The success of even an economic cripple like North Korea in building nuclear weapons demonstrates that the Clinton administration's nonproliferation policy is doomed. The policy ignores the obvious: the spread of nuclear weapons and the means to deliver them has already advanced so far that the important question is no longer how to stop their proliferation, but rather how to prevent them from being used.

Three options exist for the United States in dealing with emerging nuclear states: to persist in its current policy, which uncertainly presumes that America will extend its nuclear arsenal to regional allies and retaliate in kind against any nuclear attack; to withdraw its nuclear protection and ignore the dangers of regional nuclear conflicts as being of limited strategic interest; or to try to deter a regional nuclear aggressor through America's new conventional weapon technologies.

Only the third option offers a credible strategy that adheres to American interests. Since the end of the Cold War, the idea that the United States will use nuclear weapons to defend allies in peripheral regions has lost credibility and cannot protect either the United States or its allies from attacks by rogue states. Yet this is a danger, and a responsibility, that the United States cannot shun. As long as it remains the world's greatest economic and military power, America will be a prime target of ambitious tyrants with malignant designs.

Crossing the nuclear threshold no longer raises the prospect of engulfing the world in fire as it did when the U.S. and Soviet arsenals faced one another. But a lesser catastrophe, such as the obliteration of a single population center, is now far more likely. This new risk arises both because of the spread of nuclear weapons and also the vulnerability of powerful nations to attack by smaller ones. U.S. policymakers must assume that at least one of the 20 countries now possessing or trying to build nuclear weapons will use them. When this happens, the international strategic landscape will be irrevocably transformed.

The United States must shift its policy focus from nonproliferation to deterring or, if necessary, punishing a lesser nations use of nuclear weapons. But in a proliferated world, it is important that this aim be achieved without using the very force that America rightfully seeks to contain. While the means of deterrence must change, the idea of it does not. Foremost, America must have a credible strategy—one that guarantees that the United States will inflict so much pain on an enemy that the certain disadvantages offering a single nuclear weapon against America or its allies far outweigh the possible benefits.


The weapons of America's still formidable nuclear arsenal are a poor deterrent against the often dictatorial regimes of lesser nations for three reasons.

First is the unwillingness to use them. The utter destruction of, for example, Iraqi or North Korean civilian population centers as a response to a one or two weapon attack on their most likely American targets—military bases or personnel stationed on foreign soil—would be extremely unjust. Neither Iraqi nor North Korean civilians have any effective control over their governments. Nor did Soviet citizens. But even America's Cold War strategy of mutually assured destruction threatened first the Soviet Unions ability to launch a nuclear attack against the United States and, only subsequently, Soviet industrial and population centers. A policy of unilaterally assured destruction is not only unworthy of the United States but also falls beyond the pale what America's public conscience can safely guarantee. It would wreak such disproportionate vengeance on both the intended target and its neighbors downwind that the strategy is barely a credible deterrent.

Second, the dictators of many smaller nations have not always worried about sacrificing the lives of their citizens, especially when the regional advantages of frustrating or humiliating the United States seem within grasp. Thus even a U.S. doctrine of proportionate response to a nuclear attack—while its power to sober an enemy cannot be denied—would provide an extremely slender defense. An "irrational" despot might well calculate the costs of a nuclear exchange differently from strategists in the West—not so much in terms of lives or absolute devastation but rather in terms of what affects or furthers his own grip on power. A greater deterrent would threaten those targets he values most—the sources of his power, such as military infrastructure, bases and troops. Such a deterrent strategy could be carried out far more economically and humanely without employing nuclear weapons.

Finally, a nuclear attack on America or one of its allies must not cause the United States to subvert the very aim of its broader policy—to prevent the nuclear threshold from being crossed. A conventional response—one still capable of delivering devastating punishment—would uphold a valuable international stricture on the use of nuclear weapons and go further toward deterring a nuclear aggressor because of the certainty of the retaliation. Such a deterrent policy could help prevent a widening gyre of international chaos.

The United States should not rule out the use of nuclear weapons entirely. Uncertainty about American nuclear retaliation still forces an enemy to think seriously. For example, a newly belligerent Russia that again threatens America or its European allies would provide a large and well-armed nuclear opponent and perhaps require a return to a strategy of mutually assured destruction. But the U.S. strategic arsenal is as unqualified to defend this nation against lesser, rogue states as it is to protect America's allies against nuclear attack. A cold-blooded despot could challenge both America's reluctance to legitimize nuclear force by responding in kind and its unwillingness to retaliate against innocent civilian populations.


American policymakers need to understand the limitations of their current deterrence approach, which, like the weapons that accompany it, is based on outdated Cold War assumptions. The sad fact is that extended deterrence—the ability of the U.S. nuclear force to protect its allies—is dead. Washington's public hand-wringing about using nuclear weapons to defend states hosting U.S. forces has already undermined the credibility of the deterrent. This equivocation came both during Iraq's seizure of Kuwait and the events leading up to the current crisis with North Korea. The uncertainty stems in part from the fact that American interests in these regions are simply less pressing than the Cold War need to defend Western Europe. But the effect is plain: American nuclear protection for allies is a thing of the past.

To deal with emerging nuclear threats, the United States could refine its special operations forces and earth-penetrating missiles to destroy subterranean weapons facilities. While these instruments should be used to keep dangerous states from possessing nuclear weapons, American leaders characteristically avoid preemptive military strikes. Since the mid-1980s, for instance, the United States has had a declared policy of seeking out and destroying terrorists in advance of an attack. But this policy has never been carried out. And it is unlikely that any U.S. president would launch preemptive strikes against nuclear facilities, especially when the attack could spread dangerous radiation.

The entire weight of protecting allies and U.S. forces overseas thus falls on plans for a limited theater defense against ballistic missiles. The Bush administration had proposed spending about $6.3 billion to research and develop such a system. But between Congress and the Clinton administration, this figure has been slashed to about $3 billion, with further cuts recommended over the next five years. This program, however, is only a fragment of a shield that will not assure an ally of protection against a nuclear attack, especially one that throws a host of lethal weapons and dummies into the atmosphere. It would not defend a square foot of the United States and, most important, it is not a weapon of deterrence.

When such a theater defense is finally put into place, if the warhead of a single missile detonates over its target, the United States will confront precisely the same dilemma it faces today. The choices will be either to legitimize crossing the nuclear threshold by responding in kind, to ignore a grievous attack on an ally, or to attempt to punish the transgressor with a steadily dwindling conventional military. The political effect of such unacceptable alternatives is to undermine an effective system of American alliances. The United States could eventually find itself in the position of having to defend its principles single-handedly. Moreover, America's inability to provide an effective nuclear deterrent against hostile despots hastens international disorder and, in turn, threatens its own security.

The actual use of a nuclear device would greatly increase pressure on many countries to start or complete their own nuclear weapons programs. Even the possibility that North Korea may possess a nuclear weapon has produced a legitimate and widening discussion in Japan of what, until 12 months ago, was unthinkable: a Japanese nuclear arsenal. Tokyo's effort to achieve nuclear status would generate similar programs throughout the region. An atomic burst—of whatever size—red in anger anywhere on earth would turn the world's most economically dynamic and populous quarter into a pressure cooker heated by ancient enmities and the most powerful instruments of modern destruction.

The possibility of a direct attack on the United States will grow as more nations are able to field the most effective means of delivering unconventional weapons: rockets and submarines. Thanks to exporters such as China and North Korea, membership in the intermediate-range ballistic missile club is less and less exclusive. India, Pakistan, Saudi Arabia, Algeria, Egypt, Iraq, Iran and Afghanistan can now launch IRBMs, or are on the threshold of being able to do so. The new ballistic missile Pyongyang tested in 1993 brings Osaka, Japan, within range of North Korean fire. Director of Central Intelligence James Woolsey has told the House Select Intelligence Committee that 20 states will possess IRBM capability by the end of this decade.

Putting IRBMs aboard submarines is not an easy technical feat. But building or buying submarines, especially cheap and quiet diesel or electric subs, is not difficult. Besides the United States, 15 other nations manufacture submarines, either under licensing agreements or of their own design. Moreover, unclassified data from the recently completed global positioning satellite system gives any vessel extremely reliable information on its location, removing a significant obstacle to launching an accurate missile attack from the sea. Such a depressed-trajectory missile launched from a quiet submarine o the U.S. coast is a defense planner's nightmare. The missile would be difficult to detect and allow very little time to shoot it down—even if the United States were protected by a ballistic missile defense. In a world that had left behind the prohibition against using nuclear weapons, this nightmare turns real.

Any breach of the stricture on using nuclear weapons could slide the world toward greater and more general chaos. The effect can be likened to current concerns over Bosnia. The West's recognition of Bosnia and subsequent failure to help defend it erodes the international stricture against seizing a sovereign nation by force. It signals to ambitious, would-be rulers such as Russia's Vladimir Zhirinovsky that the revanchist path is open, that the sovereignty of Ukraine, Georgia and even Poland is important but not critical to the free nations. Greece's deputy foreign minister, George Papandreou, has seen fit to remark, "The whole concept of a multiethnic Europe has been undermined. We're tearing apart what we've built over the last 50 years in Europe."

Even a single detonation could subvert the international stricture against using nuclear weapons. Yet that possibility is only increasing. One tactical nuclear burst, for example, could be very useful in gaining the advantage before or during an armored engagement, such as a North Korean invasion across the demilitarized zone or a Russian attack on Ukraine. U.S. deterrence policy must seek not only to prevent a nuclear blast but, if deterrence fails, it should reassert the primacy of a valuable international norm to help stem any movement toward wider nuclear use. The world is clearly better off if the unwritten proscription against the use of nuclear weapons remains intact.


The objectives of U.S. policy should be both to deter nuclear danger and to contain and defeat a possible aggressor. At a minimum, the United States needs an effective shield to defend itself and its allies from a rocket-borne nuclear or chemical weapon attack. But U.S. policymakers must assume that any defensive shield against a missile attack is imperfect. Thus the United States must concentrate equally on fashioning an effective sword—one that will discourage attack without using the nuclear force that would turn an increasingly unstable world into a certifiably violent one.

The U.S. cruise missile system contains many of the essential elements that can provide conventional deterrence against a nuclear threat. Launched at a distance of several hundred miles from an enemy's border, conventionally armed cruise missiles are extremely accurate and will become more so in the near future. This accuracy allows them to be aimed, for example, at an adversary's ammunition bunkers, petroleum storage facilities and repair shops without the unnecessary waste and effort of destroying the entire base.

Once multiplied, a cruise missile attack could decimate an enemy's military, providing an effective deterrent against or meaningful punishment for a smaller nations use of nuclear force. The threat of attack against military targets by a thousand conventionally armed missiles, for example, would be a very persuasive argument against using weapons of mass destruction. And if an adversary knew that another salvo could be launched at its power-generating facilities, key communications and transportation nodes and vital industrial plants, the potential of a cruise missile conventional deterrent would be held in even greater respect. Such a response would offer a threat equal in devastation to the 60,000 sorties carried out by coalition forces during the Gulf War.

A ruler who would not be daunted by the prospect of such damage is also not likely to be deterred by any other weapon in the U.S. arsenal. A despot may willingly sacrifice the lives of his people, but because the military and its sustaining infrastructure are vital sources of his power, the prospect of their loss is far more painful. Such a conventional deterrent would allow the United States to apply formidable pressure to these sensitive points. Short of an expensive and blood-soaked invasion, no other conventional option is currently available. A cruise missile deterrent is particularly well-suited to minimizing U.S. combat casualties as well as civilian deaths. It is also consistent with America's global leadership role as well as its strategic advantage in high technology.

With about 2,300 cruise missiles in its current inventory, the United States is at least five years from fielding such an extensive deterrent system. More and better intelligence is needed to identify the critical targets of a large-scale cruise missile attack. Military planners need to know not only the exact position of key installations, but what takes place in which parts of them and, in some circumstances, when. Without this kind of precise information, the increasing precision of the weaponry is an expensive ornament. As this intelligence is refined, cruise missile accuracy should also be improved beyond current impressive levels. Further innovations must also allow military commanders fast, retargeting flexibility so that important battle field intelligence—for example, the changing positions of an adversary's mobile missiles—can be put to effective use.

Other manageable technical problems must also be solved. Large airplanes that can re missiles at a relatively safe distance from an enemy's border provide an almost immediate deterrent. Ships, however, offer the cheapest and most flexible way to move two or three salvos of a thousand cruise missiles each to within striking distance of any threat on the earth's surface. Vessels equipped to carry and launch so large and unique a military payload do not exist and would have to be specially built. Moreover, a safe and practical way to resupply ships at sea with cruise missiles has yet to be developed.

Although the 23 cruise missiles launched at Saddam Hussein's intelligence facility last June merely equaled the price of a single manned bomber, the approximately $2 million cost of the most expensive cruise missile needs to be substantially reduced. Greater privatization and competition in manufacturing the weapon as well as increased purchases would cut the cost of the missile sharply. Similar savings were realized in the Army's purchase of Patriot missiles, which cost $4.24 million each when 176 were ordered in 1982, and $1.16 million apiece when another 815 were ordered in 1990.

The costs of these improvements are slight compared to the much larger bureaucratic interest in preserving Cold War military hardware and doctrines. They are small when set alongside popular pressures to decrease the defense budget. And they are no match at all for an especially formidable obstacle to military change: Washington's stated counter-proliferation policy, which seeks to limit the spread of unconventional weapons rather than prevent their use.

The unique advantage of this deterrent is its ability to discourage a nuclear attack on allies as well as on the United States. This may not utterly solve the strategist's ancient problem of where to make one's stand. But it provides a sobering disincentive to aggression that can be moved around the globe independently of U.S. access to foreign bases. Most important, it can be used to prevent the crossing of the nuclear threshold, whereas launching any part of the U.S. nuclear arsenal would help gut this still valuable, if implicit, international prohibition.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Seth Cropsey is director of the Asian Studies Center at the Heritage Foundation. He served in the Pentagon during the Reagan and Bush administrations.
  • More By Seth Cropsey