey the Soviet Union may , and I think probably will, attack the Turkish missiles.” This led inexorably on to the proposition that “if the Soviet Union attacks the Turkish missiles, we must respond. We cannot allow a Soviet attack on the—on the Jupiter missiles in Turkey without a military response by NATO.” He continued

Now the minimum military response by NATO to a Soviet attack on the Turkish Jupiter missiles would be a response with conventional weapons by NATO forces in Turkey, that is to say Turkish and U.S. aircraft, against Soviet warships and/or naval bases in the Black Sea area. Now that to me is the absolute minimum, and I would say that it is damned dangerous to—to have had a Soviet attack on Turkey and a NATO response on the Soviet Union.52

McNamara took this sufficiently seriously, even though he was assuming in this script that his own government would take choices that he clearly thought unwise, to suspect that a nuclear war would be set in motion the next day. In reality, neither Kennedy nor Khrushchev were prepared to contemplate such a calamity and they found a way to draw back from the brink: withdrawal of Soviet missiles in return for an American promise not to invade Cuba. During the crisis there were many examples of how poorly the two sides understood each other, yet on the most fundamental issue they had a shared view. They were determined to avoid a nuclear tragedy.

Although the outcome of the missile crisis was shaped by a shared fear of nuclear war, one conclusion drawn was that with a clear head and a strong will, such crises were possible to manage. In particular, the successful outcome was used to challenge the notion of escalation. This had been not so much a strategy as something to be avoided. After the crisis, the metaphor was challenged for failing to recognize the potential for graduated moves, especially during the early stages of a conflict before serious battle had been joined. Albert and Roberta Wohlstetter observed that “there are down- escalators as well as up-escalators, and there are landings between escalators where one can decide to get off or get on, to go up or down, or to stay there; or to take the stairs. Just where automaticity or irreversibility takes over is an uncertain but vital matter, and that is one of the reasons a decision maker may want to take a breath at a landing to consider next steps.”53

Herman Kahn sought to show that even once nuclear exchanges had begun there were ways of conducting operations that might keep the pressure on the other side while avoiding Armageddon. He saw escalation as a dragon to be slain: not so much a phenomenon operating independent of human action but a possible product of inadequate intellectual and physical preparation. He introduced the idea that escalation might be a deliberate act. The noun acquired a verb when he referred to “people who wish to escalate a little themselves, but somehow feel that the other side would not be willing to go one step further.”54 Escalation was transformed from a hopelessly unruly process to one that might be tamed and possibly manipulated. In his 1965 book On Escalation, he introduced the “escalation ladder” with sixteen thresholds and forty-four steps. For most, the striking feature of the book was the possibility of anyone coming up with almost thirty distinct ways of using nuclear weapons following their first use at rung fifteen.55 The escalation up the ladder concluded, when all semblance of control had been lost, with a “wargasm.” Kahn declared himself innocent of the Freudian connotations. Luigi Nono, the radical Italian composer, used Kahn’s ladder as the theme for a musical composition dedicated to the National Liberation Front of Vietnam, which moved from (1) “crisi manifesta” to (44) “spasmo o guerra insensata.”56

McGeorge Bundy, former national security advisor to both Kennedy and Johnson, also reacted strongly to analyses such as these. He concluded that the arms race had become largely irrelevant in terms of actual international political behavior. Once both sides acquired thermonuclear weapons, there was a stalemate. The “certain prospect of retaliation” meant that there had been “literally no prospect at all that any sane political authority, in either the United States or the Soviet Union, would consciously choose to start a nuclear war.” He wrote of the “enormous gulf between what political leaders really think about nuclear weapons and what is assumed in complex calculations of relative ‘advantage’ in simulated strategic warfare.” In the think tanks, levels of “acceptable” damage could involve the loss of tens of millions of lives, so that the “loss of dozens of great cities is somehow a real choice for sane men.” For Bundy, in “the real world of real political leaders” a “decision that would bring even one hydrogen bomb on one city of one’s own country would be recognized in advance as a catastrophic blunder; ten bombs on ten cities would be a disaster beyond history; and a hundred bombs on a hundred cities are unthinkable.”57

Bundy’s belief that the more esoteric strategic debates had lost touch with reality led in 1983 to the argument that because both sides would be able to retaliate with thermonuclear weapons even after the “strongest possible preemptive attack,” there was in place a form of deterrence which he described as “existential,” resting on “uncertainty about what could happen.”58 This removed the strategic effect from particular weapons programs, preparations for employment, or doctrinal pronouncements. So long as any superpower war carried a high risk of utter calamity, it was best not to take risks. This notion proved to be extremely seductive not only because of its intuitive plausibility but because it solved all those perplexing problems of nuclear policy by rendering them virtually irrelevant, so long as they did not stray too far into the realms of recklessness and foolishness. Although in policymaking circles it was still extremely difficult to think of ways to assess the size and composition of nuclear arsenals except by reference to the assumed requirements of actual exchanges, as evidenced in numerous debates in Washington over new weapons systems, these debates eventually acquired a routine quality. The scenarios became drained of credibility. Nuclear deterrence worked for the United States because it warned of the severe dangers of disrupting the status quo. The sense of danger depended not on the rationality of a nuclear response but on the residual doubt that once the passions of war had been unleashed no reliance could be placed on the irrationality of nuclear response.

If you find an error or have any questions, please email us at Thank you!