The nuclear non-proliferation regime that helped contain the spread of nuclear weapons is crumbling. India and Pakistan became full-fledged members of the nuclear club in the spring of 1998 when they tested atomic weapons (India had previously conducted a “peaceful nuclear explosion” in 1974). North Korea detonated a nuclear device last October, and policymakers in Washington and Europe worry that Iran is using its civilian nuclear program to provide cover for a weapons program. Meanwhile, nuclear deterrence theory, the intellectual compass that guided our thinking about nuclear weapons and helped us navigate the Cold War, is condemned at best as “old thinking” and at worst as dangerously misleading.
This article addresses three questions: What is nuclear deterrence theory? Is nuclear deterrence theory still relevant, or is it a dangerous relic of the Cold War? If it is still relevant, then what does it tell us about some of the potential dangers posed by the spread of nuclear weapons? I argue that nuclear deterrence theory remains relevant, and it implies that the risk of states using nuclear weapons depends on two key factors being present at the same time. The first is a severe conflict of interest; the second is uncertainty about the balance of resolve between the states, i.e., about which state is willing to run the higher risk in order to prevail. The importance of the first factor is widely appreciated and understood, but the importance of the second is not.
What is Nuclear Deterrence Theory?
Nuclear deterrence theory can be seen as an effort to understand how political conflicts of interest play out in the shadow of nuclear weapons. The theory begins by recognizing that nuclear weapons do not eliminate political conflicts of interest. The United States and Soviet Union remained at odds after they acquired nuclear weapons; India and Pakistan continue to be divided over Kashmir.
Rather than eliminating political conflicts, nuclear weapons change the strategic setting in which those conflicts play out. Once two political adversaries acquire nuclear weapons and the means to deliver them, neither state can prevent its adversary from imposing horrendous destruction on it, should an adversary decide to do so. Thus, any crisis between these states poses a risk of spinning out of control and, ultimately, leaving each state far worse off than it would have been had it given in or acquiesced to its adversary’s original demands at the outset. For example, the original stakes of a crisis between the United States and the Soviet Union would hardly matter had that crisis ended in a massive nuclear exchange which left both societies in ruins. Likewise, the terms of a Kashmir settlement would seem unimportant after a nuclear exchange between India and Pakistan.
How does the risk of catastrophic escalation affect the way that political conflicts play out? How and to what extent can states exert coercive pressure on each other in order to further their interests in the shadow of such risk – be those interests to protect what they already have or to acquire more? A handful of scholars including Bernard Brodie, Hermann Kahn, Thomas Schelling, Glenn Snyder, and Albert Wohlstetter tried to answer these questions in the early years of the Cold War. Nuclear deterrence theory grew out of these efforts. Marc Trachtenberg’s “Strategic Thought in America, 1952-1966” in his History and Strategy provides an excellent overview of these efforts. Some of the ideas discussed here are developed at greater length in my article in International Security titled “Nuclear Deterrence Theory, Nuclear Proliferation, and National Missile Defense.”
Thomas Schelling saw that the risk of events spinning out of control transformed crises between nuclear-armed states into a kind of brinkmanship. The actions conventionally-armed states take during a crisis, e.g., mobilizing their forces and moving ships, exert coercive pressure primarily through their effects on the military balance between the states. In contrast, the actions nuclear-armed states take during a crisis, e.g., imposing a blockade as in the Cuban missile crisis or conducting a limited military probe, exert coercive power by raising the risk that the crisis will go out of control and end disastrously.
In a brinkmanship crisis, each state tries to induce the other to back down by taking steps that raise the risk that events will go out of control. This does not mean that states bid up the risk eagerly or enthusiastically. Although states may be very reticent to raise the risk, they may be still more reluctant to back down. Throughout a brinkmanship crisis, each state faces a series of terrible choices. It can quit, or it can decide to hang on a little longer and accept a somewhat greater risk, in the hope that its adversary will find the situation too dangerous and back down. If neither state backs down, the crisis goes on with each state effectively bidding up the risk until one eventually finds the risk too high and backs down, or until events actually do spiral out of control.
Brinkmanship crises, therefore, are contests of resolve, not of relative military strength. Resolve in turn depends on what is at stake in the crisis. The more a state has to gain by prevailing or the more it stands to lose if it backs down, the greater the risk of disaster it is willing to run in order to prevail, and the stronger its resolve.
A key point follows. What makes crises dangerous is the uncertainty about the balance of resolve, i.e., about which state is willing to run larger risks. Even if the stakes for both states were large enough so that each would pay a high price if it backed down, there would still be no escalation and no crisis if there were no uncertainty about the balance of resolve. The less resolute state would acquiesce to the more resolute state because the it knows that no matter how hard it pushes a crisis, the more resolute state would be willing to escalate still further. Indeed if the more resolute state has to, it is willing to push the crisis to the point where the risk exceeds the less resolute state’s resolve.
Escalation occurs only if the balance of resolve is uncertain. If each state believes that it is likely to be more resolute than its adversary, then each may escalate in the expectation that the other will back down. As the crisis continues and neither state backs down, each learns that the other is more resolute than initially believed. Eventually one state concludes that the risk is too high and the chances that the other will back down are too low to warrant further escalation. At that point that state backs down and the crisis ends – assuming, of course, that events have not already gone out of control.
Brinkmanship can be seen as a variant of an all-pay, second-price auction in which bids are measured in terms of the risk that events will go out of control. The political conflict of interest underlying the crisis defines the payoffs to winning or losing the auction. In order to win, each state bids up the risk until one of the states finds the risk too high and quits. The state that prevails is the one that is willing to hang on longer, that is, makes the highest bid. But the amount of risk that the states actually run during the crisis is determined by the state that backs down first. Thus, the “price” that each state must pay – that is, the risk of disaster each must run – is determined not by the highest bid but by the second-highest bid.
In sum, according to nuclear deterrence theory, nuclear weapons change the strategic setting in which political conflicts play out by transforming crises from contests of relative military strength into contests of resolve. We can then explain the dynamics of the latter in terms of the logic of brinkmanship. States exert coercive pressure on each other during a crisis by taking steps that raise the risk that the crisis will go out of control. States make “threats that leave something to chance,” as Schelling called them.
Is Nuclear Deterrence Theory A Cold-War Relic?
Before trying to think through the consequences of nuclear proliferation, it is important to distinguish between two questions. First, does nuclear deterrence theory still provide a useful guide to the logic of escalation and the way that crises between nuclear states play out? If the answer to this question is “no,” then we have to go back to the blackboard to see if we can develop a new post-Cold War theory of nuclear crises and escalation. If the answer is “yes,” then we can use the theory to help answer a second question. Can the United States successfully deter adversaries from challenging US interests? In other words, does deterrence work from Washington’s perspective?
Not surprisingly, policymakers and analysts in Washington tend to focus on the second question. But then they often mistakenly treat a negative answer to the second as a negative answer to the first. That is, if the spread of nuclear weapons means that Washington is less likely to get its way in future crises, then there must be something wrong with our thinking about the way crises work. Clearly this is not so. The balance of resolve will not always favor the United States, and deterrence will always seem to fail from the losing side’s perspective.
When the Bush Administration took office in 2001, many both inside and outside the government seemed to believe that nuclear deterrence theory was at best an irrelevant artifact of Cold War thinking and at worst dangerously misleading. Indeed, this was a major argument in favor of developing a national missile defense system. The United States, according to those critics, could no longer rely on deterring potential adversaries. As Secretary of Defense Donald Rumsfeld argued shortly after taking office during testimony to the Senate Armed Services Committee in 2001, failing to build a missile defense system “could give rogue states the power to hold our people hostage to nuclear blackmail – in an effort to prevent us from projecting force to stop aggression.”
Rather than demonstrate that deterrence theory is irrelevant or dangerously misleading, Secretary Rumsfeld’s comments, along with those of many other observers, suggest that we can use deterrence theory to think through the consequences of nuclear proliferation. The essence of Rumsfeld’s concern is that the spread of nuclear weapons (in the absence of an effective missile defense) will likely lead to situations in which a nuclear-armed adversary would be able to deter the United States from projecting its conventional military power in ways that it otherwise would not. Secretary of Defense Robert Gates echoed this reasoning during his confirmation hearings when he expressed his concerns about the effects of North Korea and Iran’s nuclear programs; “one of the reasons why Iran is determined to have nuclear weapons is that they see how complicated it is for us to try and deal with a North Korea that has nuclear weapons.” That is, the United States may fail to deter an opponent just as deterrence theory predicts, precisely because the balance of resolve may often favor the adversary. In brief, nuclear deterrence theory remains a useful conceptual framework and analytic tool after the Cold War.
What does nuclear deterrence theory say about the consequences of nuclear proliferation? We highlight three implications here. First, it is clear why a militarily weak state facing a much stronger adversary has an incentive to acquire nuclear weapons. A weak state is at a severe disadvantage in any confrontation with a much stronger adversary as long as the ultima ratio is a test of conventional forces. By transforming the confrontation into a contest of resolve, nuclear weapons shift the advantage to the state with the greatest resolve which will often be the conventionally weak state. Indeed, this is especially likely to be the case if the stronger threatens the weaker with regime change. Most regimes are willing to run greater risks to stay in power than others are willing to run in order to topple them.
A second implication follows immediately. Insofar as the United States will be the overwhelming conventional power in any dispute for many years into the future, the spread of nuclear weapons will, at least to some degree, reduce the United States’ ability to use its military power to influence events.
The third implication centers on the risk that nuclear weapons will be used. If nuclear deterrence theory continues to explain the dynamics of escalation, and if the United States and the Soviet Union made it through the Cold War without using nuclear weapons despite profound conflicts of interest, then why should India and Pakistan or any two other states be any more likely to use these weapons than were the United States and the Soviet Union? The spread of nuclear weapons may be undesirable according to this view, but the risk of nuclear weapons use would be very small.
The reasoning underlying this sanguine view is wrong. There is a critical distinction between a theory and the predictions derived from it. The latter depend not only on the theory but on the initial conditions as well. Predicting who wins an auction and at what price depends not only on the rules of the auction but also on how much the bidders value what is being auctioned off and on their uncertainty about the other bidders’ valuations. Even though the brinkmanship model of nuclear crises remains relevant and a useful explanatory tool, this does not mean that the nuclear relationship between India and Pakistan or other nuclear-armed adversaries will be as stable as it arguably was between the United States and the Soviet Union. Who prevails and at what risk in brinkmanship depends on the balance of resolve and the clarity of that balance. Should these factors differ in post-Cold War conflicts from what they were during the Cold War, then the Cold War experience will provide a poor guide to the future.
If the balance of resolve between new nuclear states over the issues that divide them were clear and remained so, the spread of nuclear weapons to those countries would not significantly raise the risk of nuclear use. But such clarity is unlikely. The balance of resolve between new nuclear states over politically contentious issues is likely to be fraught with uncertainty. New nuclear adversaries are quite likely to be very unsure about where the bright lines are, and this makes the spread of nuclear weapons dangerous.
Indeed, a better reading of the history of the Cold War reinforces this point. The relatively sanguine view that some have about the inherent stability of nuclear deterrence during this period is largely based on the second half of the Cold War following the Cuban missile crisis. International politics seemed much more dangerous and fraught with peril during the first part of the Cold War when the balance of resolve over various issues was much less clear and the United States and Soviet Union were figuring out which was willing to risk more over what. As Richard Betts observes in an article published by MIT in The Coming Crises, “The early phase of the Cold War, before the crises over Berlin and Cuba worked out the limits to probes and provocations, is a less reassuring model. Only with hindsight is it easy to assume that because the superpowers did not go over the edge, it was foreordained by deterrence that they could not.”
The presence of nuclear weapons changes the incentives states face and in this way alter the strategic setting in which states find themselves. Nuclear deterrence theory is an effort to explain how political conflicts play out in this arena, and the theory remains a useful compass with which to navigate the future. But the answers it provides are not comforting. The balance of resolve between new nuclear states over critical issues is likely to be opaque, and this increases the risk of escalation and use of nuclear weapon.