Foreign Relations of the United States, 1961–1963, Volume VIII, National Security Policy

82. Address by Secretary of Defense McNamara at the Ministerial Meeting of the North Atlantic Council⁰

Athens, May 5, 1962.

[Here follows an introductory section.]

I. General Nuclear War and Its Deterrence

Nuclear technology has revolutionized warfare over the past seventeen years. The unprecedented destructiveness of these arms has radically changed ways of thinking about conflict among nations. It has properly focused great attention and efforts by the Alliance on the prevention of conflict. Nevertheless, the U.S. has come to the conclusion that to the extent feasible basic military strategy in general nuclear war should be approached in much the same way that more conventional military operations have been regarded in the past. That is to say, our principal military objectives, in the event of a nuclear war stemming from a major attack on the Alliance, should be the destruction of the enemy’s military forces while attempting to preserve the fabric as well as the integrity of allied society. Specifically, our studies indicate that a strategy which targets nuclear forces only against cities or a mixture of civil and military targets has serious limitations for the purpose of deterrence and for the conduct of general nuclear war.

In our best judgment, destroying enemy forces while preserving our own societies is—within the limits inherent in the great power of nuclear weapons—a not wholly unattainable military objective. Even if very substantial exchanges of nuclear weapons were to occur, the damage suffered by the belligerents would vary over wide ranges, depending upon the targets that are hit. If both sides were to confine their attacks to important military targets, damage, while high, would nevertheless be significantly lower than if urban-industrial areas were also attacked. As an example, our studies of a hypothetical general nuclear war occurring in 1966 show that, with the conflict starting under one particular set of circumstances, and with the Soviets confining their attacks to military targets, the United States under present civil defense plans might suffer 25 million deaths and Europe might suffer somewhat fewer. On the other hand, were the Soviets to attack urban-industrial as well as military targets, the United States might incur 75 million deaths and Europe would have to face the prospect of losing 115 million people. While both sets of figures make grim reading, the first set is preferable to the second. There are others like them.

In the light of these findings the United States has developed its plans in order to permit a variety of strategic choices. We have also instituted a number of programs which will enable the Alliance to engage in a controlled and flexible nuclear response in the event that deterrence should fail. Whether the Soviet Union will do likewise must remain uncertain. All we can say is that the Kremlin has very strong incentives—in large part provided by the nuclear strength of the Alliance—to adopt similar strategies and programs. Thus, we calculate that in 1966, if the Alliance were to limit its retaliatory attack to miltary targets in the Soviet Union, while holding superior forces in reserve, the Soviets might suffer around 25 million deaths, whereas if we attack urban-industrial targets in the wake of a Soviet strike against European and American cities, the Soviets would suffer at least 100 million deaths.

Other factors besides target strategies of the belligerents would determine the damage in a thermo-nuclear war. The yields of the warheads used in a nuclear exchange would make a significant difference in the amount of blast, thermal, and fallout damage; and it is possible to match the yields to the particular targets under attack and so reduce damage to civilians. Furthermore, as the accuracy of missiles improves, the belligerents could attack targets with greater assurance of destroying them; they could also reduce the yields with which they strike. If they so choose, they could regulate the height at which they burst their weapons and thereby affect the amount of fallout that is distributed. The existence of civil defenses also could have a significant impact on the number of deaths, especially if only military targets are attacked so that the principal danger to most civilians is from fallout. Depending on these and other factors, the number of deaths could vary over a wide range—by four times or more. The more discriminating the attacks, the less the damage.

I have raised these points because we think they are relevant to allied defense policies now and in the future. In particular, we believe that they have important implications for the general war posture of the Alliance and the role that NATO should assign to nuclear forces in its grand strategy.

II. The General War Posture of the Alliance

Perhaps the most important implication of these observations is that nuclear superiority has important meanings. I want to stress that for the most relevant planning period—through the mid 1960’s—there can be little question about the ability of the Alliance to maintain nuclear superiority over the Sino-Soviet Bloc. During the coming fiscal year the United States plans to spend close to $15 billion on its nuclear weapons to assure such superiority.

[Here follows a description of the U.S. strategic retaliatory arsenal, target coverage of the threat to Europe, survivability and control, and indivisibility of control.]

III. The Role of General War Strength in Alliance Strategy

What does the Alliance accomplish by creating this complex machinery to maintain nuclear superiority over the Sino-Soviet Bloc? And what is the impact on NATO’s policies of both the grave damage that would result from nuclear war and the great variations in that damage under different strategies?

My Government feels that the strategic capabilities I have described have important political consequences. The Alliance continues to possess much of the diplomatic freedom that it has enjoyed in the past. We can confidently reject the missile threats that Mr. Krushchev so imprudently brandishes. If the Soviets or their satellites impinge on our interests we can resist with considerable confidence that our antagonists will not wish to escalate the conflict. The question at issue now is the point at which NATO, not the Soviets, would wish to escalate a non-nuclear conflict.

As the President has indicated on a number of occasions, the United States is prepared to respond immediately with nuclear weapons to the use of nuclear weapons against one or more members of the Alliance. The United States is also prepared to counter with nuclear weapons any Soviet conventional attack so strong that it cannot be dealt with by conventional means. But let us be quite clear what we are saying and what we have to face. Owing to our non-nuclear deficiencies, there is, first, a high probability that in an ambiguous situation the West, not the East, would have to make the decision to initiate the use of nuclear weapons. Secondly, there is the almost certain prospect that, despite our nuclear superiority and our ability to destroy the Soviet target system, all of us will suffer deeply in the event of major nuclear war.

The Berlin crisis exemplifies a type of threat that we should expect to face elsewhere in the NATO area. In such a crisis the provocation, while severe, does not immediately require or justify our most violent reaction. Also as such a crisis develops, as military force is threatened or becomes engaged—even in limited quantities—the increasingly alert nuclear posture of the belligerents makes the prospective outcome of a nuclear attack for both sides even less attractive.

In short, faced with the more likely contingencies, NATO, not the Soviets, would have to make the momentous decision to use nuclear weapons, and we would do so in the knowledge that the consequences might be catastrophic for all of us.

We in the United States are prepared to accept our share of this responsibility. And we believe that the combination of our nuclear superiority and a strategy of controlled response gives us some hope of minimizing damage in the event that we have to fulfill our pledge. But I would be less than candid if I pretended to you that the United States regards this as a desirable prospect or believes that the Alliance should depend solely on our nuclear power to deter the Soviet Union from actions not involving a massive commitment of Soviet force. Surely an Alliance with the wealth, talent, and expression that we possess can find a better way than this to meet our common threat.

We shall continue to maintain powerful nuclear forces for the Alliance as a whole. They will continue to provide the Alliance a strong sanction against Soviet first use of nuclear weapons. Under some circumstances they may be the only instrument with which we can counter Soviet non-nuclear aggression, in which case we should use them. But, in our view, the threat of general war should constitute only one of the several weapons in our arsenal and one to be used with prudence. On this question I can see no valid reason for a fundamental difference of view on the two sides of the Atlantic.

IV. Tactical Use of Nuclear Weapons

Our great nuclear superiority for general war does not solve all our problems of deterring and dealing with less than all-out direct assault. What, then, is the prospect that NATO can fall back on the local or tactical use of nuclear weapons? Battlefield nuclear weapons were introduced in NATO at a time when our Shield forces were weak and the Soviet atomic stockpile was small. In these circumstances it was reasonable to hope that NATO might very quickly halt a Soviet advance into Western Europe by unilateral application of nuclear weapons on or near the battlefield. Using nuclear weapons tactically might still accomplish a desired end in the early 1960’s. Consequently, we continue to maintain substantial nuclear forces within the European theater and we now have over [omission in the source text] nuclear weapons of various yields stockpiled in Europe.

But how much dependence should NATO place on these capabilities? We should succeed in deterring the Soviets from initiating the use of nuclear weapons, and the presence of these weapons in Europe helps to prevent Soviet use locally. But NATO can no longer expect to avoid nuclear retaliation in the event that it initiates their use. Even a local nuclear exchange could have consequences for Europe that are most painful to contemplate. Further, such an exchange would be unlikely to give us any marked military advantage. It could rapidly lead to general nuclear war.

To be sure, a very limited use of nuclear weapons, primarily for purposes of demonstrating our will and intent to employ such weapons, might bring Soviet aggression to a halt without substantial retaliation, and without escalation. This is a next-to-last option we cannot dismiss. But prospects for success are not high, and I hesitate to predict what the political consequences would be of taking such action. It is also conceivable that the limited tactical use of nuclear weapons on the battlefield would not broaden a conventional engagement or radically transform it. But we do not rate these prospects very highly.

Highly dispersed nuclear weapons in the hands of troops would be difficult to control centrally. Accidents and unauthorized acts could well occur on both sides. Furthermore, the pressures on the Soviets to respond in kind, the great flexibility of nuclear systems, the enormous firepower contained in a single weapon, the ease and accuracy with which that firepower can be called in from unattacked and hence undamaged distant bases, the crucial importance of air superiority in nuclear operations—all these considerations suggest to us that local nuclear war would be a transient but highly destructive phenomenon.

I realize there is a school of thought which believes that the United States and the Soviet Union might seek to use Europe as a nuclear battleground and thus avoid attacks on one another’s homelands. Not only does my government emphatically reject such a view; we also regard it as unrealistic. It ignores the basic facts of nuclear warfare I have described; it contemplates geographical limits unrelated to the actualities of target locations, and of the varied sources from which attacks would come. Any substantial nuclear operation in Europe inevitably would involve both forces and targets in the U.S. and USSR. It is possible, as I have mentioned, that a small, demonstrative use of nuclear weapons could be contained locally, and possibly, distant nuclear operations in less vital locations outside the NATO area, or at sea, may be limitable. But there is likely to be no effective operational boundary, or set of mutual restraints, which could restrict large-scale nuclear war to NATO Europe and the satellites. As we understand the dynamics of nuclear warfare, we believe that a local nuclear engagement would do grave damage to Europe, be militarily ineffective, and would probably expand very rapidly into general nuclear war.

V. Non-Nuclear Forces and Deterrence

With the Alliance possessing the strength and the strategy I have described, it is most unlikely that the Soviet Union will launch a nuclear attack on NATO. But there are other forms of aggression, and in December I mentioned our concern that the threat of general war might be adequate against many lesser Soviet actions, political as well as military. Some such hostile actions we could thwart now; others we might not. To deal with these others, how can we convincingly show that aggression, if continued, would lead to a situation where the danger of nuclear war was very great indeed? Let us assume two situations:

In the first, the NATO front is lightly covered by our forces. In the event of deep penetration by Soviet non-nuclear forces which our forces cannot prevent, the only military options open to Alliance forces are immediate nuclear response or defeat. This might be true even for a minor Soviet challenge.

In the second, we assume the NATO front firmly held under a concept of forward strategy. Ready and able to deal with any Soviet non-nuclear attack less that all-out, NATO forces guard positively from the frontier against any quick strike or ambiguous aggression. The NATO front can be broken only by massive application of Soviet power. In such a major fight, if Western forces were thrown back, Alliance nuclear action would follow.

If you were on the other side, which situation would you consider more laden with a real risk of nuclear war with all its consequences? Which would make you more inclined to refrain from a series of actions designed, step by step, to erode NATO’s interests? To us the answer is clear.

In the first situation, it simply is not credible that NATO, or anyone else, would respond to a given small step—the first slice of salami—with immediate use of nuclear weapons. Nor is it credible that a chain of small actions, no one of which is catastrophic, would evoke a response of general nuclear war. We regard it as much more evident that NATO would find it politically possible to act in effective defense of its interests from the second posture than from the first.

The development of recent events concerning Berlin may provide relevant evidence of the utility of limited but decisive action. Although it would be premature to announce the end of this crisis, and in any case we cannot be certain of the influences that most affect Kremlin policy, it is not unlikely that the NATO non-nuclear buildup conveyed to the Soviets the right message about Berlin. When the Soviets began menacing Berlin, they may have entertained doubts about Western determination; clearly they were not deterred from their initial steps by our previous nuclear threats. But the creation of greater new non-nuclear strength has reinforced our overall deterrent, and the aggression has not occurred. It was not simply the substantial increase in NATO manpower and the addition of the equivalent of four combat-ready divisions, 88 more ships and 19 more air squadrons, but the meaning which their addition conveyed of our determination that may have given the Soviets second thoughts.

For the kinds of conflicts we think most likely to arise in the NATO area, non-nuclear capabilities appear to be clearly the sort the Alliance would wish to use at the outset. The purpose of our common effort is the defense of the populations and territories of NATO. To achieve this, at least initially, with non-nuclear means requires that our non-nuclear defense begin where the populations and territories begin. A truly forward deployment, along the lines General Norstad has advocated, we consider an urgent need of the Alliance.

Let me make clear however that we do not believe that a forward defense must be able to defeat in non-nuclear action every conceivable element of Soviet strength that might be thrown against it. Our nuclear forces would rapidly come into play if an all-out attack developed. We believe the Soviets can hardly doubt that; hence, we think it quite improbable that a major attack would develop out of a crisis.

[Here follows a description of NATO force dispositions and then-current U.S. views on Alliance decisions.]

0. Source: Department of State, S/P Files: Lot 69 D 121, Europe. Top Secret. Attached to Instruction No. CW-9106, May 15. The Ministerial Meeting of the North Atlantic Council was held in Athens May 4-6. Secretary McNamara delivered a similar address in unclassified form in Ann Arbor, Michigan, on June 16. For text, see Department of State Bulletin, July 9, 1962, p. 64.