THE hard core of the West's military strength is its retaliatory forces. Their most powerful single element is the United States Strategic Air Command. . . .

In spite of the retaliatory forces, there remains the possibility that war could start because the enemy made an error in judgment or took some reckless or opportunistic action. Because of this chance--rather, mischance--we cannot omit or skimp at any point along our eastern frontiers the defense strength which the Soviet threat dictates. For if our line is not defended throughout, the enemy might trump up a pretext for crossing it. We would then face not only an accomplished fact, but also a dilemma: If we did not take immediate action, we would fail to meet the commitments of the Alliance; if we did take it, we would start a war.

On the other hand, if our line is being held in reasonable strength, and if the enemy knows this beyond doubt, then any inclination on his part to cross the line makes him face the terrible decision of detonating World War III, with a sure prospect of his own annihilation. The defensive forces deployed on our eastern boundary thus become an essential part of the deterrent.

--General Lauris Norstad, Supreme Allied Commander, Europe

IF one looks only at the second paragraph above, the case for sizable NATO ground forces seems to be based on a very understandable reluctance to initiate all-out atomic war: we need forces tailored for wars limited in scope and in the weapons employed. Then we can meet limited provocations without starting all-out atomic war.

Yet if it appears that the United States Strategic Air Command (SAC) will not necessarily be unleashed against the Soviets in response to enemy attacks in Europe, the enemy becomes less worried about initiating such an attack. Our main deterrent is weakened. Acknowledging this as they must, proponents of "graduated deterrence" none the less prefer to accept some weakening of the big deterrent in order to lessen the probability that any hostilities will degenerate into total war. NATO military commanders naturally want the best of both worlds: no weakening of the supreme air-atomic threat against the Soviets, and yet sizable land forces in case aggression does occur. But we cannot have the best of both worlds. One kind of military preparation diverts resources from another. Should the NATO compromise between land and other forces follow the new British pattern, with our ability to fight an old-fashioned ground war in Europe being further reduced?

There are many reasons for maintaining a sizable ground-war capability in Europe, but the conventional ones, premised on a strategy of total war, are suspect.

First, why maintain land forces to serve as a "trip-wire" to trigger retaliatory air blows? If the line between the NATO Powers and the Soviet Bloc is sharply defined, we and the Soviets will know when it is crossed. If they believe that crossing the line will unleash our full atomic capability, only one course of military action is sensible for them. If they attack at all, they must try first to eliminate our retaliatory airpower. Otherwise the aggressor loses any chance of destroying our air forces on the ground, so that many more bombers fly against him, greatly increasing the probable damage to his homeland. We may then destroy much of the Soviet strategic air force on the ground, limiting the size and disrupting the planning of any following Soviet air strike. Furthermore, any smaller, less well-organized Soviet strike that follows will be much less able to cope with our alerted air defenses. Enemy losses are multiplied at each stage, compounding the supreme risk he runs in foregoing the strategic air initiative. Consequently the much-touted "tripwire" function of NATO land forces is highly unlikely to be fulfilled whether our troops are many or few in number.

This simple argument is based, of course, upon two premises: (1) the enemy believes that SAC will attack in response to aggressions in Europe, and (2) the "line" between East and West is sharply defined. Altering either premise changes the strategic argument, and both must be altered to be realistic. The extent to which our extreme threat is believed in by the enemy is likely to vary widely, depending upon what we do and what we say we will do. Our utterances are conflicting enough. Nowhere in the Western Alliance are there more than pitifully ineffective civil defenses against atomic bombing--a condition which hardly contributes to the credibility of our threats of massive retaliation. And, especially relevant to this discussion, NATO offensive military forces have not as yet been reorganized to fit a massive retaliation strategy, despite all our talk about doing so. How is the enemy to interpret the lag in revamping our forces?

The more we strip our conventional land defenses, the more credence the enemy is likely to place in our expressed willingness to use, by default, the strategic airpower that is our remaining alternative, whereas the more we emphasize the continued desirability of sizable land forces, the more he is likely to suspect that we regard them as an alternative to using strategic airpower, thus reducing his belief in our intent to use the main deterrent. One can escape this dilemma only by establishing other strongly complementary uses (as distinct from alternative uses) for strategic nuclear airpower and conventional land forces when both are to be employed in total war; and the case for this is vanishing.

Beyond serving as a "trip-wire," land forces are often said to complement airpower by protecting air bases. This was notably the case in World War II when our airplanes were relatively "short-legged" and a bomb packed but one-millionth the explosive power that it can today. Then, many repetitive mass air raids were required to cause extensive damage to the enemy economy as a whole. Now, if we get in one or only a very few mass sorties, we can probably achieve devastation on a scale that dwarfs the total destruction of World War II. Many sorties per month from constantly operating overseas bases are consequently dispensable luxuries--luxuries we are highly unlikely to enjoy anyway, whether the bases are attacked by surface forces or not. Unless the enemy air force is destroyed at the outset--the "pie in the sky" case where we win overnight--the soft portions of our overseas bases, including parked aircraft on the ground, are obviously vulnerable to enemy atomic air attack. Also, of course, we must expect many of our bombers to be lost to the enemy's air defenses on our first strikes. We shall lack the wherewithal to mount many sorties, although our first strike alone may be deadly. Our system of overseas bases is a tremendous contributor to the efficacy of our big deterrent, but its utility is not dependent upon prolonged protection against surface attacks. The vital complementary rôle for surface forces lies in protecting our air bases against sabotage or surprise assault from the air, not mass ground invasion, in order that we may have a good chance of mounting one or two mass attacks.

II

Both the "trip-wire" and base-protector arguments for land forces lose most of their appeal when examined realistically in the context of fighting modern total war. But, of course, much current discussion about the utility of military preparedness properly emphasizes deterring wars, not winning them. A strong NATO land force in Western Europe certainly helps inhibit the Soviets from launching total war. The important question is, by how much? Its gross contribution to deterrence may not be great, while its costs detract from the power of the main airpower deterrent so much that the net contribution of land forces to deterrence is negative.

One can imagine a land force so strong and securely protected that it offers the enemy a high probability of having his territory occupied if he is rash enough to attack. Such a force would be a tremendous deterrent. But our army planners have never sought nearly enough trained divisions to fit a calculated offensive strategy of overmatching the 175-odd divisions of the Soviet Union. Their requirements were designed from the outset to fit a much more modest, although still formidable, objective. And in NATO there has been, as we know, an always sizable "gap" between the resources called for even by these requirements and the lesser resources actually made available. Consequently the Soviet leaders cannot suppose that NATO's strength on the ground offers them much of a threat. In contrast, they must regard as feasible the prospect of their society being devastated by an air-atomic assault. This is the main deterrent, a deterrent which depends little upon the size of NATO land forces.

Some believe that NATO land forces help deter total war merely because of their defensive capability. Do not such forces prevent the Soviets from quickly conquering the economic riches of Western Europe in compensation for a Russian economy devastated from the air? To this one can only reply: compensation to whom? How many Russians would survive to inherit the riches? Furthermore, getting intact Western factories in return for smashed Eastern factories is by no means equivalent to acquiring an entire substitute economy. The argument that the Russians would regard an occupied Western Europe as a good trade for a devastated Russia is simply incredible.

Thus if we are really committed to a total war response to any Soviet aggression against NATO Powers, even large numbers of land troops can add little to the deterrent power of our retaliatory air forces. Resources diverted to enhancing our retaliatory airpower then promise a stronger deterrent, especially when the deficiencies in the main element of our airpower, publicly dramatized by the recent Symington Committee hearings, are noted. To repeat an old but terribly important point, there can be no logical argument against spending enough on protecting our strategic air forces so that they are not obviously vulnerable to enemy attack. Our posture must preclude this extreme kind of deadly speculation by the enemy: "Why does he have so threateningly big but so unprotected a striking force? Surely such a combination is only rational if he is planning to strike first. If this is the case, should I not forestall his move, especially when my chances of destroying his air forces on the ground look so good?" Protecting SAC is imperative but not easy. We cannot even now count on the luxury of hours of radar warning, and when the enemy has operational intercontinental ballistic missiles our maximum warning will be measured in minutes.

There is a lot of uninformed talk about automatic self-protection for SAC because, though concentrated on just a few bases, "it is always ready to go in force almost instantaneously." The forthright statements by our Air Force officers make clear that this is not the case. To be always ready to go in force requires ready planes and stand-by crews, which in turn implies several crews per airplane so that a full shift is always available. As it is, we have less than one crew per aircraft. To get extra crews would mean many more training hours in flight, probably more planes, and certainly a greatly increased burden on already strained maintenance resources. The acute shortage of skilled personnel for maintenance and flying has been identified by General LeMay as SAC's most critical deficiency. To protect our strategic force adequately is bound to involve some combination of a greatly increased state of around-the-clock alertness, dispersal of vulnerably located aircraft, hardening of vital base elements against bomb damage and increased provision for warning of enemy air attack. A judicious combination of such necessarily expensive measures is our most imperative need if we seek to deter the enemy.

In consequence, the NATO strategist who wants the United States to support more land forces in NATO faces more, not less, budgetary competition from our Strategic Air Command. Funds must be provided both to protect SAC much more adequately against a growing enemy threat and to complement SAC bombers with guided missiles.

Even if one grants the force of these arguments that belittle the contributions of land defenses either in deterring or fighting total war, there remains another to be considered. Truly effective land forces promise a bonus to threatened Europeans beyond any contribution to deterrence. If adequate in number to implement the celebrated "forward strategy" of stopping a Soviet invasion, such forces can prevent the feared sequence of fighting on Western European territory, Soviet occupation, and then fighting again. To mitigate disaster is obviously important if you have failed to avert it by a policy of deterrence, and a preparedness policy which promises to do so offers an important asset. But how important? Is there a case for diverting resources to land defenses merely to mitigate disaster?

Suppose NATO land forces were adequate to contain a Soviet invasion. If the Soviets initiate total war against NATO countries by massive air assault, how can immediate casualties be held below the tens of millions in either the United States or Europe? And even if Western defenses are greatly improved, there is very little chance that casualties would be less in the future because of the steadily growing Soviet air threat. Consequently, no assured protection against catastrophe in NATO countries is possible however strong the defense ground shield. If the Soviets initiate total war, therefore, what would the defensive land shield accomplish? Who then would worry about occupation?

Our European allies are now nakedly vulnerable to Soviet air attack. They still lack an integrated radar network and interceptor missiles. More important, most of them are but minutes of jet flying time away from the Soviet threat, and for this reason it would be expensive and perhaps impossible to build a defense system assuring high attrition of the enemy. Passive defense measures, the alternative form of protection, have always been neglected by all NATO members. Nowhere in the West have we created anything approaching the hardened and dispersed underground civilizations required for protection against thermonuclear weapons. Though invulnerability cannot be achieved, some appreciable reductions in vulnerability may be purchased at reasonable cost. The very greatness of our deficiencies suggests that there may be some first-order things to do that promise sizable payoffs for modest investments. Tying together existing radar networks in Europe may be one of them; providing fall-out shelters for civilian populations may be another.

Are these things worth doing? The proponent of large armies often argues that resources should not be diverted on any sizable scale to such purely defensive purposes. But if he denies the usefulness of purely defensive measures, then the value of land forces as protectors against occupation is likewise small. In the context of total war, such forces also promise only to alleviate one kind of enormous damage. And if, on the other hand, the army proponent concedes the usefulness of purely defensive measures, the case for remedying the most glaring deficiencies against air attack becomes strong.

One recourse of the army proponent is to stress the advantages of a blend, while minimizing its disadvantages: "True, dollar for dollar we contribute less to the main deterrent than strategic airpower, but we offer a shield against invasion that it cannot provide. True, some measures to defend against air attack might do more to alleviate destruction, again dollar for dollar, than we can promise to do, but we contribute to the deterrent more than do such purely defensive measures." This is the logical argument of the army proponent who clings to a total war premise, an argument whose persuasiveness naturally declines with the spectacular growth of Soviet atomic airpower. The greater this threat, the more we must protect our own strategic air forces and insure that their protected strength is adequate. Hence the greater the tendency to pull resources away from armies to enhance airpower. With respect to air defense, the same growth of the Soviet air threat naturally increases vastly the prospects of bomb damage in the West if war occurs. As the terrors of bomb devastation loom steadily larger relative to those of occupation, the land shield loses its relative appeal as a mitigator of disaster. Hence the greater the tendency to divert resources from armies to air defenses.

Moreover, air defense measures for the United States contribute to the main deterrent beyond protecting our air striking forces. Their contribution is that of "backbone stiffening." How much the enemy is deterred from all-out air assault depends presumably upon his assessment of the security and power of our retaliatory forces and upon little else. But for aggression short of all-out air assault, his assessment is compounded by an additional factor: what is the probability that the United States would counter with everything it has? In seeking an answer to this question he necessarily assesses the resoluteness of our political leaders who must authorize the use of atomic weapons. His estimate of the risk will obviously depend upon how determined they appear, which in turn may depend upon how well equipped this country is in fact to withstand a retaliatory blow. After all, one chance in 20 that the United States would respond to a flagrant provocation by initiating total war is a tremendous risk for the Soviets. Increasing their assessment of the risk even a little by toughening our air defenses, thus appearing more resolute, would be an extremely handsome payoff for us. By how much we can increase his assessment of the risk is exceedingly difficult to quantify. But at least we see that those who desire some secondary strengthening of the main deterrent can argue as effectively for more air defense as for a stronger land shield.

The foregoing arguments all presuppose total-war planning designed to maximize deterrence of the enemy. Cling to a premise of total war and one can logically expect the other NATO Powers to follow the British example in respect to the reduction of conventional forces, thereby touching off the "chain reaction" so much feared by NATO military commanders. There is a strong temptation among NATO countries to pin their hopes upon deterring enemy provocations by the threat of SAC retaliation, for this has the great merit of permitting large budgetary economies. The cheapest solution is to have no strategic air force of one's own, to do nothing about vulnerability to Soviet air attack, and to raise only such conventional land, naval and tactical air forces as are required to take care of one's own colonial and other "brushfire" requirements. This solution, we note, implies no national contribution to collective defense, and hence strikes at the foundations of NATO.

Before criticizing the logic of this cheap solution, it is necessary to acknowledge its appeal. It is not the blind economizers alone who are attracted by it, but reasonable men who recognize that assured protection against all military eventualities is simply not procurable. They will not, as is sometimes done, overstate the main objection to a massive retaliation policy, namely that no enemy will credit you with any serious intent to retaliate massively because to do so would amount to an irrational policy of "suicide." It may be perfectly rational to commit yourself to a response which, if the enemy commits certain actions, does amount to mututal homicide. The probability that he will take these actions may be driven very low by this policy compared to alternative policies open to you. Suppose the "if" means that he sends troops across a clearly marked line which certainly could not have been crossed accidentally. Our policy does not in the least amount to saying that we value the territory in question more than our life. It amounts only to saying that since we believe the enemy values it much less than he does his life, our commitment will serve to reduce the probability of his crossing the line to the irreducible minimum.

The objection that "massive retaliation is an irrational policy because suicide is irrational," however, is less easily dismissed when applied to alliance strategy. It is more believable that America would retaliate massively against an attack upon Alaska than upon the United Kingdom, although a dispassionate analysis might well conclude that of the two the security of the United Kingdom is the more valuable to America. For America's partners in NATO, placing so great a reliance upon retaliatory power alone is uncomfortable, and this is doubly true when the retaliatory power is somebody else's. They must be sorely troubled by the thought that the credibility of a response by SAC to Soviet provocations in Europe declines as the Soviet airpower threat against America grows. On the other hand, the power of SAC grows steadily too, so that the net deterrent effect upon the Russians of a less certain but more powerful response by SAC may decline only a little or not at all. Still, depending upon an ally for the main deterrent is uncomfortable for Europeans. Yet what is their alternative?

One alternative is for each to create atomic weapons and a little strategic air force of its own. This course has the drawback of sizable expense, especially for the United Kingdom which has developed no less than three types of V-bombers and its own bombs, while it now faces, at the very least, the operating costs of the intermediate range ballistic missiles promised by the United States. They are paying a very high price for some prestige and a limited degree of strategic independence. But for other European countries, doubts about the advantage of such forces should give them greater pause than the certainty of sizable expense. First, would such forces really give them an independent capability to retaliate? Second, if such a capability is possible, would it actually strengthen Soviet inhibitions?

Doubts on the first score arise primarily because European-based planes or missiles are so close to Soviet airpower that protecting such forces is a much more formidable problem than protecting forces based in America. And protecting U.S. forces is difficult enough, while not protecting them vitiates a strategy of deterrence by inviting all-out attack. Forces in Europe are but minutes of jet flying time away from Soviet air bases and are exposed to very heavy and accurate attack with little warning. It may not be feasible to protect European-based forces against massive surprise air assault in the face of this geographical disadvantage; or, if feasible, it will almost certainly involve much more expensive forms of protection per plane or missile than to provide comparable security for SAC.

Doubts on the second score should compound those on the first. Assume that an independent retaliatory capability is attainable at reasonable cost, say a small missile force so protected that several missiles could probably survive Soviet all-out attack, and then penetrate to Soviet cities by virtue of phenomenal speed. The Kremlin's logic may then be this: "Although European Country X now has a limited capacity to retaliate with air atomic weapons, this independent capability decreases the probability that SAC will respond to our provocation, because the Americans now have an excuse for 'contracting out' of any alliance obligation. On balance, the small additional decline in the probability of the enormous American threat being made good outweighs even a sizable increase in the threat posed by Country X because its strategic force is so much smaller. Conclusion: we can feel freer to provoke." In brief, little strategic air forces in Europe can weaken the West's deterrent power either because they are so vulnerable as to invite attack, or because their existence weakens the bonds of alliance by making a SAC response less likely.

The uncomfortable but cheap solution for European governments--continued reliance on SAC--may well be the least unappealing of the poor alternatives available as long as NATO strategy is one of total war. It promises economy, while other feasible military alternatives probably offer at considerable expense little if any reinforcement of the main deterrent and little protection in the event of war. We may lament the failure of our European allies to put forth greater military effort, but we cannot upset the logic of their failure to do so. If we want more European military effort, especially a genuinely balanced collective effort, we had better reshape our NATO strategy to supply reasons and incentives for it.

III

Everything changes if America announces that tactical forces and not SAC will be used in response to limited Soviet provocation in Europe, and that, moreover, various other political constraints may be imposed upon the degree of violence with which we respond--a policy of "graduated deterrence." Under these conditions our allies in Europe have a renewed incentive to supply conventional forces because their own security depends upon the use of such forces against Soviet threats short of all-out assault.

We would necessarily pay a heavy price in adopting any such strategy, whatever specific variant we chose. If we make clear to our allies that our main deterrent threat may not be invoked, we cannot avoid making it clear to the Russians at the same time. Qualifying our main deterrent would embolden our enemy and frighten our allies. In addition to being frightened, they would face the sizable cost of reversing the decline in their conventional forces. We would run a risk of impairing their morale rather than raising it. Only one thing could justify such a heavy price, and that is lowering the probability that total war would occur. Paradoxically, such a reduction is a reasonable prospect even should our main deterrent be weakened.

Going back to the second of the main premises on which a strategy of massive retaliation is based, we must now bring it into serious question. The "line" between East and West is not clearly defined nor mutually agreed upon. The "line" in Europe may be "the most sensitive, most sharply defined political line between East and West," but this is a relative observation. It serves only to emphasize the fuzziness of the line elsewhere in the world. The trouble with threats of massive retaliation is that they can deter only those conflicts that the enemy deliberately provokes, not those that arise accidentally or because events get out of his control. Consequently massive threats may well decrease the probability of war starting, but not to the vanishing point, and only at the tremendous cost of increasing the probability that a war, once started, becomes total. What happens if each side attempts to resolve local conflicts in its favor by invoking the threat of strategic air blows? Either side loses greatly if it backs down first by withdrawing its threat, encouraging further provocation. Consequently there is a great temptation to increase the magnitude of the threatened strategic air blows to increase the chance that the other will back down first. The strategic equilibrium is unstable, and the probability that total war will be precipitated is uncomfortably high.

If NATO countries are to have a sufficient incentive to supply tactical forces designed primarily to meet local conflicts, we and our allies must be persuaded (1) that qualifying the use of the main deterrent--a policy which, to a degree, encourages small wars--will reduce the probability of total war; and (2) that for all or most of our allies the damage associated with small wars is likely to be considerably less than the damage associated with total war. If we are so persuaded, the weighty budgetary argument against supplying sizable tactical forces can be countered in its own terms. What other military alternative then promises an equally low mathematical expectation of war devastation (probability of war multipled by expected damage in war) for comparable or lower expense? If the probability of total war rises with reliance solely upon strategic airpower to deter the enemy, then only heroic air or civil defense measures promise an equivalent reduction in the expectation of war devastation. These measures can be incredibly expensive. The undeniably expensive land forces that supply an alternative are probably cheaper. True economy may best be served by the apparently uneconomical policy of preparing for two kinds of wars to be fought in considerable part by different forces.

Implementing an acceptable policy of "dual preparedness" is bound to be very difficult. It is hard to meet the second condition for acceptability noted above, namely, that damage in limited war be much less for all or most of our allies than in total war. Substituting "merely tactical" atomic weapons (how defined?) for troops does not meet this condition. "Taking war back to the battlefield" is a fine slogan, but if used to rationalize unlimited theater employment of atomic weapons it becomes a mockery to the allies over whose territory they are being employed. The battlefield will be deep, because tactical air bases, missile launch sites and logistic supply lines are obviously prime targets. Hitting them hard and early is terribly important. And military commanders have compelling reasons to use big weapons rather than little ones to compensate for inaccuracy in their delivery. The battle line itself may be fluid and deep penetration of enemy forces into allied controlled areas must be expected. These are chilling prospects indeed for our allies, for the distinction between limited and total war becomes less and less discernible to them. To cap their disillusionment, the economies promised by substituting tactical atomic weapons for troops turn out to be small and perhaps negative. True, a small force equipped with big-bang weapons can pack a large destructive power, thus permitting manpower economies for any given level of combat potential. But planning for full tactical use of atomic weapons in a context where the enemy's stock of such weapons grows steadily means that a higher and higher level of combat potential must be attained. As estimated casualties rise, so do force requirements, thereby offsetting the manpower economies hoped for by the introduction of atomic weapons.

Thus we are driven to a conclusion doubly unpalatable to some NATO commanders: the use of the main deterrent must be qualified, and there must even be severe constraints on employing atomic weapons in tactical operations. But if these commanders want strong incentives for Europeans to raise land troops, they must accept this conclusion, however reluctantly, and then face explicitly the terribly hard but enormously important practical problems of working out sensible weapons limitations. Weapons limitations must be acceptable both to the West and to the Soviets; the rules for limitation must be reasonably clear, so that violations are unlikely to be accidental and can be detected when they occur; and obviously there must be sanctions. Here, above all, is the place where strategic airpower must complement the tactical forces that may be operating under irksome restrictions, and here we must remain resolute despite the fearful stakes. SAC must continue to play the supremely important rôle of prime deterrent not only before war breaks out, but during war itself. The enemy must be persuaded that violating the weapons limitations or engaging in all-out land assault may bring against him a massive SAC strike. Then the terrible double penalty he faces operates to dampen hostilities rather than aggravate them: (1) the local war serves to alert SAC and our air defenses, so that enemy chances of precluding retaliation by an all-out assault diminish appreciably; (2) violating weapons limitations in the field, his alternative means of gaining a victory, runs a serious risk of giving SAC the enormous strategic advantage of the first strike.

The key to any policy of weapons limitations must be a reasonably clear definition of provocations that would invoke massive retaliation. We cannot expect to find an absolutely clear one, but we can certainly hope for something less fuzzy than current notions about what will trigger massive retaliation, and so reduce the chance of stumbling into total war. One limitation that possesses the great merit of clarity is simply non-use of atomic weapons, a policy, however, which maximizes the Soviet's advantage in mobilized manpower. Another radical possibility is for each side to restrict the use of atomic weapons to the open seas or to those territories clearly recognized to be one's own. This is not as clear a policy, but it does permit the defender, but not the aggressor, to concentrate his troops, thus enhancing the normal advantage of armies on the defensive and making a few troops go a long way. In these circumstances atomic weapons could be used against enemy aircraft and submarines certainly, and in a restrained way in tactical land operations probably. In land operations, each side would have powerful reasons to use fewer atomic weapons, and ones that are cleaner and smaller, because the territory affected would be his own or that of his allies. These drastic limitations have their obvious military disadvantages in terms of destroying enemy armies, but anything less drastic might well involve such levels of mutual destruction that fighting a war would be tantamount to losing it. In that case why not pin everything upon strategic deterrence?

That we can even discuss the feasibility of limited-war strategies for NATO today stems in part from our past inconsistencies. We have talked total war, but built mobilization bases that in the event of total war would probably be destroyed before they could be used, provided reserve forces that could not be mobilized, and built elaborate logistic facilities and bases that would be smashed. These legacies do possess great utility in limited war. Furthermore, we must always remind ourselves that it is the West that has by far the greater economic potential, a fact still relevant for conventional war if not for total atomic war. Moreover, NATO "dual preparedness" for old-fashioned war in Europe backed by a mainly American retaliatory force capable of waging total atomic war can take advantage of the great economies latent in a natural division of military labor. Within limits, of course, inexpensive European manpower can substitute for expensive American troops, while the advantages of geographical protection and operational concentration can be realized by strategic airpower that is primarily American. Should we throw away these advantages at a time when the contribution of Soviet satellite troops has been shown to be dubious or even negative?

The danger is that we will throw away these advantages by remodeling our tactical forces to suit exclusively the needs of a total atomic war in which their utility is problematical. Surely it is sensible to provide alternative sets of bomb-racks on all NATO bombers so that we can use either atomic or non-atomic weapons, or to supply cheap, unguided missiles to our allies, the Honest John, for example, that can be used either way. Some compromising of our ability to deliver conventional weapons is inevitable in enhancing the flexibility of our response to provocations, and this is a price to be cheerfully paid. But providing NATO with elaborate guided missile installations, revamped logistic systems, and greatly increased air defenses are very different matters. We can pass a point of no return at which our capability for limited warfare has been fatally compromised. Indeed, we should have passed this point already if our actions had been as prompt and unequivocal as our doctrine implied.

Total reliance upon threats of massive retaliation is a frightful risk to run, in Europe as elsewhere. So we return to our opening quotation, holding that General Norstad is profoundly right in appealing for maintenance of a sizable shield in Europe. But the case for maintaining the shield can no longer be based effectively on the grounds that it complements strategic airpower when both are used in total war. The possibility of keeping war limited, difficult though it may be, is what provides the incentive to retain and enlarge the shield, while strategic airpower imposes on both sides the obligation to keep the war limited.

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