The recent meeting of NATO defense and foreign ministers at Athens ended with the usual proclamations of Allied unity. A great deal was made of the United States commitment of five-and later more-Polaris submarines to NATO. Yet the significance of the meeting went far beyond this largely symbolic gesture. The Athens conference marked the point at which a reassessment of NATO strategy could no longer be avoided. It underlined the urgent need to resolve the debate of the past years about the relative role of nuclear and conventional forces, the relationship of deterrence to strategy and the control and use of nuclear weapons.

This debate was first given impetus when, shortly after the advent of the new Administration, the United States proposed that the conventional forces of NATO be strengthened-specifically, that they be brought up to the level of 30 divisions agreed upon in 1957. Our allies, in turn, inquired whether the United States was reducing its reliance on nuclear weapons, and whether our nuclear guarantee was being deprived of its credibility. We replied that the build-up of conventional forces did not imply a reduced reliance on what has been called the Deterrent. On the contrary, the strategic striking force was being expanded and made more invulnerable.1 The non- nuclear build-up, far from diminishing the credibility of our nuclear power, would actually enhance it.2 In short, the NATO policy being promulgated was not new, according to our official statements; we were simply asking our allies to live up to their previous commitment. Our goal was flexibility of response.

This curious dialogue, in which a change of emphasis in United States policy was defended in terms of traditional doctrines of NATO, in many ways only increased the uneasiness of our allies. There were a number of reasons why the NATO goals for conventional forces had never been met. An important one was the fact that in terms of the prevailing NATO strategy these forces did not make much sense.

That strategy had been developed during the period when the United States had strategic preponderance. For the greater part of the nuclear age we could win a general war whether we struck first or second. We could destroy the Soviet means of retaliation at minimum cost. This superiority enabled us not only to prevent an attack on the United States, but to protect our allies by what amounted to a unilateral American guarantee. The Soviet Union did possess superior conventional strength. But it could not set these forces in motion because such an action might have triggered our Strategic Air Command (SAC). It could not attack the United States at the same time as Europe-even after it became technically feasible-because, while this would have increased our losses, a Soviet defeat would nevertheless have been inevitable. Throughout its history, the security of NATO has depended essentially on the ability of the United States to destroy the Soviet retaliatory power-technically speaking, on our capacity to conduct a counterforce strategy.3

Any prudent planning for NATO in the 1960s must start from the assumption that the utility of a counterforce strategy for the defense of Europe is bound to decline. It is possible to accept the statements of our defense leaders that the United States is superior in strategic striking power and intends to remain so. But the strategic and political significance of this superiority is certain to diminish. For the immediate future-as long as the Soviet missile force remains relatively small-a counterforce strategy may remain technically feasible. This situation should be treated, however, as a temporary respite permitting an adjustment to inevitable realities, not as an excuse for continuing to depend on traditional policies.

For one thing, if a counterforce strategy is to retain any chance of success, the location of the targets must be known in advance. This is particularly true in the case of missiles, which cannot search for their objectives. Given the nature of Soviet society and the vastness of Soviet territory, our knowledge of Soviet launching sites is likely to be fragmentary. Moreover, as missile systems become more sophisticated, missiles will be dispersed-at least assuming reasonable prudence on the part of the opponent-thus preventing one attacking missile from destroying more than one on the ground. Also, many missiles will be protected in hardened underground sites. These two factors will greatly magnify the force required for a successful counterforce strike. Other missiles will be mobile or based at sea, creating the need for coördinating many different forms of attack.

The constantly growing complexity of staging a missile attack is bound to reduce the political utility of a counterforce strategy-perhaps even faster than is warranted by purely military considerations. It will grow ever more difficult to convince political leaders that they should rely on a strategy based on intelligence which is fragmentary, using large numbers of weapons for which there is no operational experience in wartime, and in the face of risks which are certain to mount dramatically. Even if the percentage of the opposing strategic force which we are able to destroy remains constant throughout the 1960s-an unwarrantedly optimistic assumption-the absolute number which is left will rise. Thus the Soviet capability to devastate NATO countries in a counterblow will inevitably increase.

To pretend that these factors will not reduce any President's readiness to initiate general war would be sheer irresponsibility on our part and a convenient delusion for some of our allies. Even our best efforts will not prevent the credibility of a counterforce strategy from declining in the years ahead. It will grow ever more difficult to convince the Soviets that such a capability exists or that we are prepared to run the risks it involves. If NATO does not change its strategy as the missile age develops, increasingly direct Soviet challenges will be likely.

There are, of course, those who argue that as long as the nuclear threat is sufficiently grave our hand will never be called. But when the stake is the lives of tens of millions, statesmen cannot gamble on such psychological assessments. It is simply irresponsible to base the future of freedom on the gamble that a threat one is not-or should not be-prepared to implement will never be challenged. And if the threat is implemented except as a last resort, NATO strategy would have brought on a needless catastrophe. Wherever possible, NATO should possess the capability to thwart Communist pressures by means appropriate to the provocation, shifting to the other side the responsibility for escalation.

This is not a peculiarly American problem. The reduced utility of a counterforce strategy for the defense of Europe is not the result of a desire by the United States to become a sanctuary while turning the territory of its allies into a battleground-as is sometimes alleged in Europe. Soviet strategic weapons deployed against Europe far exceed in number those aimed at the United States. A SAC counterforce strike that destroyed the same percentage of both categories of weapons would leave a far greater number directed against our allies than against us. And the European national strategic forces would not alter this situation significantly.

A general war is certain to be more destructive to our European allies than to the United States. Their inhibitions, as they confront a strategy of pure devastation, cannot be less than ours. If they understand their own interests, they should demand, rather than resist, alternatives to the present strategy for the defense of Europe.


Throughout its history NATO has suffered from the difficulty that its strategic doctrine has followed, rather than guided, the creation of its forces. Decisions have been made to meet specific crises, or on the basis of fiscal considerations, or with an eye toward what the traffic would bear politically. Too frequently, more thought has been given to expedients for getting through another NATO conference than to the shape of NATO over a decade.

But whatever the origin of particular force levels, once established they have developed a life of their own. The 30-division goal of 1957 did not make much sense when first advanced-except as an estimate of what our allies might reasonably be expected to contribute. However, that goal furnished a convenient target for the Kennedy Administration when it decided that conventional forces should be increased. As a result, goals which originally were not met because, in terms of traditional NATO doctrine, they were too high may now not be met because they are too low for sustained local defense.

Actually, and fortunately, the size of the military establishment on the Continent-some 22 divisions when the Administration took office-has always been largely inconsistent with NATO strategic theory. Had NATO carried out the implications of the prevalent strategic doctrine, the sole function of the shield forces-the forces deployed on the Continent-would have been to determine that a general Soviet advance had in fact begun. At that point, SAC would have launched its counterforce strike.

It was no accident that the 30 divisions called for in 1957 were never provided. For it did not require 30 divisions to establish that something more than a border incursion was taking place. To be sure, in order to rationalize NATO force goals various other functions were conceived- expressed in such terms as "forward strategy" and "broken-back warfare"-but they were equally inconsistent with basic NATO strategy. In case of war, the crucial factor would not be the units on the ground in Europe, but the effectiveness of SAC. Only if SAC were successful could the Soviet advance be stopped, and then the precise number of divisions would not make much difference.

The size of the military establishment in Europe reflected psychological and not strategic considerations. Our allies have always been uneasy about placing the major responsibility for their defense on an ally 3,000 miles away and with a recent tradition of isolationism. No moral commitment could entirely remove their fear that we might not implement our guarantee. They therefore sought to eliminate or at least to restrict our freedom of decision by committing us to maintain substantial ground forces on the Continent. We could not be expected to do this, however, unless our allies were prepared to make an effort of their own.

Thus there grew up a substantial military establishment on the Continent, even though prevailing NATO doctrine could assign no particular function to it. Current disputes within NATO are largely incomprehensible unless we realize that in the eyes of our allies the shield forces have always been considered a price that had to be paid for the United States nuclear guarantee, a form of mechanical trigger for our strategic forces. Our allies have kept their contribution at a level high enough to induce us to maintain our military establishment on the Continent, but not so large as to provide a real alternative to reliance on a counterforce strategy.

Another consequence of NATO doctrine has been that it has provided a maximum incentive for the creation of national nuclear retaliatory forces. Our doctrine had defined these as the ultimately decisive weapons. In order to reduce dependence on the United States, to gain influence over our actions and for reason of prestige, first Great Britain and then France embarked on the immensely costly and difficult course of building up nuclear strategic striking power.

This was the situation when the present Administration took office. Our strategic doctrine had defined any war in which United States troops engaged Soviet or Chinese Communist forces as general war. This accounted not only for the halfhearted build-up of the shield forces on the Continent, but also for the paucity of strategic reserves in the United States. In any direct confrontation with the major Communist powers, the United States had little choice between acquiescing or launching a counterforce strike.

Since taking office, the Administration has attempted to remedy this situation. However, it has complicated its task by not facing squarely what kind of flexibility it is aiming for. A true capability for "flexible response" would enable us to meet a Soviet challenge at whatever level of violence it might be presented. However, we have justified the build-up of conventional forces in a more technical sense, as providing a capability for a last warning before implementing a counterforce strategy. This argument has tended to defeat its objectives as far as Europe is concerned. An increase in our strategic reserve in the United States by three or four divisions does enhance the credibility of the nuclear deterrent for the defense of, say, Southeast Asia, by giving us the physical means to intervene.

In Europe, however, where we already are physically committed, a few additional divisions take on a different significance. There, our frequent assertions that an increase in the shield forces was designed to make a counterforce strike more likely have been bound to appear as a subterfuge for our reluctance to face nuclear war. The need for flexibility of response in Europe cannot be used to justify a counterforce strategy but to reduce our reliance on it. Until this is openly avowed, the difference between 22 divisions and 30 will seem unimportant. And if the implications of the impending invulnerability of retaliatory forces are squarely faced, the goals of the build-up will have to be redefined.

These considerations are well illustrated by the concepts which we have used to justify the strengthening of the shield forces and by the concerns these have aroused. As propounded by General Norstad and reaffirmed by President Kennedy, the shield forces-presumably the conventional ones, though this has not been made explicit-should be strong enough to enforce a "pause" in military operations in order to permit the Soviets to "appreciate the wider risks involved." The shield forces, that is, are supposed to define a "threshold" below which nuclear weapons would not have to be used.

But is it really true that at the end of a short period of conventional operations the Soviets would confront more clearly the wider risks of nuclear war? If a full-scale conventional attack were to occur, the Soviets presumably would have discounted the threat of our nuclear retaliation. Why should this threat grow more credible during conventional operations, each day of which would shift the local balance against us? Why should our bargaining position be better at the end of the period of 30 days-when our forces would have presumably been decimated-than at the beginning? What if at that point the Soviets offered to negotiate after having gained possession of their prize? If nuclear war was too risky at the start of the conflict, why should it not be even more risky when the local issue has in effect already been decided against us and when the devastation caused by a nuclear exchange may make the local situation seem irrelevant?

If, on the other hand, it is argued that the Soviets will never dare to defeat our shield forces lest this trigger a nuclear exchange, what have we gained by increasing the number of our divisions from 22 to 30? In other words, as long as a nuclear strike remains the chief weapon in our arsenal, the increase in conventional forces in Europe now projected may turn out to be too little or too much: too little for a real local defense, and too much for a counterforce strategy to remain credible.

The same difficulty applies to the concept of a "threshold." This has been defined by the Deputy Secretary of Defense, Roswell Gilpatric, as follows: "The current doctrine is that if NATO forces were about to be overwhelmed by non-nuclear attacks from the bloc countries, NATO would make use of nuclear arms."4 This raises two questions: (1) Who determines whether the NATO conventional forces are being overwhelmed? (2) Assuming this determination is made, how are nuclear weapons then to be used?

With respect to the first question, one can foresee situations in which NATO forces are simply pushed back without being overwhelmed. Will this then be the threshold for nuclear war? What if the Soviets mass clearly overwhelming power in Eastern Europe, but have not yet launched an attack? Or it could be that in an otherwise successful defense individual units of, say, division size would suffer setbacks. Given the dispersal of nuclear weapons in Europe to divisional level and below, it may be that the location of our nuclear stockpiles and delivery vehicles will decide the question of using nuclear weapons for us. This situation may well contribute to deterrence. It is not, however, consistent with an attempt to conduct sustained conventional operations.

Assuming, however, that it is possible to make a clear determination that NATO's conventional forces are being overwhelmed, the question arises how and what nuclear forces are to be used. If we assume that both sides have tactical nuclear weapons in adequate numbers and sizes, it is far from clear that the employment of these weapons would favor the defense at that stage of a battle. With the conventional defense overwhelmed, there would presumably be few shield forces left to stem the tide, even if the attacking units were reduced by nuclear power. After a breakthrough has been achieved, battlefield employment of nuclear weapons may well favor the offense, whose units can by then be dispersed, while the defense has to move its units into predictable areas. In short, if nuclear weapons are to be used tactically at all, it is likely that the optimum, perhaps the only, appropriate moment is early in military operations, or at least while the shield forces are reasonably intact and Soviet reserve forces have not yet appeared on the battlefield.

The ability to reverse a defeat-or an imminent defeat-of the conventional forces would thus depend on the efficacy of a counterforce first strike, and then we are back where we started. If we have a reliable counterforce capability, it will prove as useful when 22 divisions are defeated as when 30 are. If we do not have such a capability, 30 divisions will be insufficient.

Similar considerations bear on the troublesome question of "forward strategy" by which is meant the ability to put up a defense along the Iron Curtain and in particular to protect cities close to that dividing line. It has often been argued that this could be accomplished by a build-up to 30 divisions. If the notion is to prevent a limited coup de main, the argument is valid. If, however, the expectation is to hold the central front along the Iron Curtain in case of a major assault, it is extremely dangerous. For the most perilous deployment of conventional forces is close to the main centers of enemy strength, all the more so as the defense is necessarily spread out, while the aggressor can concentrate. Most conventional battles have been decided not by over-all, but by local superiority-by the ability to concentrate forces at a critical point and then to defeat the enemy in detail. An attempt to hold a line along the Iron Curtain with 30 divisions would repeat the mistake of the Allies in 1940 when they were surrounded in the Low Countries.

Of course, 30 divisions are a more powerful force than 22. However, their major utility is in terms of tasks which we have not defined. Shield forces of this size will greatly increase the capability of NATO to defend such allies as Greece and Turkey against satellite attacks. They will make it easier to respond to pressures against European countries which are not part of NATO, such as Austria or Jugoslavia. They may be especially necessary if a tactical nuclear defense of the central front is envisaged- though this goal seems to have been explicitly excluded.5

As for the central front, 30 divisions will improve NATO's capability to resist minor incursions. And they will enable NATO to offer a somewhat more prolonged resistance to full-scale attack. However, the definition of a minor incursion is crucial. A Soviet coup de main along the central front with, say, five divisions can be resisted by 22 divisions, if not quite as flexibly. A larger Soviet attack would be tantamount to full-scale war on the central front. It is against all reason to suppose that the difference between a Soviet decision to launch a major attack or a minor incursion depends on whether they face 22 divisions or 30. Since the risks of an attack on some 20 divisions would not be much lower than those of a major assault, the Soviet Union would be likely to use whatever force was required to overwhelm the central front as long as that front is inadequately held. In other words, if the NATO conventional build-up is to be anything other than a slight variation of traditional strategy, a basic reassessment of NATO doctrine and force levels is necessary.


NATO will have to face much more explicitly than heretofore the necessity of developing forces for a local defense of Europe. Those Europeans who believe that an emphasis on local defense reduces the credibility of the deterrent are confusing cause and effect. The absence of alternatives to a counterforce strategic strike is certain to create a weakness that invites attack. The creation of these alternatives will demand a greater effort than the relatively minor modifications in doctrine and force levels envisaged thus far.

Two general options are available. The first would seek to strengthen the shield forces and ready reserves so that they could hold, not for 30 days, but long enough for the West's superior potential to make itself felt. The objective of the conventional forces would be to achieve a local stalemate equivalent to the stalemate in strategic forces which made the build-up necessary. These forces would have to be trained in nuclear warfare and they would have to be backed up by a nuclear arsenal to guard against the introduction of nuclear weapons by the Soviet Union. However, while the tactical use of nuclear weapons would not be ruled out, NATO, if it adopted this course, would in effect attempt to match Soviet power at whatever level it is presented, thus completing the spectrum of deterrence. If this option is chosen, the goal of 30 divisions will have to be substantially increased and a reserve structure will have to be created almost from scratch.

This course is not beyond the capabilities of NATO. Western Europe alone is superior to the U.S.S.R. in industrial capacity and even in manpower, and the North Atlantic Community as a whole should find it much easier than the Soviet bloc to maintain a full spectrum of deterrent power. The frequently heard argument that NATO must retain its present strategy because its people are not prepared to make the necessary sacrifices for an effective local defense is surely inconsistent with the hope of maintaining the credibility of a counterforce first strike. Why should an aggressor believe that people and leaders who are not prepared to make the comparatively small sacrifices needed to assure a conventional local defense would be willing to face the enormous devastation implied by general nuclear war?

At the same time, a decision to rely on a largely conventional defense is not without its risks. Unless we act with great delicacy and subtlety, it could give rise to the notion that the West considers any kind of nuclear war unthinkable regardless of provocation. In such conditions, the Soviets could make NATO's conventional build-up irrelevant by threatening to use nuclear weapons against it. Moreover, despite the considerations outlined above, it may be that no Western government, including that of the United States, is willing to make the effort which a heavy emphasis on conventional defense would entail. Failure to meet these force levels would then create a sense of impotence and invite Soviet pressure. Nothing could be more dangerous than to change the doctrine without acquiring the necessary forces to sustain it.

If NATO is not willing to make the effort required for a conventional defense, its other option would be to rely more heavily on tactical nuclear weapons. A tactical nuclear war does not require less manpower at the beginning than a conventional war, and it may even require more. But it is more likely to be fought by ready forces and the use of tactical nuclear weapons should make it possible to prevent massive reinforcement of the battle area.

In this strategy, the conventional forces should be at least strong enough to stop the Soviet forces deployed in Germany and Eastern Europe. This would prevent a sudden coup de main. A decisive Soviet attack would require an advance build-up of troops. If the West's conventional forces were powerful enough so that very considerable Soviet reinforcements were required, many targets for tactical weapons would be presented. The purpose of the conventional forces would be to create optimum conditions for the use of tactical nuclear weapons. Deterrence would be achieved not by protecting against every contingency, but by confronting the Soviets with the prospect of a conflict with incalculable consequences.

The difference between this option and current strategy is threefold: (a) Major reliance would be placed on stopping a Soviet attack without resorting to a counterforce strike, (b) Nuclear weapons would be used not after a pause and when NATO's conventional forces were on the verge of being overwhelmed. Rather, they would be employed early in the operation, as soon as it was clear that a massive Soviet attack was under way. (c) The primary use of nuclear weapons would be in the battle area.6 The most convincing argument on behalf of the 30 division goal is in terms of a tactical nuclear, not a conventional, defense.

At the same time, reliance on tactical nuclear defense is plainly the more uncertain and risky course, even if it is the more attainable. It would require the West to initiate the use of nuclear weapons early in the operations. If NATO hesitated, the Soviets might wear down the defending forces to a point where even the introduction of tactical nuclear weapons could no longer restore the situation.

This strategy would thus enable the Soviets to pose challenges ambiguously and force NATO into an unfavorable war of attrition. The difficulty is compounded by the failure of the West to devise a clear doctrine for tactical nuclear war or to develop forces clearly designed for it. For a variety of reasons, the possibility of limiting a nuclear conflict has been rejected and all nuclear weapons have been stigmatized without distinction as to kind. If these trends continue, it is likely that a strategy relying on the use of even tactical nuclear weapons may involve inhibitions similar to those in a counterforce strategy, bringing about almost equivalent opportunities for Soviet blackmail.

Whatever the course selected, a number of considerations must be kept in mind. First, the choice between conventional and nuclear war is no longer entirely up to the West. Whatever its preferences, NATO has to prepare for the introduction of nuclear weapons by the opponent. Any war will be nuclear, whether or not nuclear weapons are used, in the sense that deployment-even of conventional forces-will have to take place against the backdrop of tactical nuclear weapons, and the risk of escalation, even under conditions of mutual invulnerability, can never be wholly removed.

Second, in deciding on the strength of forces required, there is the danger of too mechanical an approach. Too often, the impression has been created that a sustained conventional defense is within reach simply by assuming that the offense requires a superiority of about 3:1. This assumption, which is strangely reminiscent of certain French theories prior to World War II, cannot, even if valid, be applied so literally. The attacker does not require a three-to-one superiority all along the front. He can win by massing troops suddenly at the decisive point and overwhelming it. In conventional war, victories have very often been won when forces were roughly equal. (Germany, after all, achieved its victories in World War II with inferior forces.)7

A shift of emphasis to local defense will require NATO to address itself with determination to three other problems: the standardization of weapons, the creation of special forces for conventional war and a reconsideration of nuclear deployment. At present, NATO is equipped with several different types of tanks, artillery and small arms. Though the NATO command is integrated, each national force has its own weapons. In case of conflict, the logistics problem may prove insuperable. It is likely that each national force in the integrated command represents a separate capability in terms of both "pause" and "threshold."

Also, with present weapons a conventional battle of any duration may cause insupportable attrition of the nuclear forces. This is a particular difficulty for the Air Force. The equipment required to overcome modern air defense and to make deep penetrations of hostile territory has put a premium on complex, fast aircraft. The cost of these planes almost impels the use of nuclear weapons for economical operation. Moreover, the number of combat aircraft available today is dramatically lower than it was in World War II. In a prolonged conventional war, a substantial part of the aircraft needed for a nuclear back-up might be lost. If the conventional build-up is to be effective, NATO requires the augmentation of its tactical air forces and the development of cheap, simple planes for ground support.

This raises directly the issue of nuclear deployment and control. Nuclear weapons in Europe are now deployed so far forward that any local reverse will create enormous pressures to use them. At the same time, they are so dispersed that while the President undoubtedly controls the decision to fire, discriminating control after the decision to release them will be extremely difficult. This uneasy situation is aggravated by the fact that the term "tactical" in current usage includes any weapons not assigned to the Strategic Air Command. While some of the weapons in Europe are designed for battlefield use, others have essentially retaliatory functions indistinguishable from those of SAC.

Thus, NATO's current deployment creates a built-in pressure toward escalation, not only toward nuclear war but, once that threshold is crossed, toward its most destructive form. Whatever the level of the conventional build-up, a reassessment of the deployment and concept underlying the use of nuclear weapons is essential. Thus NATO has taken only the first tentative steps toward the reappraisal of its strategy. It has somewhat increased the number of its divisions and improved their equipment. But a serious commitment to local defense requires much more drastic efforts.Unless the ambiguities described here are resolved, our actions may be misunderstood by the Soviet Union. Within NATO, it will prove difficult to assert effective political control of military planning. Half-measures may cause NATO to combine the disadvantages of every available option.


The present deployment of NATO's nuclear capability represents an anomaly in terms of current NATO doctrine. Nuclear weapons and delivery systems for tactical use have been placed in Europe in considerable numbers. Yet Deputy Secretary of Defense Gilpatric has rejected the notion of tactical nuclear war: "I, for one, have never believed in a so-called limited nuclear war. I just don't know how you build a limit into it once you start using any kind of a nuclear bang."8 And this view seems to have been reinforced by Secretary McNamara at the NATO ministerial meeting in Athens.9

At the same time, Secretary McNamara has insisted that in case of general war our strategic forces are alone sufficient to perform any nuclear mission: "There is no question but today our strategic retaliatory forces of bombers, missiles and Polaris are fully capable of destroying Soviet targets, even after absorbing an initial nuclear surprise attack. Allowing for losses from an initial enemy attack we calculate that our forces would destroy virtually all Soviet targets without any help from deployed tactical air units or carrier task forces which, of course, have the capability of attacking these targets with nuclear weapons."10

In these terms, the nuclear establishment on the Continent is both dangerous and useless. It is dangerous because it may make inevitable the limited nuclear war both Mr. McNamara and Mr. Gilpatric have declared is certain to escalate. And it is useless because it will merely duplicate attacks on targets already covered by our strategic forces.

The validity of the statements of our highest defense officials depends, of course, on the assumptions one makes about the central problem before NATO, the efficacy of a counterforce strategy. If the opposing strategic forces are extremely vulnerable, any use of nuclear weapons is likely to escalate. On that hypothesis, however, a substantial conventional build-up would be unnecessary just as the nuclear establishment on the Continent would be superfluous. If, as is much more likely, the invulnerability of the opposing retaliatory forces increases, the controlled use of nuclear weapons takes on a different significance.

The President has spoken of the need for alternatives between surrender and general nuclear war. To provide these, a capability for tactical nuclear operations would seem to be essential. Indeed, it is the most useful- perhaps the only meaningful-role for the nuclear weapons based on the Continent.

There are those who argue that the limited employment of nuclear weapons, if judged useful, can be left to the strategic forces. According to this line of reasoning, nuclear war, even if limited, would bring about conditions so confused and fraught with peril that the local situation would be relatively unimportant. This school of thought, therefore, advocates that when the conventional battle seems to be getting out of hand, nuclear weapons be used "demonstratively," as a "shot across the bow."

But what does one hope to demonstrate? Two methods of employing the strategic forces are available: a limited attack on the opposing strategic force or a "punitive" attack on the aggressor's non-military targets. The usefulness of either method is open to serious doubt.

If the opposing strategic force is vulnerable, and if our limited attack is effective, the Soviet response may be a counterblow directed against our cities while the U.S.S.R. still has some strategic weapons remaining. If the opposing retaliatory force is relatively invulnerable, the result may be a tit-for-tat. In that case, the issue may well depend on the outcome of the battle on the central front. In almost any foreseeable situation, an all-out counterforce blow may be less risky than a limited one.

If nuclear weapons are used "punitively," the outcome is again likely to depend on the local situation. After Soviet forces have reached the Rhine, for example, the destruction of Kiev followed by a similar destruction in the West may submerge all other considerations in the desire to end the spiral toward ever greater devastation. The ensuing confusion and panic may well enable the Soviets to consolidate their gains.

The most effective method for employing nuclear weapons in a limited manner appears to be their tactical use to stop a battle and prevent a breakthrough. In the best of circumstances, the confusion attendant on any use of nuclear weapons will be considerable. It will be compounded by every country's lack of experience with them. The use of nuclear weapons to prevent a significant territorial advance by the attacker would have the virtue of observing well-defined geographic limitations. Of all conceivable limitations, this is the easiest to understand and to concert. If a standstill in military operations can be forced, the major objective of the defense will be achieved.

Heretofore, consideration of this mission has been frustrated by a variety of groups: by disciples of traditional air theory who are convinced that any limited employment of nuclear weapons is certain to produce an inconclusive outcome; by pacifists eager to polarize our options between the most extreme alternatives; by advocates of more limited disarmament schemes who want to leave the door open for such measures as a nuclear-free zone in Europe. The lack of a concept and capability for tactical nuclear war-conceived as control of the battlefield-is one of the greatest gaps in the present military posture of the United States as well as of NATO. It should be remedied as part of any reassessment of NATO strategy. Strategic weapons cannot perform this role. They lack the accuracy; they are too powerful; and their attrition in tactical operations would be too great.

A new approach to the nuclear establishment on the Continent should address itself systematically to the role of nuclear weapons in stabilizing a battle front. This is especially required if NATO cannot raise sufficient conventional forces to resist a massive attack, but it is needed, in any case, so that we may know how to deal with nuclear weapons if they are introduced by the Soviets against NATO forces. NATO should not simply seek to replace existing weapons with more modern versions designed for similar missions-such as replacing aircraft with missiles. Rather an effort should be made to reassess the missions first and develop weapons appropriate to the new concept afterwards. The term "tactical use" should be much more rigorously defined than it has been in the past. In current thinking, a tactical nuclear war is distinguishable from a general nuclear war primarily by its geographic limitation-a consideration which can be of no interest to the potential victim. All targets and the weapons appropriate for them will have to be rigorously restudied.

Any tactical employment of nuclear weapons requires discriminating control and careful advance planning. Neither objective is likely to be attained as long as nuclear weapons are such an integral part of every major military unit. In order both to make the conventional build-up effective and to permit a flexible use of nuclear weapons, the atomic arsenal on the Continent should be grouped into a separate command. If a NATO nuclear force is thought necessary, the control of tactical operations would appear to be the most sensible focus for it.11

It is easy to jibe at the "Marquis of Queensbury" rules which limited nuclear war is often said to require. Many argue that a limited nuclear war will automatically escalate because the losing side will always invest more resources to restore the situation. Thus victory is said to be impossible. However, those who ridicule the concept of tactical nuclear war should explain what alternative they propose if, at whatever level of conventional build-up, the Soviets introduce nuclear weapons against our shield forces. If we wish to avoid surrender or a general nuclear war, the tactical use of nuclear weapons may be our only possibility for avoiding defeat. The worst that could happen if nuclear weapons are used tactically is what is certain to happen in case a counterforce strategy is carried out.

At the same time, if it is true that in tactical nuclear war victory is impossible, then deterrence would be enhanced, for aggression could have no purpose. The statement is, however, only partially correct. If the shield forces disintegrate, it may come about that no amount of additional violence can restore the situation. Effective employment of tactical nuclear weapons depends, therefore, on the strength of the shield forces.


Whatever strategy is adopted, NATO will confront the issue of the control of nuclear weapons. For nearly two years, this difficult issue has been the subject of consultation and dispute. France has attempted to build up its own retaliatory force, following a path long since trodden by Great Britain. The Federal Republic of Germany, in the latest formulation of its Defense Minister, has asked for information about the nuclear weapon stockpiles in Germany, for a guarantee that these will not be withdrawn without the consent of NATO, and for a degree of unspecified joint control over weapons fired from German soil.12 The French effort is designed to reduce reliance on the United States' retaliatory force; but the German proposal bears primarily on the issue of tactical nuclear war, for the nuclear weapons on German territory are neither designed nor intended for a retaliatory strategy. Other NATO allies have speculated about some form of joint control designed less to contribute to an affirmative decision to employ nuclear weapons than to provide a veto over their use.

Our own thinking has suffered from the attempt to combine several incompatible courses of action: we would like to head off the multiplication of national nuclear establishments and we are stressing the increased importance of conventional forces; but we would also like to retain the determining voice in the nuclear policy of the alliance. It is, however, against all reason to expect our European allies to integrate their conventional forces in a joint command and to place increased reliance on a conventional defense, while one partner reserves for itself a monopoly on the means of responding to the Soviet nuclear threat and freedom of action in employing nuclear weapons.

From this point of view, the French national effort is not as senseless as is often made out. It is easy enough to demonstrate that national nuclear forces in Europe are ineffective because they cannot withstand a determined surprise attack and to use them independently against the Soviet Union would be to commit suicide. However, the key question is whether these forces are intended to be used separately. Could the Soviet Union ever conduct a surprise attack against the strategic forces of Great Britain or France without running an unacceptable risk of a counterblow from the United States or from the other European ally, temporarily spared? Alternatively, if the strategic forces of one of our European allies were to attack the U.S.S.R. without our consent or perhaps against our wishes, it is of course possible to say petulantly that we would then leave our ally to his fate. However, in the real world, such a course would be extremely unlikely. It is not only that the Soviets might retaliate against us in any case. More important, to "punish" a recalcitrant ally by leaving him at the mercy of a Soviet counterblow would be scarcely less serious for us than for the offender.

If we are honest with ourselves, we will have to admit that one of our objections to the spread of national nuclear forces within NATO is that we do not wish an ally to be in a position to force us into nuclear war. For this very reason, independent national nuclear forces, whatever their military effectiveness, will have the practical consequence of impelling the coördination of policies which President de Gaulle first proposed in his scheme for a Directorate. To be sure, President de Gaulle undoubtedly has other goals in mind as well. The question before NATO, nevertheless, is whether coördination can be achieved without an expensive and wasteful duplication of effort.

Several schemes have been advanced. In May 1961, President Kennedy formally offered to "commit to NATO . . . five-and subsequently still more-Polaris atomic submarines . . . subject to any agreed NATO guidelines on their control and use."13 This pledge was fulfilled by Secretary McNamara at the Athens ministerial meeting when he announced that five Polaris submarines had already been committed to NATO, with more to come.14 Also the President has expressed a willingness to consider "a NATO seaborne force, which would be truly multilateral in ownership and control...."15 And Secretary Rusk has reaffirmed our willingness to participate in consultations to this end. 16

When we assess these proposals and undertakings, however, it is important that their nature be clearly understood. Assigning Polaris submarines to NATO is an important symbol of America's commitment. It is not a device for effective joint control of nuclear weapons. To present it as such only invites confusion.

For what precisely does this commitment to NATO entail? As announced, the submarines remain under exclusive United States control. Does this then mean that there is a stronger obligation to employ the NATO-assigned nuclear forces in the defense of Europe than to employ SAC? The danger of creating an impression that there are degrees of our nuclear commitment to NATO must be weighed against whatever reassurance our allies derive from the act of assigning these submarines to NATO. Then, what is one to make of the phrase in the NATO communiqué at Athens that the United States and Great Britain had given "firm assurances that their strategic forces will continue to provide defense against threats to the alliance beyond the capability of NATO-committed forces to deal with"?17 Does this imply that there are two thresholds in NATO strategy: one between conventional and nuclear war and another between the use of the strategic forces assigned to NATO and the rest of SAC in combination with the British Bomber Command? If so, this alone would spell the end of a counterforce strategy, for such a strategy is beyond the capability of a Polaris force assigned to NATO- present or prospective. If, however, all strategic forces are to act in concert, the wisdom of splitting them into two parts becomes doubtful.

It may be argued that the commitment of Polaris submarines to NATO subjects them to joint guidelines-as indicated in the President's Ottawa speech. But what is the significance of guidelines for a limited number of submarines? The only joint plans that would seem to make strategic sense would involve the total nuclear strength of the alliance, including SAC and the British Bomber Command.

But a more serious question is this: Are these guidelines to be understood as a contingency plan for the employment of nuclear weapons provided the governments concerned agree on using them? Or are they intended to commit NATO to use nuclear weapons in certain specified situations, giving the NATO Commander authority in advance, without further recourse to political directives? If the former is intended, a discussion of guidelines may create as many problems as it solves. A contingency plan of this nature not only includes; it also excludes. It determines not only when nuclear weapons will be used; it also defines when they will not be employed. To attempt to be too specific may thus magnify concerns; failure to be specific may defeat the objectives of some of our allies.

If, on the other hand, the guidelines are to be construed as a directive to the NATO Commander, they would involve an unparalleled abdication of what has been traditionally considered a key element of political control. In any event, the concern of some of our allies does not focus on the deficiency of our planning but-to put it bluntly-on the deficiency of our will. Some of our allies profess to doubt, not our capacity to use nuclear weapons, but our readiness to do so.

Can this concern be met by creating a NATO nuclear force, either through common ownership of a NATO Polaris force, or by devising a multilateral central mechanism for the Polaris submarines which we may assign to NATO? Making NATO into a "fourth nuclear power" is supposed to provide a mechanism for joint planning and a substitute for national nuclear establishments. In practice, it is unlikely to provide either.

A multilateral NATO nuclear force superimposed on the existing framework would bring about three, or perhaps even four, kinds of strategic forces within the alliance: a very large one under exclusive United States control (depending on how the NATO force comes into being, this may be split into two parts, one assigned to SAC, the other "committed" to NATO); two smaller national forces; and, finally, a medium-sized NATO force in which all the countries already possessing national forces would also participate. This plethora of forces will present formidable problems of command and control. As we have seen, it is also likely to signify to the Soviets that there are degrees of intensity of the American nuclear commitment to Europe. Why else would we devise such gradations of retaliatory power?

Moreover, what is to be the purpose of such a multilateral NATO force? Presumably, if our strategic forces already cover every Soviet target without help from forces deployed overseas, as Secretary McNamara has said, it will be superfluous militarily. A multilateral NATO force can no more conduct a counterforce strategy than a small NATO force controlled by the United States. And, whatever its mission, it must be coördinated with the much larger forces under our exclusive control.

If we create a multilateral NATO nuclear force that is militarily senseless, we will not be at the end, but at the beginning, of a process. Once the joint force has been established, it will be next to impossible to head off pressures to turn it into a militarily useful instrument both by increasing its size and by giving our allies a more determining voice than is now envisaged. This may be desirable, but we should at least embark on such a course with our eyes open.

If a NATO force as now conceived is militarily unnecessary, is it a political answer to Europe's concern about our nuclear predominance? This raises squarely the issue of the American veto in the nuclear policy of the alliance.

If we retain a veto, then two types of strategic forces will exist for the defense of NATO: a relatively small one, which would require our consent, together with that of whatever control mechanism NATO devised; and another large force-SAC-which would remain under our exclusive control, though perhaps split into two parts. Our European allies could prevent the use of the smaller force, but they could not compel it. With respect to the larger United States force-from the military point of view, the crucial one-our allies could neither prevent nor compel its use.

In case of disagreement with our European allies, one of two things would happen. If we wished to use nuclear weapons and the Europeans did not, we could still employ SAC (and probably even the Polaris submarines assigned to NATO). If they desired to use nuclear weapons and we did not, both the NATO nuclear force and SAC would be inactive. In other words, a NATO nuclear force with an American veto would multiply safety catches but not triggering devices. It can easily be demonstrated to be a shift of nuclear weapons from one American headquarters to another-as the French have already maintained.

If, on the other hand, we agreed to give up our veto, a major and perhaps insoluble constitutional issue would be raised in the United States. Can a group of foreign powers commit us to nuclear war without our consent? While the present situation is not very different, a great deal may depend on maintaining this nuance. The attempt to eliminate it may evoke a debate casting in doubt the extent of our nuclear commitment to NATO.

If we did give up our veto, would we be committing ourselves to the use of the NATO nuclear force only or to the use of our entire strategic force? In the former case, we would have again raised doubts whether the NATO force represented the limit of our commitment. In the latter case, a separate NATO force would be unnecessary, for we would then be obliged to use all our nuclear forces at the behest of whatever majority had the power of decision in the control body.

The discussion about the veto illustrates that the European concern about nuclear control is not as baseless as is often alleged. If disagreements with our European allies were inconceivable, we could agree to use our nuclear arsenal whenever a certain number of our allies requested it, according to plans mutually agreed upon in advance. Since we are unwilling to do this, it must be that we wish to reserve for ourselves the right to differ with the interpretation of our allies about the nature of vital interests or the means to defend them. Jealous as we are of our prerogatives, we should have understanding, even when we do not agree, of the attempts of some of our allies to achieve a greater voice in determining their fate.

In short, the only kind of NATO force that could meet some of the concerns that gave rise to the quest for multilateral control in the first place is one that involves a modification of the United States' veto. Apart from the domestic problems this would raise, a separate NATO force may not represent the most creative use of such a dramatic American commitment. A NATO force without our veto is a European Atomic Force. If we are prepared to give up our veto, it would be better either to create a control body for all strategic forces within NATO or to support an independent European Atomic Force which would allow us to avoid our constitutional dilemma.

Indeed, the only separate multilateral strategic force within NATO that makes sense would be a European Atomic Force merging the British and French nuclear forces. This would avoid our constitutional dilemma with respect to the veto. It would not create the same doubt about the extent of our nuclear commitment as would a NATO force. The fact that a European Atomic Force could not deal with the whole gamut of nuclear threats would make, not for the dissolution of NATO as is often alleged, but for a partnership in planning between Europe and the United States. The purpose of a European Atomic Force should be to deal with problems of special concern to Europeans. In particular, it could allay the fear that emphasis on local defense would enable the Soviet Union to devastate Europe while it spared the United States. Our support for a European Force would remove the impression that we are striving to keep Europe in a state of dependence with respect to nuclear matters. Such a force is likely to be the result of European integration and could become a device for fostering it.

However, it is important to recognize that at this time neither the United States not Great Britain nor France seems ready to adopt such a course, and it must not be allowed to serve as an excuse for avoiding the issues which rend the alliance. Also, we must take care not to confuse palliatives with remedies and symbolism with reality. Assignment of nuclear submarines to NATO is a token of our commitment; it does not of itself change the existing situation. To be sure, it is highly desirable that our allies have more information about the targeting and planned employment of our nuclear forces. The exchange of information agreed to at the Athens NATO meeting is an important step. It deals, however, with only one horn of the dilemma. The other concerns the issue of the United States veto and the French effort to achieve a nuclear capability of its own.

Indeed, the greatest single obstacle to a reassessment of NATO strategy is the Franco-American difference in nuclear matters. Of course, our skittishness about the veto and our refusal even to reply to President de Gaulle's suggestion for a political Directorate at least partly explains, even if it does not justify, France's lone hand in the nuclear field. In any case, irritation at de Gaulle's tactics should not obscure the fact that, as Bismarck once said, policy is the art of the possible, the science of the relative. Our assessment of the military significance of an independent French nuclear striking force is undoubtedly correct. The question remains whether giving such high priority to thwarting it may not have the ironical consequence of forcing NATO to remain committed to an obsolete strategy.

We will have to choose between seeking to prevent the spread of nuclear weapons within NATO and placing greater emphasis on local defense. A local defense of Europe is unthinkable without a French contribution or French coöperation. The refusal to help France even with projects not directly related to nuclear development (such as guidance systems for airplanes) threatens to turn an erroneous French doctrine into an obsession. By absorbing ever more of France's resources, it also reduces her potential contribution in other respects. As long as France cherishes this doctrine, then even in the unlikely event that her nuclear effort should eventually atrophy, the result in all probability would be not a greater contribution by France to the shield forces, but a sense of impotence leading to neutralism.

The chief drawback to our present policy is that, whatever its theoretical merit, it will not work. It will not prevent the spread of nuclear weapons- not even to France. It will prevent the build-up of local defense forces. If we are to make progress on the primary objective of improving NATO's capacity for local defense, we will have to be prepared-painful as it may be-to reassess our attitude towards France's nuclear ambitions.

However, if we decide to be more forthcoming with respect to the French atomic program-if only by aiding the development of delivery vehicles-we would have a right to insist that the resources thus saved be used to strengthen the conventional shield of NATO. We should make our assistance dependent on the willingness of France, and Great Britain for that matter, to maintain the forces they have promised to the NATO shield. If France accepts, the basis for a much greater and more systematic emphasis on local defense exists. Support for a modest French nuclear force at this stage may also be the best means of bringing about the European Force described earlier. If France refuses these conditions, it will be clear that she places notions of grandeur above the common interests of the alliance and ultimately above her own best interests.


The above analysis indicates that the West confronts the following tasks:

(1) NATO must face the fact that during the 1960s the opposing strategic forces are likely to grow increasingly invulnerable. This impels a substantial shift from a counterforce strategy to some concept of local defense. Unless this issue is explicitly met, NATO is condemned to evasions and a maximum incentive will be created for the multiplication of national retaliatory forces.

(2) NATO must choose between the two available options for local defense: a primary reliance on conventional defense with nuclear forces in support or, alternatively, an emphasis on a tactical nuclear concept. To attempt to combine both-or to act as if NATO has one capability while it really has another-may invite disaster.

(3) Whichever option is selected, NATO must draw the indicated consequences. The corollary to a reliance on conventional defense is a detailed and integrated NATO mobilization plan. The corollary to a tactical nuclear defense is firm political control to ensure compliance with the strategy. In either case the shield forces will have to be increased beyond the 1957 goals.

(4) The choice between conventional and nuclear defense is not entirely the West's to make. Thus it is extremely dangerous to reject the notion of tactical nuclear defense as rigorously as we appear to have done.18

(5) Greater emphasis on local defense highlights the need not only for a conventional build-up but for the ability to conduct discriminating tactical nuclear operations. The nuclear weapons now on the Continent should therefore be given a clear tactical mission-primarily that of controlling the battle situation. A tactical nuclear command should be formed under SACEUR, charged with planning and conducting tactical nuclear operations. Nuclear weapons should be removed from the control of front- line units and placed under this command. The tactical nuclear arsenal would be a much more useful focus for a multilateral NATO command than a small strategic force, since all the weapons relevant to the tactical mission could be assigned to it. To take account of the European wish for a greater role in nuclear planning, the commander of this force could be a European, presumably French.

(6) If it makes sense for NATO to rely on local defense, the implications of this for other areas must be squarely faced. It will simply be untenable to rely on a local defense of the most vital area while depending on a strategy of nuclear retaliation every where else. Thus the mobile strategic reserve in this country will have to be augmented.

(7) With respect to strategic nuclear weapons, NATO should distinguish sharply between expedients to ward off immediate pressures and steps to achieve our long-term purposes. To have meaning, NATO guidelines for strategic forces should extend to all strategic forces of the alliance and not only to a small, some what artificial NATO force. A multilateral control system, if it is to be effective, would make most sense for a European and not a NATO force. As for the French, our best opportunity for constructive policy at this stage lies in seeking to divert them from exclusive preoccupation with their nuclear program, rather than in trying to obstruct it. However, assistance to France and Great Britain in the nuclear and related fields should be made dependent on their fulfilling their commitments to the shield forces, and undertaking not to spread nuclear weapons to other countries without our consent.

(8) The ultimate problem in NATO is not military. It is all very well to proclaim the indivisible interests of the North Atlantic Community of nations in the abstract. But on a whole series of disputes from the Congo to Berlin, the United States and our European allies have seen the matter differently. NATO cannot be welded into a unit militarily until it becomes one politically.

Some of the structural changes recommended by Alastair Buchan may help.19 But the real problem goes deeper. Nothing less is required than a fundamental reassessment of attitudes on both sides of the Atlantic. When NATO was formed, the United States was preponderant politically and strategically. The only security for our European allies was the American guarantee. They were therefore willing to accept most of our proposals without too many quibbles, whatever their real views, for their very act of submission increased our commitment. We have since then fallen into the habit of dealing with our European allies, except Great Britain, almost psycho-therapeutically; we have tended to confuse our periodic expressions of reassurance with the creation of a true partnership, and the muffling of European expressions of concern with meeting the cause of concern. Our Continental allies in turn have shown their disquiet more frequently in the hesitancy with which they have carried out agreed measures than in frankly asserting their disagreement.

In the decade ahead, we must face the fact that our mere enunciation of a policy no longer guarantees that it will be accepted. Henceforth, the Europeans are bound to examine not only the fact of our commitment, but also its nature. At the same time, our allies must accept the responsibility that goes with their new equality. If they merely show petulance, while refraining from suggesting constructive alternatives, they will drain the alliance of dynamism.

In many ways Europe's new self-assertiveness should be a source of pride for the United States. It marks the success of the policies ushered in by the Marshall Plan. Many of the difficulties described in this article arise from the very vitality and strength within the North Atlantic area. Resolving them is a worthy task, for it will liberate energies that could transform the nations of the North Atlantic into a true community. 1 Speech by Secretary of the Army Stahr to the NATO Parliamentarians Conference, November 13, 1961, cited in The New York Times, November 14, 1961. 2 Secretary of Defense McNamara, February 17, 1962. Speech to the Fellows of the American Bar Foundation, Chicago, cited in The New York Times, February 18, 1962. 3 The term "counterforce strategy" can be used in two senses: (a) as a strategy for general war, specifically for dealing with a Soviet attack on our strategic striking forces; (b) as a strategy for dealing with all challenges, including the defense of NATO. The author is of the view that in case of an attack on our retaliatory force a counterforce capability gives us the greatest degree of flexibility and may be an extremely important bargaining device. For almost every other challenge, including the defense of NATO, its utility will, however, decline constantly for the reasons outlined in this article. Whenever the term counterforce strategy is used in this article, it will be in the second sense: as a strategy for responding to challenges other than an attack on our strategic striking forces. The article does not argue against possessing such capability, where possible, but against relying on it too much. 4 The New York Times, June 7, 1961. 5 See reports in The New York Times of Secretary of Defense McNamara's presentation to the NATO Ministerial Meeting at Athens, May 6, 1962. 6 See below p. 530-1. 7 To reduce the danger of a sudden breakthrough, we should reëxamine the utility of building fortifications to protect strategic points on the central front. 8 News Conference, June 6, 1961, cited in The New York Times, June 7, 1961. 9 The New York Times, May 6, 1962. 10 Speech to the Fellows of the American Bar Foundation, loc. cit. 11 See below, p. 540. 12 See Frankfurter Allgemeine Zeitung, April 21, 1962. 13 Speech by President Kennedy at Ottawa, May 17, 1961, text in The New York Times, May 18, 1961. 14 The New York Times, May 6, 1962. 15 Speech by President Kennedy at Ottawa, op. cit. 16 See Press Conference of March 1, reproduced in the Department of State Bulletin, March 19, 1962, p. 456. 17 Text of communiqué, The New York Times, May 7, 1962. 18 See the reports of Secretary of Defense McNamara's presentation to the NATO Ministerial Conference in The New York Times and New York Herald Tribune, May 6, 1962. 19 Foreign Affairs, January 1962.

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