JUST AS last week's NATO summit was a reminder that the world faces a future in which the United States has become the sole global power, it likewise signals that it is time to re-examine the role and place of strategic nuclear weapons in American security policy. With the dissolution of the Soviet Union and the division of its nuclear forces, the concerns of U.S. strategic planning have shifted from a single nuclear threat to a complex international situation -- from Iraq to Ukraine to North Korea -- where regional aggression is more likely than it has been since before World War II.

Our experience, including the Gulf War, also teaches us that nuclear weapons are unlikely to be useful in deterring aggression in these circumstances. Rather than rely on them, therefore, the United States should consider what might seem at first glance a step backward: converting its principal strategic deterrent from nuclear weapons to a more credible deterrence based at least in part upon "smart" conventional weapons. It is a shift that could be justified as a coldly rational approach to a new security strategy and equally so as a morally correct foreign policy choice.

The case for choosing strategic, high-precision conventional weapons over strategic nuclear weapons is clear. They are safer, cause less collateral damage and pose less threat of escalation than do nuclear weapons. Thus they offer far greater flexibility in a variety of situations where use of any sort of nuclear weapon would be politically or militarily impractical.

The principal challenges to reliance on strategic conventional weapons are also clear. Can they adequately carry out their combat missions? If so, will that fact deter aggression as effectively as nuclear weapons appear to have done? I believe the answers to these questions are, in general, positive and that a strategic conventional military option may become practical for many strategic missions previously thought of as a nuclear preserve.

The Gulf War offered a spectacular demonstration of the potential effectiveness of smart weapons used in a strategic role. Against Iraq, such weapons rapidly rendered useless the military forces of a powerful dictator, in particular by neutralizing his command, control and communications facilities.

At the same time, the Gulf War showed the limited value of nuclear weapons in deterring aggression. Indeed, I would argue that there was no useful role for nuclear weapons for anyone in the Gulf War; Iraq could and did simply ignore allied nuclear weapons as virtually chimerical, even when it attacked Israel. Likewise, Iraq would have gained little by employing a nuclear device. It would not have been possible for Saddam Hussein to diminish significantly the overwhelming military superiority of the forces arrayed against him. For him to have used such a nuclear capability as he might have developed would merely have reinforced the determination of the major powers to eliminate him.

It is also true that a nuclear capability in Saddam's hands might have undermined U.S. efforts to force him to behave responsibly. Nuclear weapons used in desperation, or in a wild plan of revenge against Israel, could have resulted in great human tragedy. We cannot know whether or not Saddam would have used nuclear weapons had he possessed them. But we also cannot know whether the allied nuclear threat could ever be counted upon to deter him from using them.

After all, Saddam chose to start a nuclear weapons program in the very face of the overwhelming nuclear power of the states arrayed against him, including the Israelis he sought to provoke. There was no logical reason for Iraq to build a nuclear weapon outside of this threat of irresponsible behavior: the looming threat of a wildcard, regional nuclear power. To my mind, Saddam's decision to embark upon a nuclear program itself demonstrates that there was no nuclear deterrent at play in Iraq's evaluation of the strategic situation in the gulf.

In the aftermath of the Gulf War, as the lessons of the successes, failures and potential of conventional, smart weapons are digested by all nations, one message rings loudest: The United States, when provoked, can and will use strategic conventional weapons against whatever targets it considers appropriate.

Understanding this single lesson may offer us a way to create the first credible and therefore useful strategic deterrent we have seen since the early days of the nuclear era. It may well be that conventional strategic weapons will one day perform their primary mission of deterrence immeasurably better than nuclear weapons if only because we can -- and will -- use them.

It is now vitally important that we understand both the effectiveness and limitations of strategic conventional weapons. Unfortunately, much of the postwar popular evaluation of our weapons is based on perceptions drawn from media coverage -- coverage often well managed by the Department of Defense. The professional discussion has been superficial, often little more than speculation about the levels of nuclear weapons required in a post-Cold War strategic environment.

However, the absence of informed discussion of the capabilities of this and the next generation of strategic conventional weapons should not keep us from opening a general debate over the future of nuclear weapons. The most encouraging development for Americans is that we may begin to plan a national security strategy that does not rely as heavily on nuclear weapons. For the first time we might reasonably contemplate making nuclear weapons largely obsolete for the most practical and fundamental strategic missions.

From a policy perspective, there should be a conscious decision by the government to pursue the conversion of our strategic deterrent from nuclear to conventional weapons. It is no stretch to assert that we can and should now begin to decide not whether, but in what manner, this conversion will take place. Our discussion ought to focus on what strategic nuclear and strategic conventional weapons can and cannot do and what we should do to maintain and improve the capabilities of these weapons.

But to begin such a discussion, we must establish the truth about smart weapons, especially strategic conventional ones. Even though advanced conventional weapons appear to have performed well in combat, we must be careful as we evaluate how good a model the Gulf War provided for understanding the future utility of such weapons as a deterrent.

To much of the world viewing the Gulf War on television, smart weapons appeared a miracle weapon, a new panacea for all sorts of conflicts capable of doing the job with little loss of military personnel and limited civilian losses. This perception caught the imagination of a people with the reasonable desire to limit human suffering and loss of life under any circumstances. Unfortunately, this may be an unreasonable perception at the current stage of strategic conventional weapons development.

The lessons of the military utility of nuclear weapons must also be re-examined and frankly acknowledged. We will never be certain what has deterred the use of nuclear weapons since 1945. We can speculate that the strategic nuclear arsenals in their morbid way did stay the use of these weapons, that mutually assured destruction may have prevented the use of nuclear weapons against other nuclear powers. But in truth, using nuclear weapons has never entirely been ruled out, and much of the debate of operational nuclear strategy during the Cold War reflected this reality.

What inhibited the American use of nuclear weapons was clearly sensitivity to the implications of the destructiveness of such weapons. And however much U.S. military doctrine asserted otherwise, their use was never an easy option to the United States, and some troublesome governments have known this and exploited it as a weakness in U.S. military posture. While the McNamara-era decision to move towards flexible response certainly led to a more credible U.S. military presence and deterrence for some situations, it did not improve our strategic deterrent. We were left with a massive investment in a nuclear arsenal of limited use except in possibly deterring a nuclear attack by the Soviet Union directly against the United States. It was a one-use strategic deterrent. Developing true strategic conventional weapons offers us a flexible capability that no aggressor can discount safely in a wide range of circumstances.

Certainly, it would be wise to continue to maintain a secure and widely dispersed array of nuclear weapons and their delivery systems until we are assured that the nuclear weapons of others constitute no threat to the United States and its associates. But possession of such nuclear strategic superiority does not, by itself, answer whether it would be wise for the United States actually to use its nuclear forces even in retaliation to the initial use of nuclear weapons by another country. After all, if the country initiating such use could be effectively disarmed by conventional forces, there would be no military reason to retaliate with a nuclear strike.

But how close are we to possessing conventional weapons that can indeed replace nuclear weapons as the primary deterrent against aggression?

Today, there remains a gap between the destructive power of a first-class strategic arsenal, such as that of Russia, and the ability of American strategic conventional weapons to overcome such a threat. Understanding and overcoming this gap should become the focus of technological research into the practical obstacles of delivery, accuracy and explosive capabilities, as well as planning security strategy and tactics.

The Gulf War suggests that U.S. conventional weapons could offer an adequate deterrent against regional aggression. We must still evaluate whether other powers, such as China and Russia, have come to this conclusion. But the present threat does not come primarily from these nations but from states such as Iraq, North Korea or even Libya. While we need to understand what motivates the weapons programs of these states and try to develop a new method of deterring them, it is, unfortunately, not clear that any strategic weapon can deter the ambitions of a tyrant.

The United States should recognize its responsibility to help shape the pattern and purpose of security arrangements worldwide to the long-term interests not merely of the United States but of the world as a whole. The idea that the future peace and well being of the world should rest upon the threat of nuclear annihilation of large numbers of noncombatants is, in the long run, unacceptable. We should treat with scorn those, like North Korea, who may attempt to blackmail others with imprudent nuclear threats.

In the world as it is, we will continue to need nonstrategic conventional forces to stop aggression as it unfolds. We will also need to maintain an overwhelming nuclear strategic capability, though not necessarily to use such weapons -- even in retaliation -- if we can disarm an aggressor with smart non-nuclear strategic weapons. We must learn not merely to react, as eye for eye, or out of anger, but with wisdom and a sense of the great responsibility that comes with great power.

Paul Nitze, former arms control negotiator and ambassador-at-large during the Reagan administration, is diplomat-in-residence at the Paul H. Nitze School of Advanced International Studies, Johns Hopkins University.