NSC 68: United States Objectives and Programs for National Security
VIII. Atomic Armaments
A. MILITARY EVALUATION OF U.S. AND USSR ATOMIC CAPABILITIES
1. The United States now has an atomic capability, including both numbers and deliverability, estimated to be adequate, if effectively utilized, to deliver a serious blow against the war-making capacity of the USSR. It is doubted whether such a blow, even if it resulted in the complete destruction of the contemplated target systems, would cause the USSR to sue for terms or prevent Soviet forces from occupying Western Europe against such ground resistance as could presently be mobilized. A very serious initial blow could, however, so reduce the capabilities of the USSR to supply and equip its military organization and its civilian population as to give the United States the prospect of developing a general military superiority in a war of long duration.
2. As the atomic capability of the USSR increases, it will have an increased ability to hit at our atomic bases and installations and thus seriously hamper the ability of the United States to carry out an attack such as that outlined above. It is quite possible that in the near future the USSR will have a sufficient number of atomic bombs and a sufficient deliverability to raise a question whether Britain with its present inadequate air defense could be relied upon as an advance base from which a major portion of the U.S. attack could be launched.
It is estimated that, within the next four years, the USSR will attain the capability of seriously damaging vital centers of the United States, provided it strikes a surprise blow and provided further that the blow is opposed by no more effective opposition than we now have programmed. Such a blow could so seriously damage the United States as to greatly reduce its superiority in economic potential.
Effective opposition to this Soviet capability will require among other measures greatly increased air warning systems, air defenses, and vigorous development and implementation of a civilian defense program which has been thoroughly integrated with the military defense systems.
In time the atomic capability of the USSR can be expected to grow to a point where, given surprise and no more effective opposition than we now have programmed, the possibility of a decisive initial attack cannot be excluded.
3. In the initial phases of an atomic war, the advantages of initiative and surprise would be very great. A police state living behind an iron curtain has an enormous advantage in maintaining the necessary security and centralization of decision required to capitalize on this advantage.
4. For the moment our atomic retaliatory capability is probably adequate to deter the Kremlin from a deliberate direct military attack against ourselves or other free peoples. However, when it calculates that it has a sufficient atomic capability to make a surprise attack on us, nullifying our atomic superiority and creating a military situation decisively in its favor, the Kremlin might be tempted to strike swiftly and with stealth. The existence of two large atomic capabilities in such a relationship might well act, therefore, not as a deterrent, but as an incitement to war.
5. A further increase in the number and power of our atomic weapons is necessary in order to assure the effectiveness of any U.S. retaliatory blow, but would not of itself seem to change the basic logic of the above points. Greatly increased general air, ground, and sea strength, and increased air defense and civilian defense programs would also be necessary to provide reasonable assurance that the free world could survive an initial surprise atomic attack of the weight which it is estimated the USSR will be capable of delivering by 1954 and still permit the free world to go on to the eventual attainment of its objectives. Furthermore, such a build-up of strength could safeguard and increase our retaliatory power, and thus might put off for some time the date when the Soviet Union could calculate that a surprise blow would be advantageous. This would provide additional time for the effects of our policies to produce a modification of the Soviet system.
6. If the USSR develops a thermonuclear weapon ahead of the U.S., the risks of greatly increased Soviet pressure against all the free world, or an attack against the U.S., will be greatly increased.
7. If the U.S. develops a thermonuclear weapon ahead of the USSR, the U.S. should for the time being be able to bring increased pressure on the USSR.
B. STOCKPILING AND USE OF ATOMIC WEAPONS
1. From the foregoing analysis it appears that it would be to the long-term advantage of the United States if atomic weapons were to be effectively eliminated from national peacetime armaments; the additional objectives which must be secured if there is to be a reasonable prospect of such effective elimination of atomic weapons are discussed in Chapter IX. In the absence of such elimination and the securing of these objectives, it would appear that we have no alternative but to increase our atomic capability as rapidly as other considerations make appropriate. In either case, it appears to be imperative to increase as rapidly as possible our general air, ground, and sea strength and that of our allies to a point where we are militarily not so heavily dependent on atomic weapons.
2. As is indicated in Chapter IV, it is important that the United States employ military force only if the necessity for its use is clear and compelling and commends itself to the overwhelming majority of our people. The United States cannot therefore engage in war except as a reaction to aggression of so clear and compelling a nature as to bring the overwhelming majority of our people to accept the use of military force. In the event war comes, our use of force must be to compel the acceptance of our objectives and must be congruent to the range of tasks which we may encounter.
In the event of a general war with the USSR, it must be anticipated that atomic weapons will be used by each side in the manner it deems best suited to accomplish its objectives. In view of our vulnerability to Soviet atomic attack, it has been argued that we might wish to hold our atomic weapons only for retaliation against prior use by the USSR. To be able to do so and still have hope of achieving our objectives, the non-atomic military capabilities of ourselves and our allies would have to be fully developed and the political weaknesses of the Soviet Union fully exploited. In the event of war, however, we could not be sure that we could move toward the attainment of these objectives without the USSR's resorting sooner or later to the use of its atomic weapons. Only if we had overwhelming atomic superiority and obtained command of the air might the USSR be deterred from employing its atomic weapons as we progressed toward the attainment of our objectives.
In the event the USSR develops by 1954 the atomic capability which we now anticipate, it is hardly conceivable that, if war comes, the Soviet leaders would refrain from the use of atomic weapons unless they felt fully confident of attaining their objectives by other means.
In the event we use atomic weapons either in retaliation for their prior use by the USSR or because there is no alternative method by which we can attain our objectives, it is imperative that the strategic and tactical targets against which they are used be appropriate and the manner in which they are used be consistent with those objectives.
It appears to follow from the above that we should produce and stockpile thermonuclear weapons in the event they prove feasible and would add significantly to our net capability. Not enough is yet known of their potentialities to warrant a judgment at this time regarding their use in war to attain our objectives.
3. It has been suggested that we announce that we will not use atomic weapons except in retaliation against the prior use of such weapons by an aggressor. It has been argued that such a declaration would decrease the danger of an atomic attack against the United States and its allies.
In our present situation of relative unpreparedness in conventional weapons, such a declaration would be interpreted by the USSR as an admission of great weakness and by our allies as a clear indication that we intended to abandon them. Furthermore, it is doubtful whether such a declaration would be taken sufficiently seriously by the Kremlin to constitute an important factor in determining whether or not to attack the United States. It is to be anticipated that the Kremlin would weigh the facts of our capability far more heavily than a declaration of what we proposed to do with that capability.
Unless we are prepared to abandon our objectives, we cannot make such a declaration in good faith until we are confident that we will be in a position to attain our objectives without war, or, in the event of war, without recourse to the use of atomic weapons for strategic or tactical purposes.
C. INTERNATIONAL CONTROL OF ATOMIC ENERGY
1. A discussion of certain of the basic considerations involved in securing effective international control is necessary to make clear why the additional objectives discussed in Chapter IX must be secured.
2. No system of international control could prevent the production and use of atomic weapons in the event of a prolonged war. Even the most effective system of international control could, of itself, only provide (a) assurance that atomic weapons had been eliminated from national peacetime armaments and (b) immediate notice of a violation. In essence, an effective international control system would be expected to assure a certain amount of time after notice of violation before atomic weapons could be used in war.
3. The time period between notice of violation and possible use of atomic weapons in war which a control system could be expected to assure depends upon a number of factors.
The dismantling of existing stockpiles of bombs and the destruction of casings and firing mechanisms could by themselves give little assurance of securing time. Casings and firing mechanisms are presumably easy to produce, even surreptitiously, and the assembly of weapons does not take much time.
If existing stocks of fissionable materials were in some way eliminated and the future production of fissionable materials effectively controlled, war could not start with a surprise atomic attack.
In order to assure an appreciable time lag between notice of violation and the time when atomic weapons might be available in quantity, it would be necessary to destroy all plants capable of making large amounts of fissionable material. Such action would, however, require a moratorium on those possible peacetime uses which call for large quantities of fissionable materials.
Effective control over the production and stockpiling of raw materials might further extend the time period which effective international control would assure. Now that the Russians have learned the technique of producing atomic weapons, the time between violation of an international control agreement and production of atomic weapons will be shorter than was estimated in 1946, except possibly in the field of thermonuclear or other new types of weapons.
4. The certainty of notice of violation also depends upon a number of factors. In the absence of good faith, it is to be doubted whether any system can be designed which will give certainty of notice of violation. International ownership of raw materials and fissionable materials and international ownership and operation of dangerous facilities, coupled with inspection based on continuous unlimited freedom of access to all parts of the Soviet Union (as well as to all parts of the territory of other signatories to the control agreement) appear to be necessary to give the requisite degree of assurance against secret violations. As the Soviet stockpile of fissionable materials grows, the amount which the USSR might secretly withhold and not declare to the inspection agency grows. In this sense, the earlier an agreement is consummated the greater the security it would offer. The possibility of successful secret production operations also increases with developments which may reduce the size and power consumption of individual reactors. The development of a thermonuclear bomb would increase many fold the damage a given amount of fissionable material could do and would, therefore, vastly increase the danger that a decisive advantage could be gained through secret operations.
5. The relative sacrifices which would be involved in international control need also to be considered. If it were possible to negotiate an effective system of international control the United States would presumably sacrifice a much larger stockpile of atomic weapons and a much larger production capacity than would the USSR. The opening up of national territory to international inspection involved in an adequate control and inspection system would have a far greater impact on the USSR than on the United States. If the control system involves the destruction of all large reactors and thus a moratorium on certain possible peacetime uses, the USSR can be expected to argue that it, because of greater need for new sources of energy, would be making a greater sacrifice in this regard than the United States.
6. The United States and the peoples of the world as a whole desire a respite from the dangers of atomic warfare. The chief difficulty lies in the danger that the respite would be short and that we might not have adequate notice of its pending termination. For such an arrangement to be in the interest of the United States, it is essential that the agreement be entered into in good faith by both sides and the probability against its violation high.
7. The most substantial contribution to security of an effective international control system would, of course, be the opening up of the Soviet Union, as required under the UN plan. Such opening up is not, however, compatible with the maintenance of the Soviet system in its present rigor. This is a major reason for the Soviet refusal to accept the UN plan.
The studies which began with the Acheson-Lilienthal committee and culminated in the present UN plan made it clear that inspection of atomic facilities would not alone give the assurance of control; but that ownership and operation by an international authority of the world's atomic energy activities from the mine to the last use of fissionable materials was also essential. The delegation of sovereignty which this implies is necessary for effective control and, therefore, is as necessary for the United States and the rest of the free world as it is presently unacceptable to the Soviet Union.
It is also clear that a control authority not susceptible directly or indirectly to Soviet domination is equally essential. As the Soviet Union would regard any country not under its domination as under the potential if not the actual domination of the United States, it is clear that what the United States and the non-Soviet world must insist on, the Soviet Union at present rejects.
The principal immediate benefit of international control would be to make a surprise atomic attack impossible, assuming the elimination of large reactors and the effective disposal of stockpiles of fissionable materials. But it is almost certain that the Soviet Union would not agree to the elimination of large reactors, unless the impracticability of producing atomic power for peaceful purposes had been demonstrated beyond a doubt. By the same token, it would not now agree to elimination of its stockpile of fissionable materials.
Finally, the absence of good faith on the part of the USSR must be assumed until there is concrete evidence that there has been a decisive change in Soviet policies. It is to be doubted whether such a change can take place without a change in the nature of the Soviet system itself.
The above considerations make it clear that at least a major change in the relative power positions of the United States and the Soviet Union would have to take place before an effective system of international control could be negotiated. The Soviet Union would have had to have moved a substantial distance down the path of accommodation and compromise before such an arrangement would be conceivable. This conclusion is supported by the Third Report of the United Nations Atomic Energy Commission to the Security Council, May 17, 1948, in which it is stated that ". . . the majority of the Commission has been unable to secure . . . their acceptance of the nature and extent of participation in the world community required of all nations in this field.... As a result, the Commission has been forced to recognize that agreement on effective measures for the control of atomic energy is itself dependent on cooperation in broader fields of policy."
In short, it is impossible to hope than an effective plan for international control can be negotiated unless and until the Kremlin design has been frustrated to a point at which a genuine and drastic change in Soviet policies has taken place.