REVISITING THE ROLE OF NUCLEAR WEAPONS A Nuclear-Weapon-Free World: Making Steady Progress from Vision to Action 22 nd United Nations Conference on Disarmament Issues Saitama, Japan, 25 27 August 2010 James M. Acton Carnegie Endowment for International Peace Debates about the abolition of nuclear weapons are often polarized between two extremes. Many abolitionists argue that nuclear weapons do not deter that they contribute nothing to the security of their possessors. At the recent Non-Proliferation Treaty Review Conference, the Non-Aligned Movement argued that doctrines of nuclear deterrence neither bring about peace nor international security. At the 2005 Review Conference, a leading non-nuclear-weapon state this one in NATO stated that nuclear weapons must not be seen as an attractive option that will provide additional security. Naturally, if you believe that nuclear weapons contribute nothing to security, then the risk of a nuclear detonation whether by accident or by design constitutes a simply unassailable case for abolition, and in short order. At the other end of the spectrum, opponents of abolition I ll call them realists since that is generally how they self identify don t just argue that nuclear weapons enhance security, they claim that it is simply unimaginable that any other means of managing an anarchic world, even approaching the Bomb in its effectiveness, could ever be developed. For realists, abolition however realistically it is approached is a fool s errand. Both positions, for me at least, are misguided. It has always struck me as odd that some abolitionists can simultaneously argue that the detonation of a single nuclear weapon on a single city would be an indescribable human catastrophe, which of course it would be, and that nuclear weapons do not deter. The effects of nuclear weapons may be indescribable but they are not unimaginable. In fact, it is precisely because we have seen those effects that any state contemplating aggression against a possessor of nuclear weapons must think long and hard about the consequences of its actions. I don t want to overstate the value of nuclear deterrence. It is not infallible, nor is it risk free, nor is it relevant to many of the security challenges faced by states in the contemporary world. I merely contend that, in some cases, nuclear weapons reduce the chance of large-scale aggression and we should not want to abandon them without an alternative. As the International Commission for Non-proliferation and Disarmament, sponsored by the Japanese and Australian governments, recognized it is hard to contest 1
the almost universally held view that the absence of great power conflict since 1945 must be at least in part attributed to the fear of nuclear war. Realists make the error of overstating the effectiveness of nuclear deterrence. Nuclear deterrence can fail. Moreover, for nuclear weapons to deter there must be some probability that, if deterrence were to fail, they would actually be used. The result, to quote McGeorge Bundy, would probably be a disaster beyond history. What concerns me most is the potential for miscalculation: the possibility that a leader who wrongly believed that a nuclear attack was imminent might launch a nuclear counterattack. Recognition that nuclear deterrence can fail that it carries real risks leads, or at least ought to lead, to the realization that the pursuit of an alternative is worthwhile and, moreover, that this alternative need not be perfect. I can only see one alternative to nuclear deterrence: conventional power balancing backed by international law, robustly enforced through collective action authorized by the United Nations. For abolition to actually occur, this system would have to be effective enough that the states that possess nuclear weapons today, and their allies, would be at least as secure without them. Many of you will think that this criterion is unfair. You re right. The so-called principle of undiminished security for all does prioritize the interests of states that possess nuclear weapons. Nevertheless, as George Perkovich and I wrote a couple of years ago: Reality has put the states that possess nuclear weapons in the driver s seat, albeit while travelling a road that may lead over a cliff. They cannot be forced to eliminate these weapons. They will choose to do so only if they judge that they will not become more endangered as a result. Moreover, if the elimination of nuclear weapons were to increase the chance of a conventional war between China and the United States, or India and Pakistan, or Russia and NATO, the consequences would be felt worldwide; the suffering would not be limited to the belligerents. Ensuring that the elimination of nuclear weapons does not lead to the thawing of frozen conflicts is in all our interests. Creating a rules-based alternative to deterrence is daunting. And, unfortunately, the record of the international community in enforcing the rules has been unimpressive. Take Iran, for instance. The United Nations failed to act when Iraq invaded Iran in 1980. It failed to act when Iraq used chemical weapons against Iran a few years later (or, for that matter, when Iran used such weapons against Iraq, albeit on a smaller scale). The problem quite simply was that too many states viewed meaningful action as being contrary to their interests. More recently, of course, Iran has been the violator, not the victim. It has been found in non-compliance with its nuclear safeguards obligations for serious and sustained violations stretching back over 20 years. It refuses to cooperate fully with the International Atomic Energy Agency to resolve key questions related to weaponization. Yet, once again the 2
international community s response has been anaemic. Once again, too many states do not view enforcing the rules as being in their interests. So, when it comes to enforcement, it remains to be seen whether states are willing to make the necessary sacrifices for international law to become a viable alternative to nuclear deterrence. Over the last year, I have been working on the other end of the problem: the question of how far along the road to abolition it is possible to go, without serious progress towards a much stronger system of international law. Are deep reductions in nuclear weapon numbers possible? Or, would they increase the chance of conflict? My conclusion is a positive. Deep reductions are both possible and desirable. I won t try to run through all the analysis right now. I ll just shamelessly encourage you to read my work as it is published over the next six months or so. Today, I want to focus on one of the bureaucratic hurdles to low numbers and the vital role that US allies have in overcoming it. Shortly before New START negotiations, an initial study by US Strategic Command concluded that reductions much below the United States current level of 1,700 deployed strategic warheads would be impossible under existing guidance. The last three words under existing guidance are the important ones here. STRATCOM was not saying that the United States necessarily needed 1,700 warheads to deter threats that determination is fundamentally a political one left to civilians. Rather, they were saying that they could not implement their guidance with significantly fewer than 1,700 warheads. In the end this did not become an issue in New START. Russia was not willing to go lower than 1,550 deployed strategic warheads and STRATCOM decided, with a bit more study, that this number would be sufficient to implement their guidance. Future reductions will not be so straightforward. A US administration that wants to go to 200 or 500 or even 1,000 deployed strategic warheads is going to have to write new guidance. This will certainly open up a debate about US deterrence requirements and will almost definitely lead to vitriolic accusations of playing fast and loose with national security. But, it won t just be the security of the United States that the administration will be accused of undermining; it will be the security of US allies too. In Washington, US allies are invoked in opposition to every arms control agreement and every progressive change to US posture or doctrine. Some former US officials argue against ratification of the Comprehensive Test Ban Treaty on the grounds that, if the United States gave up the right to test, its allies would lose confidence in the effectiveness of the US arsenal. (I hasten to add that no one in the current administration makes this argument.) Senator Jim DeMint, one of the few Senators so far who has actually announced an intention to vote against New START, has said that because of US allies it is absurd and 3
dangerous that America should seek parity with Russia when it comes to nuclear weapons. And, you can be virtually certain that US allies will be invoked in opposition to deep cuts. So, what will the argument be? How will critics argue that the new guidance required to facilitate deep reductions would undermine the security of US allies? Answering this question requires an understanding of current US war planning and why it requires so many warheads. I should acknowledge right away that because nuclear panning guidance is so highly classified, my analysis is necessarily speculative. However, there are some interesting and revealing nuggets available. Consider, for instance, the definition of deterrence given by General Kevin Chilton, the current commander of STRATCOM: The postulate that a decision maker s behavior can be affected by holding at risk something he or she values, and by denying the benefits he or she seeks. The more valued the thing held at risk, and the more the decision maker believes we can deny what is sought, the more the behavior can be influenced. Chilton s emphasis on denial is both honest and interesting. Denial was an important element of US nuclear planning during the Cold War. A conventionally inferior NATO planned to use nuclear weapons to help defeat a conventional invasion of Europe and hence deny the Soviet Union its supposed war aims. Today, of course, the Soviet Union no longer exists and the United States is no longer conventionally inferior. This raises the question: what denial objectives are nuclear weapons still required for? The answer, I believe, is damage limitation, that is, the attempt, in the event of a nuclear war, to destroy as many of the enemy s nuclear forces and as much of its nuclear infrastructure as possible, before they can be used against the United States. Of the various pieces of evidence that suggest that damage limitation is still an important element of US nuclear planning, I d point in particular to the US military s deterrence manual, Deterrence Operations: Joint Operating Concept, which is available online and explicitly endorses damage limitation. The important point here is that if damage limitation is your goal, you probably do need a large number of nuclear weapons. By this point, some of you might be thinking that I am starting to sound a bit like General Turdigson in Dr. Strangelove. So let me be clear. I don t believe that meaningful damage limitation in a nuclear war is possible. I don t think the US President believes that meaningful damage limitation is possible. Oddly enough, I don t even think that many in the US military believe that meaningful damage limitation is possible they are just implementing the guidance they have been given. In spite of all of this, however, damage limitation appears to be part of US planning guidance. The United States would have to abandon this concept to reach lower numbers. Here is where the US allies come in. 4
Historically, extended deterrence has always been the major justification for damage limitation. The argument runs something like this: the United States makes nuclear threats on behalf of allies. But, if it ever executed these threats, the US could find itself embroiled in a nuclear war in which tens of millions of Americans were killed. As a result, American military planners worry that extended deterrence is not credible. They fear that an adversary would attack their allies and simply assume that the United States would not sacrifice its security to defend its allies. Damage limitation was developed in the frozen depths of the Cold War as a solution to the credibility problem. The theory is that if the United States could significantly reduce the devastation it would suffer in a nuclear war, its adversaries might believe that US would actually defend its allies if push ever came to shove. As I said just now, I think this argument is flawed because meaningful damage limitation is simply not possible. But, to be frank, I m a think-tank analyst and the US government probably doesn t care all that much what I think. Much more important is what the governments of US allies think. Their voices carry weight. And, they could have a significant, positive effect on the pace of disarmament efforts if they were to say to Washington, both publicly and privately, that, in their view, deep reductions would not undermine extended deterrence. Let me give you one example of how US allies really can make a difference. During the recent US Nuclear Posture Review, the fate of a weapons system called TLAM/N (the nuclear-armed Tomahawk Land Attack Missile) proved particularly contentious. Advocates of this system acknowledged that the United States did not need it for selfdefence. Instead, they argued that this system was required for extended deterrence, in particular, to assure Japan. For much of 2009, the fate of this system appeared genuinely in the balance. Then, on 24 December last year, Japanese Foreign Minister Okada wrote to US Secretary of State Clinton and made it clear that, while Japan valued extended deterrence, it had no opinion about TLAM/N. Three and half months later, the Nuclear Posture Review Report announced the demise of TLAM/N. We will not know for certain how much Foreign Minister Okada s letter contributed to this decision until years from now when key documents are declassified. The preliminary evidence, however, suggests that it played a significant role. The Foreign Minister s letter was a very real example of non-nuclear-weapon state action on Article VI of the Non-Proliferation Treaty. The first seven words of that article Each of the Parties to the Treaty often get ignored. Yet, they are important. They make it clear that disarmament is a shared responsibility. To be clear, shared responsibility is not the same thing as equal responsibility. Nuclearweapon states have a clear obligation to lead and to shoulder most of the burden. But, non- 5
nuclear-weapon states also have a role to play. Part of their role a pivotally important part is to ensure that they are not a reason for inaction. Thank you. 6