We are four Americans who have been concerned over many years with the relation between nuclear weapons and the peace and freedom of the members of the Atlantic Alliance. Having learned that each of us separately has been coming to hold new views on this hard but vital question, we decided to see how far our thoughts, and the lessons of our varied experiences, could be put together; the essay that follows is the result. It argues that a new policy can bring great benefits, but it aims to start a discussion, not to end it.

For 33 years now, the Atlantic Alliance has relied on the asserted readiness of the United States to use nuclear weapons if necessary to repel aggression from the East. Initially, indeed, it was widely thought (notably by such great and different men as Winston Churchill and Niels Bohr) that the basic military balance in Europe was between American atomic bombs and the massive conventional forces of the Soviet Union. But the first Soviet explosion, in August 1949, ended the American monopoly only one month after the Senate approved the North Atlantic Treaty, and in 1950 communist aggression in Korea produced new Allied attention to the defense of Europe.

The "crude" atomic bombs of the 1940s have been followed in both countries by a fantastic proliferation of weapons and delivery systems, so that today the two parts of a still-divided Europe are targeted by many thousands of warheads both in the area and outside it. Within the Alliance, France and Britain have developed thermonuclear forces which are enormous compared to what the United States had at the beginning, although small by comparison with the present deployments of the superpowers. Doctrine has succeeded doctrine, from "balanced collective forces" to "massive retaliation" to "mutual assured destruction" to "flexible response" and the "seamless web." Throughout these transformations, most of them occasioned at least in part by changes in the Western view of Soviet capabilities, both deployments and doctrines have been intended to deter Soviet aggression and keep the peace by maintaining a credible connection between any large-scale assault, whether conventional or nuclear, and the engagement of the strategic nuclear forces of the United States.

A major element in every doctrine has been that the United States has asserted its willingness to be the first-has indeed made plans to be the first if necessary-to use nuclear weapons to defend against aggression in Europe. It is this element that needs reexamination now. Both its cost to the coherence of the Alliance and its threat to the safety of the world are rising while its deterrent credibility declines.

This policy was first established when the American nuclear advantage was overwhelming, but that advantage has long since gone and cannot be recaptured. As early as the 1950s it was recognized by both Prime Minister Churchill and President Eisenhower that the nuclear strength of both sides was becoming so great that a nuclear war would be a ghastly catastrophe for all concerned. The following decades have only confirmed and intensified that reality. The time has come for careful study of the ways and means of moving to a new Alliance policy and doctrine: that nuclear weapons will not be used unless an aggressor should use them first.


The disarray that currently besets the nuclear policy and practices of the Alliance is obvious. Governments and their representatives have maintained an appearance of unity as they persist in their support of the two-track decision of December 1979, under which 572 new American missiles of intermediate range are to be placed in Europe unless a satisfactory agreement on the limitation of such weapons can be reached in the negotiations between the United States and the Soviet Union that began last November. But behind this united front there are divisive debates, especially in countries where the new weapons are to be deployed.

The arguments put forward by advocates of these deployments contain troubling variations. The simplest and intuitively the most persuasive claim is that these new weapons are needed as a counter to the new Soviet SS-20 missiles; it may be a recognition of the surface attractiveness of this position that underlies President Reagan's striking-but probably not negotiable-proposal that if all the SS-20s are dismantled the planned deployments will be cancelled. Other officials have a quite different argument, that without new and survivable American weapons which can reach Russia from Western Europe there can be no confidence that the strategic forces of the United States will remain committed to the defense of Western Europe; on this argument the new missiles are needed to make it more likely that any war in Europe would bring nuclear warheads on the Soviet Union and thus deter the aggressor in the first place. This argument is logically distinct from any concern about the Soviet SS-20s, and it probably explains the ill-concealed hope of some planners that the Reagan proposal will be rejected. Such varied justifications cast considerable doubt on the real purpose of the proposed deployment.

An equally disturbing phenomenon is the gradual shift in the balance of argument that has occurred since the need to address the problem was first asserted in 1977. Then the expression of need was European, and in the first instance German; the emerging parity of long-range strategic systems was asserted to create a need for a balance at less than intercontinental levels. The American interest developed relatively slowly, but because these were to be American missiles, American planners took the lead as the proposal was worked out. It has also served Soviet purposes to concentrate on the American role. A similar focus has been chosen by many leaders of the new movement for nuclear disarmament in Europe. And now there are American voices, some in the executive branch, talking as if European acceptance of these new missiles were some sort of test of European loyalty to the Alliance. Meanwhile some of those in Europe who remain publicly committed to both tracks of the 1979 agreement are clearly hoping that the day of deployment will never arrive. When the very origins of a new proposal become the source of irritated argument among allies-"You started it!"-something is badly wrong in our common understanding.

A still more severe instance of disarray, one which has occurred under both President Carter and President Reagan, relates to the so-called neutron bomb, a weapon designed to meet the threat of Soviet tanks. American military planners, authorized by doctrine to think in terms of early battlefield use of nuclear weapons, naturally want more "up-to-date" weapons than those they have now; it is known that thousands of the aging short-range nuclear weapons now in Europe are hard to use effectively. Yet to a great many Europeans the neutron bomb suggests, however unfairly, that the Americans are preparing to fight a "limited" nuclear war on their soil. Moreover neither weapons designers nor the Pentagon officials they have persuaded seem to have understood the intense and special revulsion that is associated with killing by "enhanced radiation."

All these recent distempers have a deeper cause. They are rooted in the fact that the evolution of essentially equivalent and enormously excessive nuclear weapons systems both in the Soviet Union and in the Atlantic Alliance has aroused new concern about the dangers of all forms of nuclear war. The profusion of these systems, on both sides, has made it more difficult than ever to construct rational plans for any first use of these weapons by anyone.

This problem is more acute than before, but it is not new. Even in the 1950s, a time that is often mistakenly perceived as one of effortless American superiority, the prospect of any actual use of tactical weapons was properly terrifying to Europeans and to more than a few Americans. Military plans for such use remained both deeply secret and highly hypothetical; the coherence of the Alliance was maintained by general neglect of such scenarios, not by sedulous public discussion. In the 1960s there was a prolonged and stressful effort to address the problem of theater-range weapons, but agreement on new forces and plans for their use proved elusive. Eventually the proposal for a multilateral force (MLF) was replaced by the assignment of American Polaris submarines to NATO, and by the creation in Brussels of an inter-allied Nuclear Planning Group. Little else was accomplished. In both decades the Alliance kept itself together more by mutual political confidence than by plausible nuclear war-fighting plans.

Although the first years of the 1970s produced a welcome if oversold détente, complacency soon began to fade. The Nixon Administration, rather quietly, raised the question about the long-run credibility of the American nuclear deterrent that was to be elaborated by Henry Kissinger in 1979 at a meeting in Brussels. Further impetus to both new doctrine and new deployments came during the Ford and Carter Administrations, but each public statement, however careful and qualified, only increased European apprehensions. The purpose of both Administrations was to reinforce deterrence, but the result has been to increase fear of nuclear war, and even of Americans as its possible initiators. Intended as contributions to both rationality and credibility, these excursions into the theory of limited nuclear war have been counterproductive in Europe.

Yet it was not wrong to raise these matters. Questions that were answered largely by silence in the 1950s and 1960s cannot be so handled in the 1980s. The problem was not in the fact that the questions were raised, but in the way they seemed to be answered.

It is time to recognize that no one has ever succeeded in advancing any persuasive reason to believe that any use of nuclear weapons, even on the smallest scale, could reliably be expected to remain limited. Every serious analysis and every military exercise, for over 25 years, has demonstrated that even the most restrained battlefield use would be enormously destructive to civilian life and property. There is no way for anyone to have any confidence that such a nuclear action will not lead to further and more devastating exchanges. Any use of nuclear weapons in Europe, by the Alliance or against it, carries with it a high and inescapable risk of escalation into the general nuclear war which would bring ruin to all and victory to none.

The one clearly definable firebreak against the worldwide disaster of general nuclear war is the one that stands between all other kinds of conflict and any use whatsoever of nuclear weapons. To keep that firebreak wide and strong is in the deepest interest of all mankind. In retrospect, indeed, it is remarkable that this country has not responded to this reality more quickly. Given the appalling consequences of even the most limited use of nuclear weapons and the total impossibility for both sides of any guarantee against unlimited escalation, there must be the gravest doubt about the wisdom of a policy which asserts the effectiveness of any first use of nuclear weapons by either side. So it seems timely to consider the possibilities, the requirements, the difficulties, and the advantages of a policy of no-first-use.


The largest question presented by any proposal for an Allied policy of no-first-use is that of its impact on the effectiveness of NATO's deterrent posture on the central front. In spite of the doubts that are created by any honest look at the probable consequences of resort to a first nuclear strike of any kind, it should be remembered that there were strong reasons for the creation of the American nuclear umbrella over NATO. The original American pledge, expressed in Article 5 of the Treaty, was understood to be a nuclear guarantee. It was extended at a time when only a conventional Soviet threat existed, so a readiness for first use was plainly implied from the beginning. To modify that guarantee now, even in the light of all that has happened since, would be a major change in the assumptions of the Alliance, and no such change should be made without the most careful exploration of its implications.

In such an exploration the role of the Federal Republic of Germany must be central. Americans too easily forget what the people of the Federal Republic never can: that their position is triply exposed in a fashion unique among the large industrial democracies. They do not have nuclear weapons; they share a long common boundary with the Soviet empire; in any conflict on the central front their land would be the first battleground. None of these conditions can be changed, and together they present a formidable challenge.

Having decisively rejected a policy of neutrality, the Federal Republic has necessarily relied on the nuclear protection of the United States, and we Americans should recognize that this relationship is not a favor we are doing our German friends, but the best available solution of a common problem. Both nations believe that the Federal Republic must be defended; both believe that the Federal Republic must not have nuclear weapons of its own; both believe that nuclear guarantees of some sort are essential; and both believe that only the United States can provide those guarantees in persuasively deterrent peacekeeping form.

The uniqueness of the West German position can be readily demonstrated by comparing it with those of France and the United Kingdom. These two nations have distance, and in one case water, between them and the armies of the Soviet Union; they also have nuclear weapons. While those weapons may contribute something to the common strength of the Alliance, their main role is to underpin a residual national self-reliance, expressed in different ways at different times by different governments, which sets both Britain and France apart from the Federal Republic. They are set apart from the United States too, in that no other nation depends on them to use their nuclear weapons otherwise than in their own ultimate self-defense.

The quite special character of the nuclear relationship between the Federal Republic and the United States is a most powerful reason for defining that relationship with great care. It is rare for one major nation to depend entirely on another for a form of strength that is vital to its survival. It is unprecedented for any nation, however powerful, to pledge itself to a course of action, in defense of another, that might entail its own nuclear devastation. A policy of no-first-use would not and should not imply an abandonment of this extraordinary guarantee-only its redefinition. It would still be necessary to be ready to reply with American nuclear weapons to any nuclear attack on the Federal Republic, and this commitment would in itself be sufficiently demanding to constitute a powerful demonstration that a policy of no-first-use would represent no abandonment of our German ally.

The German right to a voice in this question is not merely a matter of location, or even of dependence on an American nuclear guarantee. The people of the Federal Republic have demonstrated a steadfast dedication to peace, to collective defense, and to domestic political decency. The study here proposed should be responsive to their basic desires. It seems probable that they are like the rest of us in wishing most of all to have no war of any kind, but also to be able to defend the peace by forces that do not require the dreadful choice of nuclear escalation.


While we believe that careful study will lead to a firm conclusion that it is time to move decisively toward a policy of no-first-use, it is obvious that any such policy would require a strengthened confidence in the adequacy of the conventional forces of the Alliance, above all the forces in place on the central front and those available for prompt reinforcement. It seems clear that the nations of the Alliance together can provide whatever forces are needed, and within realistic budgetary constraints, but it is a quite different question whether they can summon the necessary political will. Evidence from the history of the Alliance is mixed. There has been great progress in the conventional defenses of NATO in the 30 years since the 1952 Lisbon communiqué, but there have also been failures to meet force goals all along the way.

In each of the four nations which account for more than 90 percent of NATO's collective defense and a still higher proportion of its strength on the central front, there remain major unresolved political issues that critically affect contributions to conventional deterrence: for example, it can be asked what priority the United Kingdom gives to the British Army of the Rhine, what level of NATO-connected deployment can be accepted by France, what degree of German relative strength is acceptable to the Allies and fair to the Federal Republic itself, and whether we Americans have a durable and effective answer to our military manpower needs in the present all-volunteer active and reserve forces. These are the kinds of questions-and there are many more-that would require review and resolution in the course of reaching any final decision to move to a responsible policy of no-first-use.

There should also be an examination of the ways in which the concept of early use of nuclear weapons may have been built into existing forces, tactics, and general military expectations. To the degree that this has happened, there could be a dangerous gap right now between real capabilities and those which political leaders might wish to have in a time of crisis. Conversely there should be careful study of what a policy of no-first-use would require in those same terms. It seems more than likely that once the military leaders of the Alliance have learned to think and act steadily on this "conventional" assumption, their forces will be better instruments for stability in crises and for general deterrence, as well as for the maintenance of the nuclear firebreak so vital to us all.

No one should underestimate either the difficulty or the importance of the shift in military attitudes implied by a no-first-use policy. Although military commanders are well aware of the terrible dangers in any exchange of nuclear weapons, it is a strong military tradition to maintain that aggressive war, not the use of any one weapon, is the central evil. Many officers will be initially unenthusiastic about any formal policy that puts limits on their recourse to a weapon of apparently decisive power. Yet the basic argument for a no-first-use policy can be stated in strictly military terms: that any other course involves unacceptable risks to the national life that military forces exist to defend. The military officers of the Alliance can be expected to understand the force of this proposition, even if many of them do not initially agree with it. Moreover, there is every reason for confidence that they will loyally accept any policy that has the support of their governments and the peoples behind them, just as they have fully accepted the present arrangements under which the use of nuclear weapons, even in retaliation for a nuclear attack, requires advance and specific approval by the head of government.

An Allied posture of no-first-use would have one special effect that can be set forth in advance: it would draw new attention to the importance of maintaining and improving the specifically American conventional forces in Europe. The principal political difficulty in a policy of no-first-use is that it may be taken in Europe, and especially in the Federal Republic, as evidence of a reduced American interest in the Alliance and in effective overall deterrence. The argument here is exactly the opposite: that such a policy is the best one available for keeping the Alliance united and effective. Nonetheless the psychological realities of the relation between the Federal Republic and the United States are such that the only way to prevent corrosive German suspicion of American intentions, under a no-first-use regime, will be for Americans to accept for themselves an appropriate share in any new level of conventional effort that the policy may require.

Yet it would be wrong to make any hasty judgment that those new levels of effort must be excessively high. The subject is complex, and the more so because both technology and politics are changing. Precision-guided munitions, in technology, and the visible weakening of the military solidity of the Warsaw Pact, in politics, are only two examples of changes working to the advantage of the Alliance. Moreover there has been some tendency, over many years, to exaggerate the relative conventional strength of the U.S.S.R. and to underestimate Soviet awareness of the enormous costs and risks of any form of aggression against NATO.

Today there is literally no one who really knows what would be needed. Most of the measures routinely used in both official and private analyses are static and fragmentary. An especially arbitrary, if obviously convenient, measure of progress is that of spending levels. But it is political will, not budgetary pressure, that will be decisive. The value of greater safety from both nuclear and conventional danger is so great that even if careful analysis showed that the necessary conventional posture would require funding larger than the three-percent real increase that has been the common target of recent years, it would be the best bargain ever offered to the members of the Alliance.

Yet there is no need for crash programs, which always bring extra costs. The direction of the Allied effort will be more important than its velocity. The final establishment of a firm policy of no-first-use, in any case, will obviously require time. What is important today is to begin to move in this direction.


The concept of renouncing any first use of nuclear weapons should also be tested by careful review of the value of existing NATO plans for selective and limited use of nuclear weapons. While many scenarios for nuclear war-fighting are nonsensical, it must be recognized that cautious and sober senior officers have found it prudent to ask themselves what alternatives to defeat they could propose to their civilian superiors if a massive conventional Soviet attack seemed about to make a decisive breakthrough. This question has generated contingency plans for battlefield uses of small numbers of nuclear weapons which might prevent that particular disaster. It is hard to see how any such action could be taken without the most enormous risk of rapid and catastrophic escalation, but it is a fair challenge to a policy of no-first-use that it should be accompanied by a level of conventional strength that would make such plans unnecessary.

In the light of this difficulty it would be prudent to consider whether there is any acceptable policy short of no-first-use. One possible example is what might be called "no-early-first-use;" such a policy might leave open the option of some limited nuclear action to fend off a final large-scale conventional defeat, and by renunciation of any immediate first use and increased emphasis on conventional capabilities it might be thought to help somewhat in reducing current fears.

But the value of a clear and simple position would be great, especially in its effect on ourselves and our Allies. One trouble with exceptions is that they easily become rules. It seems much better that even the most responsible choice of even the most limited nuclear actions to prevent even the most imminent conventional disaster should be left out of authorized policy. What the Alliance needs most today is not the refinement of its nuclear options, but a clear-cut decision to avoid them as long as others do.


Who should make the examination here proposed? The present American Administration has so far shown little interest in questions of this sort, and indeed a seeming callousness in some quarters in Washington toward nuclear dangers may be partly responsible for some of the recent unrest in Europe. But each of the four of us has served in Administrations which revised their early thoughts on nuclear weapons policy. James Byrnes learned the need to seek international control; John Foster Dulles stepped back somewhat from his early belief in massive retaliation; Dwight Eisenhower came to believe in the effort to ban nuclear tests which he at first thought dangerous; the Administration of John F. Kennedy (in which we all served) modified its early views on targeting doctrine; Lyndon Johnson shelved the proposed MLF when he decided it was causing more trouble than it was worth; and Richard Nixon agreed to narrow limits on anti-ballistic missiles whose large-scale deployment he had once thought indispensable. There were changes also in the Ford and Carter Administrations, and President Reagan has already adjusted his views on the usefulness of early arms control negotiations, even though we remain in a time of general stress between Washington and Moscow. No Administration should be held, and none should hold itself, to inflexible first positions on these extraordinarily difficult matters.

Nor does this question need to wait upon governments for study. The day is long past when public awe and governmental secrecy made nuclear policy a matter for only the most private executive determination. The questions presented by a policy of no-first-use must indeed be decided by governments, but they can and should be considered by citizens. In recent months strong private voices have been raised on both sides of the Atlantic on behalf of strengthened conventional forces. When this cause is argued by such men as Christoph Bertram, Field Marshal Lord Carver, Admiral Noel Gayler, Professor Michael Howard, Henry Kissinger, François de Rose, Theo Sommer, and General Maxwell Taylor, to name only a few, it is fair to conclude that at least in its general direction the present argument is not outside the mainstream of thinking within the Alliance. Indeed there is evidence of renewed concern for conventional forces in governments too.

What should be added, in both public and private sectors, is a fresh, sustained, and careful consideration of the requirements and the benefits of deciding that the policy of the Atlantic Alliance should be to keep its nuclear weapons unused as long as others do the same. Our own belief, though we do not here assert it as proven, is that when this possibility is fully explored it will be evident that the advantages of the policy far outweigh its costs, and that this demonstration will help the peoples and governments of the Alliance to find the political will to move in this direction. In this spirit we go on to sketch the benefits that could come from such a change.


The first possible advantage of a policy of no-first-use is in the management of the nuclear deterrent forces that would still be necessary. Once we escape from the need to plan for a first use that is credible, we can escape also from many of the complex arguments that have led to assertions that all sorts of new nuclear capabilities are necessary to create or restore a capability for something called "escalation dominance"-a capability to fight and "win" a nuclear war at any level. What would be needed, under no-first-use, is a set of capabilities we already have in overflowing measure-capabilities for appropriate retaliation to any kind of Soviet nuclear attack which would leave the Soviet Union in no doubt that it too should adhere to a policy of no-first-use. The Soviet government is already aware of the awful risk inherent in any use of these weapons, and there is no current or prospective Soviet "superiority" that would tempt anyone in Moscow toward nuclear adventurism. (All four of us are wholly unpersuaded by the argument advanced in recent years that the Soviet Union could ever rationally expect to gain from such a wild effort as a massive first strike on land-based American strategic missiles.)

Once it is clear that the only nuclear need of the Alliance is for adequately survivable and varied second strike forces, requirements for the modernization of major nuclear systems will become more modest than has been assumed. In particular we can escape from the notion that we must somehow match everything the rocket commanders in the Soviet Union extract from their government. It seems doubtful, also, that under such a policy it would be necessary or desirable to deploy neutron bombs. The savings permitted by more modest programs could go toward meeting the financial costs of our contribution to conventional forces.

It is important to avoid misunderstanding here. In the conditions of the 1980s, and in the absence of agreement on both sides to proceed to very large-scale reductions in nuclear forces, it is clear that large, varied, and survivable nuclear forces will still be necessary for nuclear deterrence. The point is not that we Americans should move unilaterally to some "minimum" force of a few tens or even hundreds of missiles, but rather that once we escape from the pressure to seem willing and able to use these weapons first, we shall find that our requirements are much less massive than is now widely supposed.

A posture of no-first-use should also go far to meet the understandable anxieties that underlie much of the new interest in nuclear disarmament, both in Europe and in our own country. Some of the proposals generated by this new interest may lack practicability for the present. For example, proposals to make "all" of Europe-from Portugal to Poland-a nuclear-free zone do not seem to take full account of the reality that thousands of long-range weapons deep in the Soviet Union will still be able to target Western Europe. But a policy of no-first-use, with its accompaniment of a reduced requirement for new Allied nuclear systems, should allow a considerable reduction in fears of all sorts. Certainly such a new policy would neutralize the highly disruptive argument currently put about in Europe: that plans for theater nuclear modernization reflect an American hope to fight a nuclear war limited to Europe. Such modernization might or might not be needed under a policy of no-first-use; that question, given the size and versatility of other existing and prospective American forces, would be a matter primarily for European decision (as it is today).

An effective policy of no-first-use will also reduce the risk of conventional aggression in Europe. That risk has never been as great as prophets of doom have claimed and has always lain primarily in the possibility that Soviet leaders might think they could achieve some quick and limited gain that would be accepted because no defense or reply could be concerted. That temptation has been much reduced by the Allied conventional deployments achieved in the last 20 years, and it would be reduced still further by the additional shift in the balance of Allied effort that a no-first-use policy would both permit and require. The risk that an adventurist Soviet leader might take the terrible gamble of conventional aggression was greater in the past than it is today, and is greater today than it would be under no-first-use, backed up by an effective conventional defense.


We have been discussing a problem of military policy, but our interest is also political. The principal immediate danger in the current military posture of the Alliance is not that it will lead to large-scale war, conventional or nuclear. The balance of terror, and the caution of both sides, appear strong enough today to prevent such a catastrophe, at least in the absence of some deeply destabilizing political change which might lead to panic or adventurism on either side. But the present unbalanced reliance on nuclear weapons, if long continued, might produce exactly such political change. The events of the last year have shown that differing perceptions of the role of nuclear weapons can lead to destructive recriminations, and when these differences are compounded by understandable disagreements on other matters such as Poland and the Middle East, the possibilities for trouble among Allies are evident.

The political coherence of the Alliance, especially in times of stress, is at least as important as the military strength required to maintain credible deterrence. Indeed the political requirement has, if anything, an even higher priority. Soviet leaders would be most pleased to help the Alliance fall into total disarray, and would much prefer such a development to the inescapable uncertainties of open conflict. Conversely, if consensus is re-established on a military policy that the peoples and governments of the Alliance can believe in, both political will and deterrent credibility will be reinforced. Plenty of hard questions will remain, but both fear and mistrust will be reduced, and they are the most immediate enemies.

There remains one underlying reality which could not be removed by even the most explicit declaratory policy of no-first-use. Even if the nuclear powers of the Alliance should join, with the support of other Allies, in a policy of no-first-use, and even if that decision should lead to a common declaration of such policy by these powers and the Soviet Union, no one on either side could guarantee beyond all possible doubt that if conventional warfare broke out on a large scale there would in fact be no use of nuclear weapons. We could not make that assumption about the Soviet Union, and we must recognize that Soviet leaders could not make it about us. As long as the weapons themselves exist, the possibility of their use will remain.

But this inescapable reality does not undercut the value of a no-first-use policy. That value is first of all for the internal health of the Western Alliance itself. A posture of effective conventional balance and survivable second-strike nuclear strength is vastly better for our own peoples and governments, in a deep sense more civilized, than one that forces the serious contemplation of "limited" nuclear scenarios that are at once terrifying and implausible.

There is strong reason to believe that no-first-use can also help in our relations with the Soviet Union. The Soviet government has repeatedly offered to join the West in declaring such a policy, and while such declarations may have only limited reliability, it would be wrong to disregard the real value to both sides of a jointly declared adherence to this policy. To renounce the first use of nuclear weapons is to accept an enormous burden of responsibility for any later violation. The existence of such a clearly declared common pledge would increase the cost and risk of any sudden use of nuclear weapons by either side and correspondingly reduce the political force of spoken or unspoken threats of such use.

A posture and policy of no-first-use also could help to open the path toward serious reduction of nuclear armaments on both sides. The nuclear decades have shown how hard it is to get agreements that really do constrain these weapons, and no one can say with assurance that any one step can make a decisive difference. But just as a policy of no-first-use should reduce the pressures on our side for massive new nuclear forces, it should help to increase the international incentives for the Soviet Union to show some restraint of its own. It is important not to exaggerate here, and certainly Soviet policies on procurement are not merely delayed mirror-images of ours. Nonetheless there are connections between what is said and what is done even in the Soviet Union, and there are incentives for moderation, even there, that could be strengthened by a jointly declared policy of renouncing first use. At a minimum such a declaration would give both sides additional reason to seek for agreements that would prevent a vastly expensive and potentially destabilizing contest for some kind of strategic advantage in outer space.

Finally, and in sum, we think a policy of no-first-use, especially if shared with the Soviet Union, would bring new hope to everyone in every country whose life is shadowed by the hideous possibility of a third great twentieth-century conflict in Europe-conventional or nuclear. It seems timely and even urgent to begin the careful study of a policy that could help to sweep this threat clean off the board of international affairs.


We recognize that we have only opened this large question, that we have exhausted no aspect of it, and that we may have omitted important elements. We know that NATO is much more than its four strongest military members; we know that a policy of no-first-use in the Alliance would at once raise questions about America's stance in Korea and indeed other parts of Asia. We have chosen deliberately to focus on the central front of our central alliance, believing that a right choice there can only help toward right choices elsewhere.

What we dare to hope for is the kind of new and widespread consideration of the policy we have outlined that helped us 15 years ago toward SALT I, 25 years ago toward the Limited Test Ban, and 35 years ago toward the Alliance itself. Such consideration can be made all the more earnest and hopeful by keeping in mind one simple and frequently neglected reality: there has been no first use of nuclear weapons since 1945, and no one in any country regrets that fact. The right way to maintain this record is to recognize that in the age of massive thermonuclear overkill it no longer makes sense-if it ever did-to hold these weapons for any other purpose than the prevention of their use.

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
  • McGeorge Bundy was Special Assistant to the President for National Security Affairs from 1961 to 1966 and President of the Ford Foundation from 1966 to mid-1979. He is currently Professor of History at New York University.
  • George F. Kennan is Professor Emeritus at the Institute for Advanced Study, Princeton. He was U.S. Ambassador to the Soviet Union, 1952, and to Yugoslavia, 1961-63, and is the author of Soviet-American Relations, 1917-20 (2 Vols.); Memoirs (2 Vols.) and other works.
  • Robert S. McNamara was Secretary of Defense from 1961 to 1968 and President of the World Bank from 1968 to mid-1981.
  • Gerard C. Smith was Chief of the U.S. Delegation to the Strategic Arms Limitations Talks (SALT) from 1969 to 1972, and is the author of Doubletalk: The Story of SALT I. He also served as Special Assistant to the Secretary of State for atomic energy affairs (1954-57), Director of the Policy Planning Staff of the Department of State (1957-61), a full-time consultant on the Multilateral Force (1961-64), and Ambassador at Large and Special Presidential Representative for nonproliferation matters (1977-80).
  • More By McGeorge Bundy
  • More By George F. Kennan
  • More By Robert S. McNamara
  • More By Gerard C. Smith