Lessons of the Next Nuclear War

It doesn't take a superpower to pose a nuclear threat. A small, poor country with a few nuclear explosives and the means to deliver them could wreak terrible damage on the United States. Even if never used, a handful of nuclear weapons merely in the possession of an unfriendly country could change a regional balance of power against the United States. Thus, the major military danger now facing the United States in the post-Soviet world is not a particular country but rather a trend: nuclear proliferation.

Because they enhance national power, nuclear weapons are potentially attractive to a wide variety of countries. Yet relatively few have these weapons. The five permanent members of the U.N. Security Council -- the United States, Russia, Great Britain, France, and China -- all do. Several others either have or are very close to having operational nuclear weapons. The number, however, is far smaller than expected in the early stages of the nuclear age. In 1963 President John F. Kennedy predicted that 15 to 20 countries would have nuclear arms by 1975. Overall, Cold War efforts to prevent the spread of nuclear weapons have been successful.

The international spotlight falls on that effort in April, when the fifth Nonproliferation Treaty review conference convenes at the United Nations to reconsider, revise, and extend the treaty. The NPT, which came into force in 1970, now has 168 adherents. The treaty is useful; its extension, highly desirable. The course of nuclear nonproliferation in the post-Cold War era, however, will depend less on what happens at the United Nations in 1995 than in Washington thereafter. The main obstacle to the spread of nuclear weapons is not the NPT but the United States, nor is nonproliferation a single issue: it is composed of three separate problems. Each is now more complicated and urgent than in the past because the end of the Cold War has either weakened or removed the principal restraints on both the demand for and the supply of nuclear weapons. Three different types of states are candidates for nuclear armaments, and three different American policies will be required to discourage or thwart their nuclear ambitions.

The countries in the first and most important category are not ordinarily considered part of the proliferation problem, yet their acquisition of nuclear weapons would have the most powerful impact on international policies. They are the allies. Germany and Japan forswore nuclear weapons during the Cold War because they received security guarantees from the United States. Whether they continue as nonnuclear states depends on whether those guarantees continue.

The second group of would-be nuclear states are the orphans. They feel seriously threatened but lack the nuclear protection the allies have enjoyed. None has become a full-fledged nuclear power, but each is close. The orphans, particularly Pakistan, Israel, and Ukraine, are the objects of a different American policy -- diplomatic efforts to end the conflicts that have made nuclear armaments attractive to them.

The third category includes those countries at the center of concern about nuclear proliferation. The rogues -- Iraq and North Korea most prominent among them -- are openly hostile to the United States and are seeking or have sought nuclear weapons. Thwarting them will require strengthening the restrictions on bomb-related material, which the United States was instrumental in creating and enforcing and that the collapse of the Soviet Union has weakened. But these efforts may not suffice. The prevention of proliferation may ultimately require destroying those states' nuclear programs by force. Here again the chief responsibility will fall to the United States. If the nuclear future of the allies depends on American alliance commitments, and that of the orphans on American diplomacy, the nuclear aspirations of the rogues may become one of the chief objects of American military policy.

THE ALLIES

Why are nuclear weapons not more widely dispersed? In part, deterrence, a guiding concept of the nuclear powers' military strategies for most of the Cold War, has kept proliferation at bay. While the two strongest powers deterred each other from using nuclear weapons, other pairs of states deterred each other from acquiring such weapons. For example, India and Pakistan each has a robust nuclear weapons program. But neither state has built a bomb, in part out of concern that the other would follow.

Despite the impact of deterrence, the NPT has not been irrelevant. The organizations and policies that monitor and attempt to restrict the worldwide supply of nuclear materials contribute to nonproliferation. The treaty clarifies and publicizes which states have the bomb and which do not. But for a number of nonnuclear signatories to the NPT, adherence to the treaty was the consequence of a decision to abjure the bomb, the cause of which lay elsewhere. What was the basis for this widespread decision?

One possible basis, which has come to be known as the "nuclear taboo," is the feelings of horror that nuclear weapons arouse. This distaste is certainly widely felt but like the NPT is a doubtful explanation for the degree to which the world has been spared the spread of nuclear weapons. In the first place, the taboo is against using, not acquiring, these weapons. Hiroshima, not the Soviet-American arms race, evokes horror. Moreover, like all taboos, this one will be violated under necessity. Individuals will eat forbidden foods, even one another, if the alternative is starvation; nations will acquire and use forbidden weapons if they deem it necessary for survival.

Another possible basis for adhering to the NPT is the two benefits nonnuclear states are supposed to receive in return: assistance in obtaining technology for the peaceful use of nuclear energy and a good-faith effort by states with nuclear weapons to reduce their own arsenals, with the implication that, at some time in the future, all nuclear armaments will be gone.

The first of these promises is hollow. The economic benefits of nuclear technology are at best modest; its diffusion has complicated the effort to prevent nuclear proliferation; and whatever the advantages of nuclear power plants, they are irrelevant to security, the central purpose that nuclear weapons serve. The second promise is false. Steep reductions in and the ultimate abolition of the arsenals of the nuclear "haves" would promote rather than discourage nuclear proliferation. These reductions would undercut the very thing that has kept the spread of nuclear weapons in check: the guarantees of nuclear protection extended by the great powers, above all the United States.

During the Cold War those guarantees were credible in part because the United States had a large number of nuclear armaments, many of them deployed on territories distant from North America. If American nuclear weapons disappeared altogether, the guarantees that rest on them would be worthless. Those guarantees persuaded a class of countries with genuine security concerns and the capacity to equip themselves with nuclear weapons to forgo them. U.S. treaties and the local deployment of American forces kept the German Federal Republic and Japan from acquiring nuclear weapons. Both had the technical capacity. Both had a territorial dispute with the great nuclear rival of the United States, involving the Soviet occupation of the eastern part of Germany and the southern islands of the Kuril chain.

The protective American nuclear umbrella extended beyond West Germany to all NATO members. It even extended after a fashion to countries, such as Switzerland, Sweden, and Finland, that were not formally part of the North Atlantic alliance. These countries could reckon that American guarantees to their neighbors made Europe safe enough that they, too, could dispense with nuclear weapons.

In his 1872 book, The English Constitution, Walter Bagehot distinguished between the "dignified" parts of the British system of government, "which excite and preserve the reverence of the public" and the "efficient" parts, "those by which it, in fact, works and rules." For the global constitution of nonproliferation during the Cold War, the NPT constituted the dignified part, but its "efficiency" in limiting the spread of nuclear weapons stemmed principally from the American system of alliances and guarantees.

However urgent Germany's and Japan's perceived need for nuclear protection during East-West confrontation, that need has become far less pressing. It has not, however, disappeared completely. Russia and China still have nuclear weapons. Without some means of offsetting those armaments, the allies would have to consider that, at some point, a disagreement with one or the other of the two nuclear-armed giants could arise in which the possession of a nuclear arsenal would confer a decisive advantage. For the entirely legitimate and wholly defensive purpose of short-circuiting this situation, the German and Japanese governments would be moved to seek a substitute for the American nuclear guarantee. The plausible substitutes are nuclear arsenals of their own.

During the Cold War, the United States made elaborate efforts to sustain German and Japanese confidence in its security guarantee, both pledging to defend them and deploying troops and weapons on German and Japanese soil. These measures testify to the importance for the United States of thwarting proliferation, even among its closest and most important allies. Yet what Washington has worked so hard to prevent would not be a self-evidently damaging development. Japan and Germany are not only close allies, but they are democracies, with records over the last four decades -- admittedly in contrast with the previous four -- of spotlessly peaceful foreign policies.

The United States has at least one reason to welcome German and Japanese nuclear weapons. They would relieve Americans of defending two countries sufficiently wealthy and powerful to defend themselves and separated from North America by large oceans. The dangers of German and Japanese nuclear weapons, however, outweigh the advantages. Germany and Japan are large, powerful states. Their acquisition of nuclear weapons would cause more than a ripple in international politics: it would make waves. The change would usher in a multipolar nuclear order, which would supplant the more or less bipolar arrangement of the Cold War. A multipolar order would by some reckonings make the world more dangerous -- less stable, less certain, and less easily managed. In multipolar systems, none of the great powers can ever be certain who will side with whom. [1]

The emergence of Germany and Japan as major nuclear powers would likely set off a chain reaction. This emergence would signal the end of the American system of nuclear guarantees; if Germany and Japan gave up their American protection no country could count on the American shield. This development would trigger a wholesale recalculation of security requirements throughout Europe and Asia, which would have to take into account not only the retreat of American military power but also a sharp rise in the military status of Germany and Japan.

No country, least of all Germany and Japan themselves, wishes to see these countries become nuclear powers. Avoiding this eventuality will require keeping the American nuclear commitment in good working order, which in turn means perpetuating in some form NATO and the Japanese-American Treaty of Mutual Cooperation and Security. Because Germany and Japan are, at least for the moment, less threatened than during the Cold War, this commitment should require less imposing American military forces; to ward off uncertainty ought to be less expensive than to deter the Soviet Union. For this purpose, perhaps no American forces will have to be stationed on the European continent or the Japanese archipelago.

But precisely because Germany and Japan as well as the United States are less threatened, even a modest American commitment may prove difficult. The American public may be unwilling to support the deployment of major military forces to reassure prosperous democratic allies that they do not need nuclear weapons. The American people were willing, for four decades, to spend trillions for defense and deterrence; what they will be willing to spend for reassurance, and for how long, remains to be seen.

THE ORPHANS

The second cluster of nuclear aspirants were also aligned with the United States during Soviet-American confrontation but felt more threatened and less protected than the allies. These countries feared for their lives, having neighbors that they felt did not accept the legitimacy of their existence as sovereign states.

International orphans lacked as firm an American security commitment as the North Atlantic pact and the Japanese-American security treaty but were not entirely alone. They had the benefit of military assistance from and diplomatic cooperation with the United States. But they were orphaned in that they had neither formal treaties of alliance nor American troops on their territories.

Absent an American nuclear umbrella, these orphans mounted substantial nuclear weapons programs and refused to sign the NPT. While they neither formally declared themselves nuclear weapons states nor conducted official nuclear tests, by the end of the Cold War they were capable of assembling working bombs on short notice. They engaged in what might be called partial proliferation, stopping just short of full nuclear status out of deference to the nonproliferation sensibilities of the United States as well as a desire not to provoke their neighbors. While the United States disapproved of their programs, the interests it shared with these countries were such that Washington took no effective action against them.

The two prototypical orphans are Pakistan and Israel. India and Taiwan have some features in common with them. The end of the Cold War has had only a modest effect on their military policies. For another group of countries with important features in common with Israel and Pakistan, however, the end of the Cold War has had a transforming effect.

This group consists of the countries of central and Eastern Europe once dominated by the Soviet Union, as well as the former republics of the Soviet Union other than Russia. The end of the Soviet Union has, if not created, then brought into the open a demand for nuclear weapons. Former provinces of the Soviet empire have either recovered or achieved independence. They can now choose which military forces to have, and for some, nuclear weapons are a plausible choice. Their memories of Soviet and pre-Soviet Russian subjugation are fresh. The politics of the new Russia are not yet reliably democratic or respectful of its neighbors' sovereignty. To some of these new states, their recently achieved independence is bound to seem precarious, and nuclear weapons, as guarantors of independence, therefore attractive.

One such former province, Ukraine, resembles Israel and Pakistan. In Russia, Ukraine has a large, powerful nuclear-armed neighbor, some of whose leading political figures do not fully accept Ukrainian independence. Also, like Israel and Pakistan, while Ukraine is not a full-fledged nuclear weapons state it has engaged in partial proliferation. Israel and Pakistan each has something short of a fully assembled (or at least publicly acknowledged and tested) bomb. Ukraine already possesses fully assembled nuclear weapons on its territory but has at best only partial control of them; it has physical custody, but presumably lacks the actual means to launch the nuclear-tipped missiles it inherited from the Soviet period. Finally, Ukraine, like Israel and Pakistan, has looked to the United States to reinforce its security. Washington pressed Kiev to send all nuclear weapons to Russia. In return, Ukraine asked for security assurances. These Kiev has received but in the most general possible form, as a repackaging of those extended in connection with the NPT and the Final Act of the Conference on Security and Cooperation in Europe, the largely symbolic set of agreements signed by the United States and all European countries, including the Soviet Union, at Helsinki in 1975. The government of Ukraine has promised to give up all its nuclear weapons and ratify the NPT. But the process of doing so is likely to be protracted, and depending on the course of events in Russia the nuclear option may prove tempting.

Nuclear weapons would be less consequential in the hands of an orphan than in those of Germany or Japan and less dangerous than in the case of a rogue state. The United States has done less to keep nuclear weapons out of the hands of this second group of potential proliferators than the other two categories, but Washington has not resigned itself to inaction. Pakistani, Israeli, and Ukrainian nuclear weapons pose hazards that the American government has sought to avoid.

The nuclear weapons programs in South Asia raise the specter of nuclear war between India and Pakistan. The Israeli bomb makes it difficult to rally support for curbing the nuclear ambitions of other Middle Eastern countries dangerous to the West. Ukrainian nuclear weapons are undesirable because they may reestablish nuclear-armed political rivalries in Europe, where the end of the Cold War has brought a welcome end to the 40-year nuclear standoff between the United States and the Soviet Union. A Ukrainian nuclear arsenal, moreover, may trigger a decision by nearby Germany to acquire nuclear weapons.

Washington has therefore sought during the Cold War and afterward to limit and perhaps reverse these nuclear weapons programs through diplomatic efforts to reconcile the threatened countries with their adversaries. The United States has been a party to almost continuous negotiations between Israel and its Arab neighbors for two decades beginning in 1974. Peace was clearly the necessary condition for nonproliferation. Nuclear programs could be eliminated in the region only in the wake of an end to the conflict that had driven Israel down the path to an independent nuclear arsenal.

The United States adopted a similar pattern with Russia and Ukraine, serving as a mediator and broker. On January 14, 1994, the presidents of the three countries signed a tripartite accord under the terms of which Ukrainian nuclear weapons were to be moved to Russia. Ukrainian President Leonid Kravchuk subsequently announced during a state visit to the United States in November 1994 that his country would ratify the NPT. As with Arab-Israeli peacemaking, the United States promised financial assistance to Ukraine, which was implicitly contingent on Kiev's pledge to give up the nuclear weapons it had inherited from the Soviet Union.

The United States also intervened diplomatically, albeit less frequently and forcefully, in the continuing conflict between India and Pakistan. In response to reports that the two countries were close to war, in which each would likely have targeted the nuclear facilities of the other, in 1990 the Bush administration dispatched a special envoy to prevent the outbreak of hostilities, which in the end did not take place.

Thus if the nuclear future of the allies depends on whether the United States sustains its Cold War treaty commitments, the nuclear programs of the orphans will be influenced, perhaps decisively, by the success of post-Cold War American diplomacy.

THE SUPPLY SIDE

In addition to discouraging states from seeking nuclear weapons, the nonproliferation regime erects barriers to acquiring them. It addresses the supply of as well as the demand for weapons and the materials to make them.

By the terms of the NPT, the nuclear weapons states promise not to transfer the bomb to any nonnuclear states. The NPT also created the International Atomic Energy Agency (IAEA), which has a mandate to monitor nuclear power-generating facilities as well as fuel and waste storage sites of nonnuclear signatories. The purpose is to ensure these materials are not used to produce materials for nuclear weapons.

The work of the IAEA is supplemented by the efforts of an informal group of countries that manufacture the complicated and expensive machinery for uranium enrichment and plutonium reprocessing. Known as the Nuclear Suppliers Group, its aim is to restrict the commerce in nuclear fuel fabricating facilities that can also be used to make the core of a bomb. Another association outside the formal bounds of the NPT but whose purpose is consistent with the spirit of the treaty was created in 1987, the Missile Technology Control Regime. The aim of the MTCR is to restrict the diffusion of ballistic missiles, the most dangerous vehicles for the delivery of nuclear weapons.

Although the restraints on supply are necessarily multilateral, as are the measures to suppress demand, the United States has assumed the leading role in implementing them. The American government has been far more active than any other in setting the supply-side rules, trying to enforce them, and monitoring compliance.

The different parts of the supply-side regime, like all international agreements, share a common weakness: none includes any power of enforcement. If a sovereign state violates any of its norms, no international mechanism can compel compliance or mete out punishment. Nor does the regime have a perfect record in detecting violations. Iraq mounted a substantial clandestine nuclear weapons program even after signing the NPT and submitting its declared nuclear power-generating plants to IAEA inspections. Moreover, imperfect as the system of supply restraints was during the Cold War, it has come under increased pressure from two post-Cold War developments.

The first of these developments is the improvement and diffusion of technology of all kinds, including what is required to make nuclear weapons and ballistic missiles. By some estimates, 15 to 20 countries -- not all of them allies or orphans -- will have or be able to have such missiles by the end of the century. [2] The second development is the surge in the availability of nuclear materials brought about by the end of the Soviet Union. Its disintegration led to the weakening or collapse of controls on tens of thousands of weapons, many of them portable, scattered throughout the former Soviet Union; thousands of scientists and engineers, the mainstays of the Soviet military-industrial complex; and hundreds of tons of fissile material from laboratories, reactors, submarines, and weapons to be dismantled under the terms of one or another international agreement. Where the materials necessary for bomb-making were concerned, the Soviet Union was, in effect, an impregnable bank vault. When the U.S.S.R. collapsed, the vast territory on which it had stood turned into a massive bazaar.

THE ROGUES

The post-Cold War pressure on supply restraints raises the increased likelihood that a select group of countries whose possession of nuclear weapons would run radically counter to the interests of the United States will manage to get them. Most current discussion of proliferation refers to this third category of nations, the rogues, and most of the nonproliferation efforts of the American government are devoted to them. This is so because the political goals that underlie the rogues' desire for nuclear weapons are incompatible with American interests.

In some cases those goals are aggressive. Iraq and Iran want to expand their influence, which would almost certainly not be exercised in ways compatible with American interests. In other cases nuclear weapons may be sought for ostensibly defensive purposes but are unacceptable from the American point of view. North Korea is concerned with its own survival, which at least partly motivated its nuclear weapons program. But a nuclear-armed North Korea would be better able to pursue its commitment to unifying the Korean peninsula, by force if necessary, under communist control. Even if it did not try to attack, subvert, or coerce the South, a North Korea with nuclear weapons would be tempted to sell weapons-related material and equipment, perhaps even bombs, to other countries.

The number of rogue states is relatively small. The list of them invariably includes North Korea, Iran, Iraq, and sometimes Syria, Libya, and Algeria. Each is influenced by an ideology -- Marxism-Leninism, Islamic fundamentalism, or Arab socialism -- with anti-Western and anti-American features. All suffered politically and militarily from the collapse of the Soviet Union, which, by depriving them of a patron and protector, or at least a reliable supplier of weapons, gave added impetus to their nuclear ambitions. None is a full-fledged democracy.

The United States is at the forefront of efforts to prevent the rogues from acquiring nuclear weapons. The American role is based not only on a general aversion to nuclear proliferation but also on commitments, carried over from the Cold War, to the protection of South Korea and the security of Middle Eastern allies.

At least three methods of coping with the prospect of nuclear weapons in the hands of rogues are available. One is to provide countries that might become targets with the means to defend themselves. This approach, however, is technically problematic. Defending against a few missiles fired by a rogue is less taxing than the task President Ronald Reagan set for the United States in 1983 -- the protection of the continental United States against massive missile barrages that the Soviet Union was (and its Russian successor continues to be) capable of launching. Still, the capacity of current technology to produce even more modest defenses is questionable.

The second method is to offer the rogue generous inducements to give up its nuclear aspirations, the approach the United States has taken with North Korea. American policy has been to buy the North's nuclear program. Even if successful, this approach encourages blackmail. The policy is all too likely to make nuclear weapons programs seem like useful bargaining chips to be cashed for a bonanza of economic assistance from wealthy countries. [3]

Because the number of rogue states is limited, it may be possible to concentrate anti-proliferation resources on them, such as surveillance, diplomacy, and economic assistance. But because supply restraints can, sooner or later, be defeated by time, money, and determination, and because even if the world is willing to buy nuclear programs not every rogue will be willing to sell, more forceful methods may be required in order to keep nuclear weapons out of their hands.

More forceful methods have been used twice against the most determined and wealthiest of the rogues, Iraq. Both the raid on the Osiraq reactor by Israel in 1981 and Operation Desert Storm, the military campaign to evict Iraq from Kuwait a decade later, count as successful anti-proliferation measures. Each set back Iraqi President Saddam Hussein's quest for nuclear weapons. Neither, however, is an altogether promising precedent.

Would-be proliferators can insure themselves against a single crippling strike, such as that Israeli warplanes delivered, in the same way that the United States and the Soviet Union during the Cold War protected their nuclear programs from a disarming first strike. They can multiply, disperse, conceal, and shield the component parts of the program so that a single raid could not entirely destroy it. This is exactly what Iraq proceeded to do after 1981.

As for Operation Desert Storm, the anti-Saddam coalition fielded a large, sophisticated army that only the United States could provide. In future wars of nonproliferation, the United States will again be obliged to assume the leading role. The American military has the capacity to conduct such operations, but military operations require political support, and here Desert Storm may prove a weak precedent.

The crippling of Iraq's nuclear program was a byproduct of the war. The official purpose was to evict Iraq from Kuwait. Restoring Kuwait's sovereignty was, in fact, a less important American interest than keeping Saddam Hussein from controlling a large part of the Persian Gulf's oil reserves. A sovereign Kuwait was also less important than a nonnuclear Iraq. But because it constituted an unambiguous violation of international law, the invasion and occupation of Kuwait provided the basis for the political support the war attracted.

The Bush administration probably could not have rallied such extensive domestic and international support exclusively to eliminate Iraq's nuclear weapons program. A military campaign waged solely for that purpose would have been a preventive war, which has neither a basis in international law nor a well-established historical precedent.

Under present circumstances, it is difficult to envision 168 countries supporting an arrangement for preventive war as they do the NPT. The one clear instance of nuclear preemption, the Israeli raid on Osiraq, met with general condemnation. Moreover, such an arrangement would have to sanction preventive war against some countries but not others. Its targets would be Iraq and Iran but not Israel, India, Pakistan, Germany, or Japan. To compound the difficulty, such an international convention would also require defining a point in the development of nuclear weapons when a preemptive attack would be justified.

Even if the major powers could agree on the propriety of a preventive war against a dangerous nuclear weapons aspirant, one of them would have to take the lead. The logical candidate would be the United States, which would mean the American public would have to embrace preventive war as a norm, something it was never asked to do during the Cold War.

LESSONS OF THE LAST WAR

A presumption in favor of fighting when not directly attacked has historically emerged from a traumatic experience of armed conflict. Such a presumption represents the lesson of the last war, learned at great cost. The policies that could in retrospect have prevented or won that war become the axiomatic approach to the next one. American foreign policy during the Cold War was based on the lessons of World War II. The commitment to containing the Soviet Union and communism stemmed from the conviction that the failure to check Hitler's ambitions before 1939 led to a terrible and avoidable war.

The Cold War was not a comparably wrenching experience. The next nuclear war, however, the next nuclear shot fired in anger, almost surely would be. It would shock and horrify the world. To be sure, the post-Cold War era has already occasioned horror at the starvation in Somalia, the "ethnic cleansing" in Bosnia, and the slaughter followed by epidemic in Rwanda. But a nuclear explosion would have a more powerful psychological impact on Americans because they will be able to imagine this horror happening to them.

The next nuclear war is likely to reshape the nonproliferation policies as well as attitudes and policies well beyond nuclear proliferation of the international community and the United States. Just how it will do so, however, cannot be foreseen.

The next nuclear war could, like World War II, give rise to a powerful bias in favor of preemptive intervention. Once one nuclear weapons program led to nuclear war, the United States might decide to take whatever steps necessary to put a stop to similar programs in other countries. The next Hiroshima could create in American public opinion a consensus in favor of preventive war to keep the bomb out of the hands of rogue states.

The next war could, however, have the opposite effect. Its legacy could be more like that of World War I, which reinforced the historical American determination to remain aloof from the political and military quarrels of Europe. It could revive a foreign policy of isolation, although the geographic scope might differ from the past. Traditional isolation applied to Europe but not Latin America or East Asia. From the next nuclear war Americans might learn to eschew political engagement in any part of the world where nuclear weapons are located or might be used.

Either lesson if translated into policy would mark a basic shift in the American role in the world. In this sense, for American foreign policy, the most important event of the post-Cold War era has not yet taken place.

[1] The argument is John Mearsheimer's. See his "Back to the Future: Realism and the Realities of European Security," International Security, Winter 1990/1991. The argument follows Kenneth N. Waltz, Theory of International Relations, Reading: Addison-Wesley, 1979, chapter 8. But Waltz elsewhere makes a contrary argument that the spread of nuclear weapons will be a force for political stability for the same reasons that they imposed a standoff on the United States and the Soviet Union. See The Spread of Nuclear Weapons: More May Be Better, Adelphi Paper 171, London: International Institute for Strategic Studies, 1981.

[2] See for example, Peter van Ham, Managing Non-proliferation Regimes in the 1990s: Power, Politics and Policies, New York: Council on Foreign Relations Press, 1994, p. 24 and Seth Cropsey, "The Only Credible Deterrent," Foreign Affairs, March/April 1994, p. 17.

[3] In the case of North Korea this course has seemed to the American (and South Korean and Japanese) governments preferable to the alternative: a confrontation that would run the risk of war.