Risks of Nuclear Proliferation

NEW DIRECTIONS IN ARMS CONTROL AND DISARMAMENT

THREE and a half years ago when John J. McCloy gave an accounting in this journal of where we stood in efforts for disarmament the picture which he presented was understandably somber.[i] In the previous September there had been some progress in the negotiation of a "Joint Statement of Agreed Principles," an agreement between the United States and the Soviet Union on certain criteria that would have to be met if progress was to be made on substantive measures. But that one hopeful development had to be balanced against a record of discouragingly little progress in any other negotiations, growing difficulties over Berlin culminating in the erection of the Wall, and the termination, also in September 1961, of the understanding on the nuclear-test cessation.

Since then significant changes have occurred in Soviet-American relations and these have substantially affected the prospects for arms control and disarmament. The resolution of the Cuban missile crisis appears to have been a major turning point in this difficult search. Since that time, the Soviet Union appears to have appreciated to a degree not apparent before that its long-term interests might best be served by increased emphasis on internal development and a relaxation of the cold war. In the field of arms control and disarmament, this has been reflected in increased interest in small steps which, though they lack some of the popular appeal of general and complete disarmament, are more realistically attainable in the foreseeable future. There has been some real progress: creation of the so- called hot line between Washington and Moscow; the United Nations resolution banning the placement in space of weapons of mass destruction; some evidence of progress through mutual example in the simultaneous announcement of cutbacks in planned production of fissionable materials and in apparent reductions in military budgets; and, most notably, the treaty prohibiting nuclear tests except underground.

II

But as we consider the future, another change in attitudes is likely to be at least as significant as that in Soviet-American relations-namely, a greatly increased sense of urgency in dealing with the problem of proliferation of nuclear-weapons capabilities among nations. This is not, of course, a new interest-United States concern about control of the atom was reflected as early as in the Baruch plan of 1946-but it is a matter that has commanded far more attention lately. Three major factors probably have led to this reassessment of priorities: an increasingly realistic appreciation that agreement on general and complete disarmament will not be achieved early enough so that we can count on it to control the atom; a realization in the last few years that very significant quantities of plutonium were likely to be produced in nuclear power plants in many countries in the very near future; and, very importantly, and perhaps somewhat belatedly, an appreciation of the full implications of the Chinese nuclear tests.

It is clear that the Chinese tests have had an unsettling effect throughout much of Asia and particularly in India. The reaction there, and to a lesser degree in Japan, seems to have developed out of fear of the nascent military threat implicit in the tests and out of concern that Indian (and Japanese) prestige and influence would suffer. The degree of India's concern is due in part, of course, to the impression that she had been dealt a severe blow as a consequence of the Chinese attack in late 1962. Undoubtedly the test explosions have further enhanced Peking's prestige, particularly because they were the first nuclear tests by a non-white nation. However, the magnitude of the accomplishment has been exaggerated. The success of China's program is in large measure due to the very substantial assistance provided by the Soviet Union during the 1950s. In other respects the Indian program for exploitation of the atom is almost certainly more impressive than that of China; and so is Japan's. But despite the fact that the progress being made by these countries in the peaceful applications of nuclear energy, and in other areas, dwarfs in many respects the progress of the Chinese, in the contest for leadership among the Asian powers China's status has been enhanced as a result of its nuclear tests.

Fortunately, political leaders in India and Japan as well as in other countries such as Israel or Sweden appear to be convinced that it would be a mistake to engage in a nuclear-weapons program, despite pressure to do so. Whether they will be able to hold out against pressure to "go nuclear" will depend substantially on whether or not measures to prevent nuclear proliferation are negotiated in the months ahead, and whether or not there is other evidence that the great powers are really prepared to deal with the problem.

We continue, of course, to be concerned about the Soviet threat; and the Soviets are no doubt worried about what they must regard as a threat to them implicit in our large stock of nuclear weapons and intercontinental delivery systems. Both of us must regard it as important that we continue to seek ways to reduce the probability of a Soviet-American conflict, and also to mitigate the consequences should it occur. But whereas a few years ago these were our major concerns, it is no exaggeration to argue that we should now be at least as concerned about the problem of nuclear proliferation; and that in evaluating the implications of agreements which might affect the Soviet-American confrontation, we must give at least as much weight to the effect of such agreements on the proliferation problem as to their effect on the Soviet-American balance per se. In considering this reassessment of priorities, it is important to recognize that despite occasional backward steps (of which the present controversy over Viet Nam is a worrisome example) the trend in Soviet-American relations since the death of Stalin and particularly since the Cuban crisis has been in a generally favorable direction, whereas the trend in nuclear proliferation seems decidedly unfavorable. And there is a difference in the time scale for action in the two areas. For a decade and a half the Soviet Union has had nuclear weapons; hence the prospect of a delay of, say, a year or two in reducing the capabilities of the Soviet Union and the United States to damage each other may not seem terribly critical in itself. But a delay of a year or so, or perhaps even of months, in the implementation of measures bearing on the nuclear-proliferation problem could well mean the difference between failure and success.

III

Clearly implicit here is the view that further nuclear proliferation would have very undesirable effects on our security interests. Though this is the view of the U. S. Government, there is a contrary argument that might be made to the effect that in some selected instances the further spread of nuclear-weapons capabilities may actually be desirable. For example, it might be argued that, in the absence of any realistic possibility of inducing Communist China to give up her nuclear aspirations, it would be desirable that other selected countries in Asia have nuclear weapons to offset potential Chinese threats. In support of this, one might speculate that, if other Asian nations were to acquire nuclear capabilities, our burdens in defending the Asian periphery against China would be lessened, and that in the event of war there it might be less necessary for the United States to involve itself. This line might be further supported with the argument that the spread of nuclear weapons probably cannot be prevented anyway, or that if it can be, the political and other costs to the United States would be unacceptably great.

The case for "limited" proliferation seems to me to be based on two premises that are both implausible and inconsistent with the attitude we have taken with respect to Europe: first, that proliferation could be controlled as selectively as we might like; and second, that a country with the world-wide commitments of the United States could avoid involving itself in any conflict on a scale where nuclear capabilities would be significant.

That the first assumption is dubious becomes apparent when one considers what are almost certainly the most immediate cases in point, India and Israel. With the present unsettled state of affairs between Pakistan and India, it seems extremely improbable that either would be prepared to forego indefinitely the acquisition of a nuclear-weapons capability once it had become apparent that the other had decided to obtain one. The same is true of Israel and the United Arab Republic-a point which it is to be hoped will be fully appreciated by all elements in these countries whenever the question of acquiring nuclear weapons arises. Having five nations with nuclear weapons is bad enough, and if the number is to be limited, the prospects are almost certainly better at five than at six or any higher number.

In the short run we might successfully avoid involvement in, say, an Asian conflict in which nuclear weapons had a role. But any such success would, I believe, be short-lived and bought at a price that would prove unacceptable in the long run. That price would be a renunciation of our commitments and involvement all over the world-an attempt to return to isolationism at a time when the world is shrinking so rapidly as to make any such policy at best wishful thinking and quite possibly a blueprint for disaster.

Despite the gloomy prospects implicit in such a policy, it is highly likely that, in a world of many nuclear powers, considerable pressure would develop-and perhaps it would prove sufficient-to force the United States in just such an isolationist direction. This possibility must surely be considered a major argument for our attempting to stop the spread of nuclear capabilities. There are two others.

Very advanced nuclear capabilities are demanding in their requirements for scientific and technical manpower, as well as in material resources, but they are relatively cheap, and getting cheaper, in terms of the destructive power involved. With improvements in technology, the initial cost of nuclear weapons has been greatly reduced. Moreover, in many cases much of the development cost might be written off as chargeable to peaceful nuclear- power programs. Depending somewhat on whether or not defensive weapons are deployed, delivery systems may also become less costly. In our own case, the Minuteman systems are much less costly to build and particularly to operate and maintain than the B-52 and the first-generation I.C.B.M. All of which means that strategic nuclear-weapons systems could prove to be the great equalizer in international affairs that guns were in the case of individual combat. When we consider the costs to us of trying to stop the spread of nuclear weapons, we should not lose sight of the fact that widespread nuclear proliferation would mean a substantial erosion in the margin of power which our great wealth and industrial base have long given us relative to much of the rest of the world.

Finally, there is the simple fact that the probability of nuclear weapons being used will almost certainly increase as the number of fingers on the trigger increases. Moreover, the increase in probability will be more than proportional to the increase in numbers, particularly as, in a world of many nuclear powers, there may well be some who, unlike the United States and the Soviet Union, have relatively little to lose if nuclear weapons are used. Of course, the use of a few nuclear weapons by any power-even of one such weapon and even with an intent to localize the effect-might lead to their use in large numbers by other powers, with cataclysmic consequences.

In the light of such consequences of proliferation, we would seem justified in accepting rather large costs in an effort to prevent it. Of course, even our best efforts may not lead to success. Considering the lateness of the hour, the incentives that presently exist for the acquisition of nuclear weapons, and the prospect that they may be acquired with increasing ease, one is forced to conclude that a really major effort involving many kinds of actions will be required if there is to be any reasonable prospect of stemming the tide. These must include actions both to make the acquisition of nuclear capabilities more difficult and to reduce the incentives to acquire them.

IV

One of the central facts with which we have to deal is the very great overlap between the technology for the peaceful exploitation of the atom and that needed for weapons programs. In the enthusiasm of the late 1940s and early 1950s we and others were perhaps oversold on the potentialities of the peaceful atom and in particular on its application to the production of power. In consequence, a number of countries have been enabled to acquire nuclear reactors and technically trained people perhaps somewhat more rapidly than economics and other considerations could have realistically justified, and certainly under controls that are in some instances less rigid than would be desirable. Actually, the American record with respect to controls has been very good. The Baruch proposals, our Atomic Energy Act of 1946, our efforts, particularly in recent years, to get applied to nuclear programs throughout the world safeguards of the type established by the International Atomic Energy Agency, and our willingness to go further than any of the other nuclear powers have done in opening our reactors to international inspection-all are evidence of this. One reason we must be as concerned as we are today about the possibility of certain nations deciding to "go nuclear" is partly because the U. S. example has not, unfortunately, been emulated as widely as we might have hoped.

We are of course now at the point where the use of nuclear power is becoming economically advantageous in many situations. As a result, we must expect that within the next few years a number of countries will each be producing enough by-product plutonium to sustain a modest weapons program. Hence it is increasingly important that there be a more widespread, and stricter, application of controls to the traffic in fissionable materials and to the technology which may be useful either for peaceful or military purposes. Particularly important are international agreements on uniform standards to prevent critical materials and equipment from being offered for sale with inadequate controls, in order to realize economic or political advantage.

The most noteworthy direct efforts to make acquisition of nuclear weapons difficult has been, of course, the negotiation of a test-ban treaty which achieved at least partial success in 1963. Those efforts have been documented in great detail elsewhere. Nevertheless, in view of the importance of the present test ban as an anti-proliferation measure and its value as a precedent, it may be useful to comment further here, particularly regarding the possible risks and advantages of the treaty.

Those supporting the treaty offered varying appraisals of the relative importance of the objectives: the prevention of fallout; the directly inhibiting effect on those governments which might otherwise wish to test but which, having been forced to accede to the treaty by public and world opinion, would be reluctant to violate its terms; and the indication to the world of a willingness on the part of the great powers at least to begin to bring the nuclear-arms race under control. The last point is perhaps as important as the second in inhibiting other nations from going ahead with nuclear-weapons programs.

The extent to which these objectives could be realized by the treaty was carefully weighed within the U. S. Government. The military advantages to us were balanced against such risks and disadvantages as there may be in the treaty, including, of course, consideration of the military advantages that the Soviet Union might realize. The possibility of faithful compliance with its terms had to be borne in mind no less than the possibility of evasion, with some small risk that this would go undetected.

As might have been expected, this balancing of risks and advantages proved to be exceedingly difficult, particularly since the effects on nuclear proliferation, on international relations generally, and on further progress in arms control and disarmament were not susceptible to technical analysis as were other elements in the problem. The latter included fallout, the probability of successful evasion, the implications for weapons technology and the consequent impact of the agreement on the Soviet- American military balance under various assumptions regarding compliance.

These latter elements of the problem were the ones to which greatest attention was paid during the debate, in part simply because they were susceptible to orderly technical analysis. But the emphasis accorded them was at least as much due to the fact that by coincidence they were also matters of great general concern at that time. There was seemingly a relative lack of concern about nuclear proliferation and the less predictable relationship of the test ban to international affairs in general.

As we consider other significant arms-control or disarmament agreements-or even unilateral policy decisions-we shall have to go through a comparably difficult analysis, forcing us to weigh risks that are relatively easy to define against nebulous advantages (and vice versa), and the probability (generally not the certainty) of long-range gains against short-term costs.

The question of balancing long-term against short-term consequences is worth stressing. We sometimes have great difficulty in getting the right perspective on the time factors involved, often seeming far too concerned about the immediate effects of suggested measures on international affairs. Similarly, but at the other extreme, we seem at times to be unduly concerned about long-term risks which, while admittedly real, may be quite overrated because we discount the probability that conditions will change; what might be risky today may not seem so in a different international environment some years hence. For example, we may do ourselves a disservice if we discount the possibility of changes in the international environment and give too much weight to such factors as whether or not the Soviet- American strategic balance might be upset 15 years hence by weapons systems that the Soviet Union might develop as a result of successful evasion of some agreement under consideration today. There are, of course, some changes of which we can be reasonably certain. We can be sure that technology will spread and advance at least as rapidly in the next decade as in the last, and we can probably state with confidence that there will be substantial political realignments. We can be reasonably sure that nuclear capabilities will spread unless we take action very promptly to stem the tide. Weighing the relative risks is extremely difficult; but we do not minimize the over-all risk to us if we take too myopic a view of particular risks rather than facing up to the totality of dangers in all their complexity.

We are of course continuing our efforts to extend the nuclear test-ban treaty to prohibit testing underground as well as in the other environments. In this we face a somewhat different problem in balancing risks and advantages than when we contemplated a comprehensive test-ban agreement in 1962 or the partial test ban which we achieved in 1963. There is certainly less to be gained from extending the test ban to include underground explosions than there was when we had no agreement at all. Fallout is no longer an issue and much of the debate about military implications of weapons tests is not relevant; nor is there the possibility, for example, of very high-yield tests and of weapons-effects tests at high altitudes. Further, much of the possible inhibiting effect which a full test-ban treaty would have on nuclear proliferation has been achieved with the partial test ban. Nevertheless, extension of the treaty does seem desirable as an anti-proliferation measure. Though none of the nuclear powers began its weapons program with underground tests, it would not be difficult for others to do so while still complying with the treaty; and it does seem desirable to foreclose that possibility. Those elements, particularly in democratic states, which oppose their countries' acquiring nuclear weapons would be in a much stronger position if a treaty existed prohibiting all such tests than if the option of legitimate underground testing still remained open. Finally, extension of the treaty would be evidence of willingness on the part of the great powers to take further steps to control the arms race.

Fortunately, technical improvements may make it easier to discriminate between earthquakes and possible underground explosions. While all present scientific evidence indicates that a number of inspections continue to be necessary to provide verification, the United States continues to be willing to explore what would constitute an adequate verification system in the light of these recent and prospective developments in our capabilities for detection. If such exploration indicates that verification requirements can be satisfied by a different number and type of inspections from those previously discussed, the United States will certainly take those facts into account. Whether we can thus close the gap between the American and Soviet positions remains to be seen in further discussions.

There are other multilateral measures that may be of direct value as components of a program to stop the spread of nuclear capabilities; for example, agreements on nuclear-free zones and agreements restricting the sale or transfer of strategic aircraft and missiles, the means for delivering nuclear weapons.

But probably most important would be negotiation of a non-proliferation agreement-that is, an agreement on the part of the nuclear powers not to transfer control of nuclear weapons to other countries and an agreement on the part of the non-nuclear powers not to acquire them. Such a measure seems so overwhelmingly in the interest of both the Soviet Union and the United States that it may seem surprising that they have been unable to reach agreement; but such is so far the case.

The principal problem appears to be the Soviet refusal to go along with an agreement permitting the NATO Multilateral or Atlantic Nuclear Force (M.L.F. and A.N.F., respectively) and the Western refusal to renounce these projects. In part, of course, they sprang from the belief that any incipient desire for independent control of nuclear weapons by Europeans, particularly in Germany, could be most safely dealt with by giving them a role in a collective enterprise. But the Soviet Union has refused to concede that the M.L.F./A.N.F. idea may have value as an anti-proliferation measure, and indeed has argued the contrary. It insists that the implementation of such a concept would lead to control of nuclear capabilities by the German Federal Republic. To a certain extent this argument may reflect genuine concern, but it also seems clear that the Soviets have been using the M.L.F./A.N.F., and its connection with a non- proliferation agreement, for propaganda value and for whatever divisive effect it may have in NATO.

Whether we will be able to overcome this impasse will depend in large measure on the Soviet assessment of the relative advantages of actually getting a proliferation agreement or of continuing to use the M.L.F./A.N.F. issue for its political effect. In making this balance, the Soviets could assume that they could include in any non-proliferation agreement that might be negotiated a withdrawal clause similar to that found in the limited test-ban treaty.

For the near and medium term, the incentives for others to acquire nuclear weapons can undoubtedly be reduced, if not controlled, by adequate security assurances or guarantees-a point of particular importance since the Chinese nuclear test. The United States has already gone some distance toward assuring non-nuclear powers of our support in the event of nuclear attack, but we may have to go further if our assurances are to be considered truly adequate. This could mean either an increase or a decrease in the risks of our involvement in actual conflict, depending on the appraisal of others as to our seriousness. In some cases, too, would-be nuclear powers may ask us to provide assurances against conventional attack. There is also the question of obtaining assurances from a number of nuclear powers-an arrangement that the Indians, for example, would feel was more adequate than a U. S. (or Soviet) assurance alone.

Whether any assurances or guarantees will be considered credible depends of course to a substantial degree on the relative strength of the powers involved. One of our difficulties in NATO-and in France in particular-has been the growing feeling, perhaps most clearly articulated by the Prime Minister of France, that our commitment to NATO was of diminishing credibility as Soviet capacity to damage the United States increased.[ii] With the passage of time a similar erosion of confidence might occur with respect to any assurances designed to counter the Chinese threat. But we probably do have a number of years during which Chinese nuclear capabilities will be so small relative to those of the United States (or of the U.S.S.R.) that American (and/or Soviet) assurances would be entirely credible, assuming of course at least some degree of American (and/or Soviet) concern about the viability of the country in question.

In the longer term we must deal with the problems of the growing Chinese nuclear capability and of the role of nuclear weapons in world affairs. In a sense, all of the measures discussed above ought to be considered primarily as a means of buying time in which to get at these more fundamental problems.

Unless the non-nuclear powers are persuaded that their interests are best served by not acquiring nuclear weapons they will ultimately acquire them. A necessary, though perhaps not sufficient, condition for so persuading them is to offer clear evidence that the Soviet Union and the United States are prepared to exercise leadership in the world on a basis of strength other than that inherent in their nuclear capabilities. It is for this reason that agreements to freeze production and to start reductions in fissionable materials and in nuclear delivery systems are so important.

Such measures appear to be in our mutual interest, for they would effect substantial savings and would reduce both the probability of war and the damage incurred if war were not avoided. One of the most hopeful signs is the growing appreciation of the fact that the kind of nuclear superiority which the United States enjoyed in the 1950s and to which the Soviets perhaps at one time aspired is a thing of the past. Secretary Robert McNamara made this especially clear in recent testimony before the House Armed Services Committee. He pointed out that only a portion of our strategic capabilities would be required to inflict unacceptable damage on the Soviet Union, but that the remainder, if used against Soviet offensive capabilities, would be effective only under the improbable contingency that the Soviets withheld the bulk of their force after the beginning of a thermonuclear exchange. Soviet forces in excess of what would be required to assure a high level of destruction against U. S. industry and population could be used even less effectively to reduce damage to the U.S.S.R. With these points in mind, it is hard to avoid the conclusion that we both would be well advised to stop the strategic nuclear-arms race and destroy some strategic capabilities on a reciprocal basis. It does not make good sense to continue the race in the hope that each may be able to use some of its capability to destroy the other's offensive strength, a task that is daily becoming more difficult.

In stressing that such measures as reductions in Soviet and American nuclear capabilities are important if we are to succeed in dealing with nuclear proliferation, it should be made clear that it is not a question of our setting a good example, a factor of regrettably little influence in international affairs, but rather the fact that we would, by negotiating such measures, be giving evidence of our determination to reverse the arms race and move toward a world order in which the role of nuclear weapons would be diminished. Lacking at least reasonable prospects of movement in this direction, it is hard to see how, in the long run, we can hope to put any limits on the membership in the nuclear club.

Actually, the perceived utility of nuclear weapons has diminished to some extent in recent years. This is particularly true in the United States. Ten years ago there was little interest in the Department of Defense in any weapons other than nuclear. The situation is drastically changed today. In part this has occurred because we have come to understand the limited role which nuclear weapons can play. Somewhat belatedly there is also a broader appreciation of the fact that in many combat situations nuclear weapons would not be useful-even apart from the hazards of escalation and retaliation. With each succeeding day that their use is avoided, it is possible that the likelihood of future use will further diminish. Thus, in this one respect time seems in our favor. The world would do well to bear this in mind as crises develop which might lead to their use.

If through short-term measures we can succeed in buying time, there is no place where the need to use it effectively is more urgent than in our confrontation with Communist China. Yet we can now make progress in arms control and disarmament without Chinese participation. A comprehensive test- ban treaty to which China did not accede would be better than none at all and should be acceptable to us. With or without a treaty, our testing program would not be contingent on what China might do in the next few years. Similarly, a non-proliferation agreement to which China did not accede would be better than none at all. And we could negotiate a freeze- even a substantial reduction-of strategic delivery systems without fear that we were compromising whatever nuclear capability was needed to deal with China.

But a few years hence, none of these things will be true unless we make progress in bringing China into arms-control and disarmament agreements. Nor will U. S. or Soviet or even joint guarantees for China's neighbors be very credible. Clearly, progress in dealing with China is as essential a long-term requirement for success in a non-proliferation effort as Soviet- American coöperation is in the near term.

Over the next decade or two we cannot expect major changes in China's objectives, which are in so many respects antithetical to ours. But with changes in Chinese leadership, which are inevitable soon, and with economic growth, which will give China more to lose through war, it is at least possible that the Chinese may conclude-as apparently have the Soviets-that they have much to gain by accepting the concept of peaceful coexistence. While this hope is based on the optimistic belief that time may bring restraint and wisdom to the Chinese leaders, the alternative is bleak indeed.

V

Having dealt at some length with means and possibilities of stopping nuclear proliferation, and with the consequences should we fail, we need to appraise the costs of effective agreement. It would be irresponsible to suggest that they will not be substantial.

Some interference with the exploitation of the atom for peaceful purposes seems inevitable. A number of the newer nations, and the Soviet Union particularly, may view international controls as unacceptable infringements on sovereignty, though the voting record of the U.S.S.R. on the extension of controls by the International Atomic Energy Agency is encouraging in this respect. U. S. insistence on controls, and unwillingness to permit export of certain equipment and technology that might have application in weapons programs may be, as it has been, cause for some international friction. Fortunately, if nations and industry are willing to accept adequate controls, most socially and economically desirable applications of the atom, such as nuclear power and the use of radioisotopes for medical purposes, can go ahead with little actual impediment. This may not, however, be true in the application of nuclear explosions for peaceful purposes, of which the economics are, incidentally, still in doubt. Some such explosions are, of course, already precluded by the present test-ban treaty, but we may have to accept the cost of forgoing any such explosions. Alternatively, we might agree on provisions which permit them; none that has been so far considered, however, is without awkward features. For example, some people fear that information on weapons design might be compromised; others assert that the great powers might proceed with weapons development under the guise of peaceful explosions; and still others feel that it might be necessary to employ relatively primitive devices that would not be altogether suitable.

A heavier cost could be the erosion of alliances resulting from the high degree of U. S.-Soviet coöperation which will be required if a non- proliferation program is to be successful. Within NATO, there could be concern that the détente would lead to a weakening of our commitment to Western Europe. The problem will be particularly acute in Germany where there will be the added concern that the amelioration of the East-West confrontation could lead to an increased acceptance of the status quo in Central Europe. On the other side, it is to be expected that any move toward détente will lead to an exacerbation of the Sino-Soviet split and of the contest for influence within the Communist world. Finally, there is the simple fact that the strains inherent in any alliance are likely to become more prominent, and the cohesive forces less so, if the threat which was a major part of its rationale is perceived to be less worrisome. We should not be surprised if a move toward East-West détente leads to some changes in both the Atlantic Alliance and the Warsaw Pact. What would be troublesome is if there were a weakening of NATO which proceeded at a more rapid rate than the actual diminution in the threat justified.

But proliferation of independent nuclear capabilities could also be expected to have a highly divisive effect on NATO, the Warsaw Pact and other alliances, for the reasons outlined earlier. In France's relationship to NATO, we have already seen the divisive effects both of relaxation of tensions and of nuclear proliferation. While there are other factors involved, we can surmise that the independent attitude of France toward NATO is in large part a consequence of her perceiving the Soviet threat as less worrisome than a decade ago and in part a consequence of her acquisition of some nuclear capability. We must accept the fact, then, that either nuclear proliferation or its successful prevention is likely to weaken alliances.

I would suggest that, as regards NATO, the best we might hope, and strive for, would be movement toward a substantial measure of East-West détente- hopefully sufficient to enable us to hold the line on nuclear proliferation- with perhaps progress on European union offsetting at least in part the fact that the cement holding NATO together will weaken with any further erosion of the fear of attack.

Balancing the risks and costs of letting nuclear proliferation run its course against those that may be incurred in a determined effort to stop it is clearly one of the most difficult problems in international relations today-and far more complex than the one we faced in deciding upon the limited test-ban treaty. It is particularly trying because success in stopping the spread of nuclear weapons may elude us despite the best efforts and intentions of many nations.

But stopping nuclear proliferation is a problem from which the world cannot shrink, and one which requires very prompt action if there is to be any reasonable hope of success. May others conclude-as we have-that, all things considered, a most serious and urgent effort is justified!

[i] "Balance Sheet on Disarmament," Foreign Affairs, April 1962.

[ii] Speech on the French military program before the French National Assembly, December 2, 1964.