IN the first decade of the nuclear age the predominance of the United States in nuclear weapons steadily declined. History will note the paradox that, when the military advantage that nuclear weapons gave the United States was unchallenged, our considered policy was to restore the conventional strength we had lost in a hasty postwar demobilization and an ill-considered reduction of our standing forces, in order that we might not become wholly dependent upon the new weapons for our defense. But during the later years, when our nuclear advantage was clearly fading, the defense policy of the New Look reversed the emphasis and, without abandoning conventional weapons and forces, placed ever-increasing dependence on nuclear weapons.

The tenor of the new policy was made evident by the pronouncement, on January 12, 1954, of the "doctrine of massive retaliation" to "contain the mighty landpower of the Communist world."[i] This doctrine became manifestly unsatisfactory as the comprehensive strategic basis of our national defense policy as soon as the Russians were able to retaliate massively in their turn. Hence, although "massive" retaliation is still the threat that is supposed to deter an aggressor from all-out attack upon the United States, we now have among our strategies the "graduated deterrence" of "measured" retaliation--the threat that is supposed to deter an aggressor from actions hostile to our interests that are not "worth" total nuclear war. The full range of nuclear threats in support of our foreign policies may be termed "nuclear deterrence."[ii]

Meantime, the President of the United States and the Premier of the U.S.S.R. have both said that total nuclear war would end civilization as we know it--a view confirmed by the scientists who unlocked the secret of the atomic nucleus. The results of the Geneva Conference of 1955, called the "Geneva spirit," have been described as official, though tacit, recognition that war on the traditional total pattern must be excluded from among the instruments of national power. War must be banished from the human community, or it must become something quite different from what it has been in the past.

The necessity is underlined, and the difficulties are multiplied, by the continuing struggle between two great social revolutions in which each side holds that the other is bent upon its destruction. In this conflict the threat of war is explicit. For us the "Geneva spirit" is not enough. We must avoid total war, but without yielding to the pressures of the newer and more aggressive revolution to the East.

Nor is this the only source of conflict in the world. After the tragic events of October and November 1956, it is hardly necessary to discourage hopes that war can, in fact, be banished from the earth, or even that war between the Great Powers can permanently be avoided. The only attainable safeguard seems to be the limitation of war to levels of destruction compatible with civilization.

The popular long-range solution of the problem is to make war (at least big wars) impossible by means of controlled disarmament. But experience tells us that an arms race is more the symptom than the cause of international tensions. The symptom itself is harmful and calls for treatment when treatment is possible, but no one expects the treatment of symptoms to cure the indicated malady. If controlled disarmament becomes possible it will be indicative of a basic change in the struggle, including greatly lessened danger of war. Even so the world will not be guaranteed against the appearance of new causes of tension, of new aggressive forces in the society of nations.

The distinguishing feature of the concept of nuclear deterrence is the belief that by exploiting the power given us by nuclear weapons, by committing ourselves in advance to their employment in any open conflict with Communist imperialism, we can both reduce the chance of conflict and ensure that if it comes it will be a "limited" action in which we can defend our national interests without committing suicide in the debris of our civilization. Combined with the feeling that this is the only way, in any case, that we can guarantee our security without jeopardizing our economic stability, the belief accounts for the increasing emphasis upon nuclear defense so evident during the last four years. This article will examine this belief and the attendant economic rationale. It will inquire into the prospects of warfare in the nuclear age, and, with emphasis upon the requirements of our defense, compare various suggestions for its limitation. The thesis that will be developed is that the best way to limit war is to eliminate the employment of nuclear weapons, and to keep even conventional war limited, and that a defense policy aimed at making this kind of limitation possible is in the best interest of the United States. The relevant facts are all available to the public. The conclusion follows from the way they are put together.


Arguments for nuclear deterrence based upon the assumption that we enjoy a decisive nuclear superiority obscure the essential issue. If we are decisively superior we clearly can deter aggression in nearly all cases, and, if aggression occurs as a result of the aggressor's miscalculation, we can set acceptable limits to the resulting hostilities. But if our superiority is recognized by a potential aggressor, real threats are unlikely to develop except by his misadventure or our lack of firmness in dealing with him. It is notable how much of recent discussion of United States military and foreign policy has been devoted to our alleged lack of determination to use the advantages we are assumed to have in nuclear power, and how little to the question whether we really have such advantages, and what we should do about it if we do not have them, or if, having them temporarily, we are about to lose them.

Sir Winston Churchill, in his address to the House of Commons of March 1, 1955, in which he announced the formulation of a British defense policy for the nuclear age, called the consequences of growing nuclear stockpiles "saturation." He described it as that point at which "although one Power is stronger than the other--perhaps much stronger--both are capable of inflicting crippling or quasi-mortal injury on the other with what they have got." The common American term for Churchill's saturation is "nuclear plenty."

The stage of nuclear plenty has already been reached--or soon will be reached with the growth of the Russian long-range bomber fleet--in the sphere in which an intercontinental nuclear war would be fought. This is the stage at which each side has the weapons and the delivery means to destroy the other, given the existing state of defenses against such an attack.

There appears to be nothing that we can do to avert the loss of our decisive advantage and the advent of nuclear plenty at the strategic level. Building more bombers is not the answer; it matters not how many times over we can destroy the Russians if they can destroy us just once. Nor are technological developments in prospect that give assurance of restoring our advantage. Intercontinental ballistic missiles may reduce the number of weapons lost in delivery. This will merely reduce the stockpile levels required for plenty. Even if we are first to achieve the I.C.B.M. it will not alter the situation so long as the Russians can still retaliate successfully with annihilating attack by manned bombers. Improving our defenses against air attack will raise the cost of attack and the levels of stockpiles and air power at which the stage of nuclear plenty obtains; but no one has been able to show, and few believe, that defenses can be so perfected as to prevent the delivery of the quite small numbers of big bombs required to produce widespread destruction and general paralysis. Defense against the I.C.B.M. has been called possible, but again a virtually perfect defense seems unlikely. All this is not to say, of course, that we should fold our hands. Both the ability to attack and the ability to defend are essential elements in maintaining the "balance of terror." We cannot permit the balance of advantage to slip to the other side; maintaining the strategic stalemate is vital.

A like situation is approaching in every sphere in which nuclear weapons might be employed. A far greater number of "small" bombs will be required to establish nuclear plenty for "tactical" war, but there must be some upper limit to what is needed, and there is no reason to believe that the ceiling exceeds the capacity of the Communist bloc to manufacture, stock and employ that number. Though we still do possess superiority, perhaps even a decisive one, in this sphere, it is likely to be as transient as was our strategic predominance when "massive retaliation" was declared to be the basis of our defense policy.

When both the "strategic" and the "tactical" advantages are lost we shall be faced with nuclear plenty across the board. Only then will our defense problem in the nuclear age fully emerge. It is against this coming that nuclear deterrence must be measured.

Given the qualitative variety of delivery means now available, it is characteristic of nuclear weapons that they immensely increase the flexibility with which military power can be employed. This results from the economy of effort with which nuclear destructive power can be delivered. Any military force that possesses nuclear power in plenty and has the ability to deliver nuclear weapons on all the appropriate targets can increase the power it exerts by easy stages, from any starting point up to the level of total annihilation, anywhere within range of its delivery means; and this increase can be accomplished, unless the action is strongly opposed, with much less effort than would be required to mount a conventional force for a minor "police effort." Mr. Dulles must have had this characteristic in mind when he emphasized the importance of highly mobile sea and air power as the most logical means of delivering deterrent nuclear power in the Pacific.[iii]

When the stage of nuclear plenty is reached, the inherent flexibility of nuclear power is an asset to both parties to a nuclear conflict. The Secretary of State has said that "the essential thing is that a potential aggressor should know in advance that he can and will be made to suffer for his aggression more than he can possibly gain by it."[iv] But if his weapons and delivery means are comparable to ours it will be no easy matter to convince him. We may assume that he wants to avoid total nuclear war; he correctly makes the same assumption about us. The contemplated action then must be limited. Who defines the limits?

"Limitation," in a situation of nuclear plenty, appears to be the pawn in a game of wills. Either side is capable at any time of exerting force of such magnitude that the other side cannot defeat it, or make him "suffer" for it, without extending the existing limits of the action. The outcome of the game of wills must be determined by a balancing of acceptable limits, and the political advantage would appear to lie with the side that is the more willing to risk nuclear war.

The effectiveness of our threat to make the aggressor "suffer from his aggression more than he can possibly gain by it" will depend upon his estimate of the limits we will tolerate compared with the limits he can tolerate. So long as his threshold of tolerance is below what he believes we will tolerate the deterrence works. But let him believe that his limits are above ours and he may not be deterred by our threat of retaliation. If then we take the action we have threatened we shall find ourselves in nuclear war, which, unless some limit is accepted, promises to grow into total war. But the side that first reaches its maximum limit admits defeat. As the defender we might be willing to settle for stalemate. But, just because of its flexibility, once nuclear power is committed it probably cannot be effectively stalemated short of total war, certainly not unless both sides accept the same limits. And for them to be able to do so the limits must be identifiable and stable. This, as will be seen later, is the major shortcoming of nuclear limits--that they are neither identifiable nor stable. In actual nuclear conflict, then, military initiative lies with the side that is the more willing to run the risk of total nuclear war, while the tenuousness of the limits makes the risk of total war very nearly incalculable.


Before we can compare the prospect of limiting war by eliminating the employment of nuclear weapons with the prospect of limiting nuclear war, we must establish the feasibility of conventional war in the nuclear age and dispose of claims that if the United States gives up the employment of nuclear weapons it will be surrendering definite military advantages in the cold war.

The most difficult objection to dispose of is that arising from what is called fate: the bald assertion that nuclear weapons will be used in any future conflict between Powers possessing them simply because they exist. The assertion denies that mankind can control its destiny on the brink of annihilation. If so, it may be true that the nuclear dilemma defies solution. But if nuclear war can be "limited," human prudence is not powerless to control events. Hence it is not necessarily the case that we shall fail, if we undertake to establish as the first limit on future war the rule that nuclear weapons shall not be used.

A different argument is sometimes heard. It states that conventional and nuclear military postures cannot coexist, and that, in a sort of military Gresham's law, the "bad" currency of nuclear force will drive out the "good" currency of conventional force. This might be called the "any fine morning" justification for a commitment to nuclear arms: if we dispose our troops for conventional war, it runs, "any fine morning" the enemy can attack with nuclear weapons and wipe them out, while if we deploy them for nuclear war they cannot fight conventionally.

Of course we can adopt defense policies that will compel us to fight with nuclear weapons if we must fight at all, or we can let ourselves drift into such a situation. We can provide ourselves with forces that cannot fight effectively any other way, we can design our mobilization plans exclusively for nuclear war, and we can deploy our troops in the field in such numbers and in such a manner that they would be certain to be overrun by the enemy if they did not employ nuclear weapons. And all these actions will force our enemy to use nuclear weapons too, if he has them. But these things are matters of choice, not of fate.

There is no denying that all future wars will be fought in the shadow of nuclear power. The fact that three nations now have the means to translate any conflict in which they become engaged into a war of nuclear annihilation will have its effect even upon those conflicts in which they are not engaged. And as other Powers acquire nuclear weapons, the shadow of total nuclear war will become more, not less, foreboding. This well may mean that "forces will have to deploy as if nuclear weapons might be used"[v] in any conflict. Does it follow that they cannot fight conventionally? It is impossible at this time to foresee with clarity the impact of nuclear weapons on future battlefields, whether these weapons are actually being used or just casting their shadow. But a reasonable case for the conduct of conventional war in the nuclear age can be made.

Conventional forces do not have to be massed in the manner of World War II, and it is quite unlikely that they ever again will be. Not only would the belligerents fear "any fine morning," but also those that could do so must support both conventional and nuclear forces and maintain reserves against the prospect of the conflict becoming nuclear. But an even more decisive consideration is the necessity that the conventional action be limited, which in World War II it was not. Unlimited, or total, conventional war in the age of nuclear plenty is a contradiction. Not only would the side that mobilized its resources and deployed its forces for total conventional war invite nuclear attack under circumstances most favorable to its enemy, but also it would be exerting its force to such an extent that its enemy must almost certainly be driven to nuclear retaliation. Total war is designed to win, or to avoid total surrender, and "victory" in the familiar sense is incompatible with war's limitation.

The "any fine morning" argument assumes that tactical spearheads, which are the only combat elements in conventional war that must concentrate sufficiently to make attractive nuclear targets, would be so attractive that the other side could not resist attacking them with nuclear weapons. But if failure to resist temptation meant opening the Pandora's box of nuclear war it is difficult to believe that success in any single battle, except the last extremity, would be worth it. And the last extremity, which involves the survival of the losing side, is ruled out by the nature of limited war.

There is no self-evident incompatibility between a military effort generally so organized and deployed as to minimize losses should the enemy suddenly shift to nuclear weapons, and the temporary grouping of forces in the field in such a way as to attack or to defend successfully with conventional weapons. Limited conventional war in the nuclear shadow is almost certain to be far more open anyway, with smaller forces in contact, and with manœuvre playing a major rôle. Those who state so confidently that troops deployed to minimize the danger from nuclear attack cannot fight conventionally seem to ignore the fact that both sides must be so deployed, because both must equally be prepared for nuclear attack.

Perhaps they are reflecting their feeling that neither side can hope to win in these circumstances, because of the indecisive nature of the military actions that would be possible. But it depends on what they mean by "win." They generally agree that nuclear war can and must be limited, which means that they rule out total victory and unconditional surrender. On the nuclear battlefield, then, if the action is to be limited, either a stalemate must be reached or one side or the other must abandon its effort when the risk of total war becomes excessive in a conflict that is not "worth" it. Conventional warfare in the nuclear shadow means precisely the same, but without the additional risk inherent in the tenuousness of nuclear limitation.

Let us now examine military arguments that urge the advantages to the United States of nuclear defense.

The first maintains, paradoxically enough, that the employment of nuclear weapons in "tactical" warfare gives the defense the same decided advantage that their employment in "strategic" air war gives the offense. It has been said that this is so because, in the history of war, advances in mobility have generally favored the offense, while advances in firepower have generally favored the defense.[vi]

Regardless of its historical validity, the rule is difficult to apply in the case at hand. The mobility of aircraft is combined with the firepower of nuclear weapons to give the strategic stalemate its character. It is not clear which of the two advances is determining. On the ground, firepower that cannot be freely moved and deployed may aid the defense to the disadvantage of the attacker. This is apparently what the machine gun did during the early years of World War I. But mount the machine gun on the tank and you have the mobile firepower to break the stalemate on the battlefield. Planes, missiles and even ground vehicles can deliver nuclear weapons on future battlefields. This looks like more mobile firepower. The destructiveness of nuclear weapons might eventually lead to heavy dependence on deep concrete emplacements for survival on the battlefield and this would certainly stagnate the action. But mobility is an alternative, particularly if the action is to be limited.

Another version of the case that nuclear weapons favor the defense rests upon a tactical forecast. In this view, a defensive force armed with nuclear weapons will be able to use its nuclear firepower to prevent the enemy from concentrating for attack; hence, provided it has sufficient strength to guard its front against enemy infiltration, it cannot be attacked. Again the result would be stalemate on the battlefield. The argument must assume that the attacker will not be using nuclear weapons.

Given nuclear plenty, the attacker will use his nuclear firepower against the defender's nuclear weapons at the point of attack. When he has neutralized them he can concentrate to overrun the defender's screen of troops. There seems to be no reason to question that the attacker can blast his way through any defense, provided he has the means and is willing to pay the price. The balance of the opposing forces in weapons, training, morale and leadership, and the "hardness" of the defense, that is, the extent to which there are underground fortifications, will doubtless play something like their familiar rôles, allowance being made for the high probability that percentage casualties on both sides will be far greater than in recent wars.

It is possible that space limitation may play a new rôle on nuclear battlefields. When dispersal is essential to survival there may be an upper limit to the number of nuclear weapons that either side can employ, a limit set not by stockpiles or numbers of troops but by the ability of their delivery means to survive nuclear attack. If so, the decisive factor may not be the number of weapons used on both sides but the ratio of attacking weapons to delivery means being attacked. The ratio will favor the offense only if the defense occupies a smaller area. The consequence of a ratio unfavorable to either side would be a tendency to raise the limits of the area of conflict in order that more weapons can be used or to increase the chance of survival of delivery means, which comes to the same thing. Either side may open another theatre or attack the sources of the enemy's nuclear power outside the theatre. The prospect illustrates the difficulty, given the flexibility of nuclear power, of limiting a nuclear action. If there is no critical area limitation, as, for example, might be the case in Western Europe, it seems evident that whatever advantage one side may gain by using nuclear weapons can promptly be topped by the other side by using more of them, using bigger ones or using them on targets not previously attacked.

Another military argument for committing ourselves to the employment of nuclear weapons is the so-called fire brigade theory. One of our problems is how we can honor our treaty obligations to the small Powers on the Communist periphery. The earlier notion that we could do this merely by using our strategic or naval air to deliver nuclear bombs on the aggressor has been found wanting. Despite the Secretary of State's claim that such action could be taken without endangering "unrelated civilian centers," nuclear bombing of this sort resembles total war too closely for comfort. What happens, for example, to "related" civilian centers? And as for offering the ground forces of our small allies close support with nuclear weapons from the air, we know too well the difficulty of providing effective conventional air support to our own forces to be hopeful of the results. With nuclear weapons only a few mishaps could be disastrous for the defenders.

The proposed solution, then, is to provide highly mobile ground forces, armed with nuclear weapons in the "tactical" sizes, that can be air-lifted anywhere in the world on short notice. These "fire brigades" would be sent in to support the armed forces of our allies threatened with attack, or under attack, by the superior ground forces of the Communist aggressor. Because of the tremendous firepower inherent in the employment of nuclear weapons, such forces (according to the theory) would be the equivalent of conventionally armed forces many times their size. And so they would be.

The theory is sound as long as we alone are able to organize and deploy such forces, or as long as the aggressor has no effective defense or counterattack capability against them. But while that condition prevails it is doubtful that we need them except in token strength to prove what we can do. The problem of defending our allies does not become an acute one until the aggressor believes that an attack on them might succeed. And he will not believe it until he, too, has the weapons and delivery means to mount such forces, or to counter them.

The fire brigade theory is another application of the notion that we can defeat aggression with "limits" to the action that suit our convenience; actually the plentiful possession of nuclear weapons and means of delivery permits the aggressor to expand the scope of the action to limits that suit his convenience. It does not follow that we should not develop such forces; there is an imperative necessity that we be able to apply our nuclear power flexibly and promptly if forced to do so. It does follow, however, that the provision of such forces is no panacea for the defense problems arising from our treaty obligations in the age of nuclear plenty.

It appears, then, that all claims that dependence upon nuclear weapons will yield us a return that we cannot afford to give up, despite the risks, rest upon some advantage that we are supposed to have over the aggressor that we cannot exploit without using nuclear weapons. But upon examination these advantages turn out to be questionable or, at best, temporary. In neither case do they justify the greater risk of total nuclear war implicit in reliance on them, unless there are better grounds for hope that nuclear war can be limited than have been found so far.


The alternative proposals for limitation of war in military terms may now be examined and compared.

Advocates of nuclear deterrence commonly rest their case that nuclear war can be limited on the possibility of confining the employment of nuclear weapons to "tactical" uses of the smaller sizes. Though one of them has said that "there exists no way to define a limited war on purely military terms" because "wars can be limited only by political decisions, by defining objectives which do not threaten the survival of the enemy," he adds that "the possibility of keeping a limited nuclear war limited depends on our ability to extend the range of low-yield weapons of a kiloton and below, and to devise tactics for their utilization on the battlefield."[vii]

It should be added that "military necessity" is not a "purely military" finding. It was not a decision made by the military that committed us to unrestricted submarine warfare in World War II, nor to the massive "strategic" bombing of enemy cities, nor to the use of the atom bomb on Hiroshima and Nagasaki. It will be said that without these actions the war would have dragged on for years, at the cost of many more American lives. This objection illustrates precisely what is meant by military necessity.

Under the dictates of military necessity the attempt to bar "strategic" employments of nuclear weapons would only have the result that any uses that are deemed necessary are regarded as "tactical." This is illustrated when the proponents of this particular limitation say that urban areas in the immediate "battle area" would be exempt from immunity as "strategic" targets, and that the battle area might be 50 miles deep. But why just 50 miles? Why not 100 miles, or 1,500--the range that identifies the intermediate range ballistic missile, the I.R.B.M.? Would the enemy's air bases or his missile launching sites be exempt? Would he exempt our ports? Certainly the combat zone, even excluding attack upon general reserves and war "potential," will be hundreds of miles deep. If we wanted to ensure that our ports would be spared, for example, in the defense of Western Europe, we should have to grant immunity to many targets we considered "tactical." Military necessity would dictate that we choose limits in terms of advantage and disadvantage rather than for clarity and stability. It is questionable whether we could resist this necessity, and quite unlikely that the enemy would long tolerate our enjoying an advantage.

Efforts to limit the employment of nuclear weapons in terms of explosive power would run into similar conflicts with military necessity. But there is an additional difficulty, that as the gradations are potentially continuous there are no practical cutoff points. Imagine our trying to maintain a limit at, say, 50 kilotons or less in the face of the claims from our men in the field that the enemy was using weapons of 100 kilotons or more, a claim that could neither be confirmed nor confuted without extended and time-consuming scientific detection and analysis. Limitation below megaton sizes, to reduce the fallout hazard and limit indiscriminate killing of civilians, might survive until their explosive power, and fallout, became a military necessity "to save American (or Russian or British) lives."

Small bombs are not being developed for humanitarian reasons, but for the military reason that in the employments for which they are designed they are more, not less, destructive. They can be used in the proximity of our own troops, where the enemy must concentrate his forces for attack. But not all "tactical" targets would be close to our own troops, and those that were farther back, that were "harder" or whose location was not precisely known, could only economically be attacked with more powerful weapons. It will be objected that even small nuclear weapons are large by comparison with World War II T.N.T. weapons and that only a comparatively few of them would be required to do the job of a larger bomb. This is true, but it does not eliminate the problem of economy in lives and delivery means, which are likely to be the really scarce articles in nuclear war. Considerations of economy alone would dictate the use of thermonuclear bombs against certain "tactical" targets.

But the critical consideration, for purposes of this analysis, is the difficulty of making any nuclear limit work. Even if the risk of thermonuclear war inclined men to want to make the limits work, the odds would be against them. Military necessity, given the flexibility and the awful destructiveness of nuclear weapons, would place them in the grip of forces they could probably not control. To permit the enemy to enjoy the advantage that he would gain by broadening the scope of the conflict, or raising its limits, perhaps even for a single day, could be disastrous. By the same token the temptation to seize such an advantage might be irresistible. And at the same time, in the general near-chaos of the nuclear battlefield it would be virtually impossible to know what the existing limits were and whether they were actually being observed or violated.

In contrast to the fatal uncertainty of nuclear limitation, the limitation of war by eliminating the use of nuclear weapons has a decisive advantage. It is one of practicality. Because of the sharpness of the limit there could be no marginal, no hidden, no unwitting violations. The action would be slower, less chaotic; judgment would have a better chance to function. One side might still win in a day if it could have a day's exclusive use of nuclear weapons. But neither side could hope to get away with it, because its offense would be known instantly, and it would know what retaliation to expect. In these circumstances it is possible to believe that a viable limit might be set to military necessity.

It has already been said that there would have to be limits on conventional action as well, to bolster the distinction between conventional and nuclear destruction and to avoid rendering the loser so desperate that he will resort to nuclear weapons in defiance of common sense. Conventional war can approach uncomfortably close to the destructiveness of nuclear war in special circumstances, such as our fire-bombing of Japanese cities. These additional limits would be subject to marginal errors and violations, but the urgency to retaliate would be of an order quite different from that following the violation of a nuclear limit, and the interest of both sides in maintaining the ban on nuclear weapons would dispose them to avoid ambiguous action and to curb their retaliatory impulses in cases of apparent violation. Conventional war in the age of nuclear plenty both must and can be limited; it is very unlikely that nuclear war can be.


The military arguments for nuclear deterrence do not exhaust the case for it. They may, in fact, have played a minor rôle in the change in our defense policy that was mentioned at the outset. Arguments derived from economic considerations have been more prominent, though of course they cannot be considered in isolation from the military prospect.

Mr. Dulles said, when he proclaimed massive retaliation, "We want, for ourselves and for the other free nations, a maximum deterrent at bearable cost." For him and for most of the advocates of nuclear deterrence, costs are bearable only if nuclear weapons are to be employed.

It is characteristic of New Look defense planning that nuclear weapons are thought of as labor-saving devices. We have made a good thing of such devices, from hand tools to automation. They have enormously increased the productivity and hence the value of our workers. Why then should we take one of them off the production line, where he is worth perhaps $6,000 a year to our gross national product, and put him in the battle line against a Russian or a Chinese whose presence at the front is costing Russian production little and Chinese production nothing? If we work with machines, why not fight with machines, the ultimate war machine being, of course, the nuclear weapon?

The argument would be quite persuasive if the Russians and Chinese had no nuclear weapons. No one has ever doubted that a small nuclear force could defeat a far larger non-nuclear force. But this is not the problem. If nuclear weapons reduce military manpower requirements, in some real terms, the effect must be roughly the same on both sides. On whether they do or not, the experts have not yet decided. The problem is in good part one of definition, because if using nuclear weapons enables a military force to produce greater destructive power with fewer men, using more nuclear weapons with the same number of men will produce even greater destructive power. It has already been said that the density of nuclear battlefields will be lower, but it does not follow that superior numbers will not be an advantage to the side that can provide them, equip them with nuclear weapons and use them. Even when the battle area is restricted there must be a high proportion of replacements in being somewhere. By proposing to use nuclear weapons we do not dispose of the need to determine how many forces we should maintain, and that determination must be influenced by the number of nuclear forces the prospective enemy has or can have.

In addition, the argument based on our productive superiority breaks down to the extent that the Russians, and eventually the Chinese, too, by arming their forces with nuclear weapons, are enabled at a single stride to overcome our tremendous production advantage. The Red soldier may still be worth only a pittance in terms of his potential contribution to the national product, but his nuclear weapons are just as deadly as ours, and not many more of him are needed to deliver them. Emphasis upon nuclear warfare, in fact, to the degree that it enables the Communists to concentrate their scarce resources in a limited area of production, appears to be tantamount to abandoning just that much of the advantage given us by our general productive superiority. For this reason a war fought with complex and expensive conventional weapons, but with the prospect of a far smaller loss of valuable lives than in nuclear war--it being assumed that the war must be limited in either case--would appear to be the better bargain for us.[viii]

There are ways of using our productive superiority and our technical ingenuity to save manpower, and lives, in conventional war. We have proved this in recent wars. The lesson of Korea in this regard has often been overlooked, for most critics have been preoccupied with visions of what might have been accomplished if we had raised the limits of that action. The lesson is that the forces on our side consistently opposed forces two to four times their number. This was true not only of American and British Commonwealth troops, but also, during the later period, after they had been trained and equipped, of Korean troops. During the last year of the action we proved that we could actually mount effective offensives against superior numbers. We used our productive capacity and our technical ingenuity to compensate for the inferiority of our numerical strength. And the forces opposing us were not just peasant masses, they were the best that the North Koreans and Chinese could put in the field with massive Russian support. Furthermore, we accomplished this when much of the product of our rearmament effort was going into our strategic reserves and to our European allies against the possibility that the war might become general.

Even so Korea was fought on the pattern of World War II. If it was the first war of the atomic age, in which the limits of the action were imposed by the existence of nuclear weapons, it was also possibly the last war fought without full realization of the impact of nuclear weapons. The nature of conventional war in the nuclear age is no more evident than the details of nuclear war itself. It seems reasonable to assume, however, that conventional combat would be hardly less dispersed than nuclear combat, that mobile firepower will play the major rôle, hence that our ingenuity and productive skill will be fully challenged to supply our forces with the latest tools of war. Air transport, air supply and air support are obvious fields for further exploration. Missile technology has possibilities that are not restricted to the destruction of cities with hydrogen bombs. Communications and command problems of extraordinary difficulty and complexity will tax our abilities, but they will be taxing abilities in which we are superior to our enemies.

The power to inflict destruction cheaply is not the same thing as the best or even the most economical defense. It is true that a handful of men on both sides can most economically wipe out civilization if that is the object. The cheapest way to do it at the moment, despite the increasing cost of modern aircraft, is with strategic bombers and thermonuclear bombs. In a few years long-range missiles will make for even greater economy. But this economy is not real because it is not measured by something that is desired. What we desire is a defense that does not lead to total war.

For us, of course, there is no such thing these days as a force committed only to conventional weapons. The alternative is between a preclusive commitment to nuclear combat and the maintenance of our ability to fight nuclear or conventional war. The latter does require extra effort and extra expense; there is no disputing the saving inherent in developing a single capability, whether nuclear or conventional. But the additional cost--perhaps largely in the maintenance of artillery weapons, troops and supply--seems a small price to pay for the improved chance of our ultimate survival if war can be kept conventional.

It cannot be said with certainty, then, that the number of troops we should maintain is importantly different whether the prospect we face is limited nuclear war that we hope to keep limited or conventional war that we hope to keep conventional and limited. It will depend, as always heretofore, upon a variety of circumstances, many of which have their origin outside the field of military competence.


The reader must of course ask whether our cold-war enemies will benefit if we decide that a ban on nuclear weapons is essential to the limits that must be imposed on war. Will it only incline them to use their nuclear weapons, or will it enable them to exploit their ground-forces advantage in conventional war? The answer goes back to the first assumption--total nuclear war is an imminent and deadly peril to all mankind. It plays no favorites, and only the nuclear Powers can do anything about it. We and the Communists are forced to collaborate in the search for an effective limit to war, regardless of other differences. It is reasonable to expect them to follow the same line of speculation that we have followed and to come to the same conclusions. It would, in particular, be grossly shortsighted of them to assume that nuclear plenty leaves their conventional masses in possession of the field. These masses are appropriate only to unlimited conventional war, and the only unlimited war in the nuclear age must be nuclear.

Each side in the great conflict must feel its way. If they must recognize that conventional force cannot be used without limit, we must abandon our preference for crusades, our feeling that wars are only justified if they end in total victory over a hated enemy, that limited wars are "phony." Major emphasis upon avoiding war is justified because the risks of total war are so much better known than the effectiveness of the steps that can be taken to avoid them. But we cannot treat "peace" as an end in itself, and they are not likely to do so. "Peace" in this sense would tend to petrify the status quo and bottle up the dynamic forces loose in the world. An eventual explosion would be inevitable, upon which it might be impossible to impose limits. Limited war may be regarded as a safety valve to prevent such an explosion. It will fail if overloaded--if, for example, the Communists try to use their manpower advantage to gain their ends under protection of the nuclear ban, or if we lose patience and try to use war to effect the kind of epochal resolution of conflict that we have expected of great wars in the past.

All these things we see darkly. But our vision is needlessly obscured by the two related false notions: one, that nuclear power is just another weapons development, like gunpowder, that can be used to strengthen the arsenal of freedom; and the other, that a comfortable nuclear lead is the guarantee of our security. Nuclear destruction is not like gunpowder, which gradually changed the military art over centuries and had negligible effects on the balance of power among Western nations. Nuclear power, in little more than a decade, has introduced an entirely new dimension into human affairs--the dimension of mutual annihilation. And our nuclear lead is perishable. We cannot much longer avoid facing the nuclear age.

The prospect is disturbing, particularly to those who have thought that we could depend upon our nuclear advantages. It was not in the cards that we should owe our security to divine favor. The future counsels prudence, but not faintheartedness. While using every opportunity to reduce international tensions and to extend the reign of order among nations, we must work positively for the limitation of war. To this end we must exert ourselves to the utmost in the technological competition to prevent the balance of advantage from shifting to the other side, and we must make it quite clear that we are prepared to risk annihilation itself to prevent Communist conquest by default, either by threat of nuclear terror or by conventional arms under cover of the nuclear ban. We must, in short, guarantee that only effectively limited hostilities can be rationally undertaken.

Moreover, we must be prepared to fight limited actions ourselves. Otherwise we shall have made no advance beyond "massive retaliation," which tied our hands in conflicts involving less than our survival. And we must be prepared to lose limited actions. No limitation could survive our disposition to elevate every conflict in which our interests are affected to the level of total conflict with survival at stake. Armed conflict can be limited only if aimed at limited objectives and fought with limited means. If we or our enemy relax the limits on either objectives or means, survival will be at stake, whether the issue is worth it or not. But saying that we must be prepared to lose does not mean that we shall lose, particularly in the long run. Our strengths are many, not least the fact that our revolution offers a better promise to mankind than the Communist alternative.

[i] Speech by Secretary of State John Foster Dulles before the Council on Foreign Relations, New York, January 12, 1954.

[ii]Cf. John Foster Dulles, "Policy for Security and Peace," Foreign Affairs, April 1954. Also Henry A. Kissinger, "Force and Diplomacy in the Nuclear Age," Foreign Affairs, April 1956.

[iii] Remarks at a press conference, December 21, 1954.

[iv]Foreign Affairs, op. cit., p. 358.

[v] Kissinger, op. cit., p. 356.

[vi] Paul Nitze, "Atoms, Strategy and Policy," Foreign Affairs, January 1956.

[vii] Kissinger, op. cit., p. 357, 361.

[viii] See "Military Policy and National Security," W. W. Kaufmann, ed. (Princeton: Princeton University Press), 1956, p. 250.

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
  • JAMES E. KING, JR., of The Johns Hopkins University; Executive Secretary, Office of the U.S. High Commissioner for Germany, 1949-52
  • More By James E. King Jr.