Foreign Relations of the United States, 1958–1960, Volume III, National Security Policy; Arms Control and Disarmament, Microfiche Supplement

449. Memorandum From Killian to Eisenhower¹

Washington, March 31, 1959

SUBJECT

• Technical Factors Relating to Arms Limitation and to the Geneva Conference on Nuclear Test Cessation

In the attached memorandum I have outlined the principal technical factors which may have a bearing on policy decisions affecting nuclear test negotiations.

On the chance that you may find this summary statement of some help as background material, I send it along but I hasten to point out that the views expressed have not been coordinated with the Departments of State or Defense nor with the Atomic Energy Commission.

J.R. Killian, Jr.

Attachment

Memorandum by Killian

March 30, 1959

MEMORANDUM ON SOME TECHNICAL FACTORS INVOLVED IN POLICY DECISIONS ON ARMS LIMITATIONS AND SPECIFICALLY ON THE LIMITATION OF NUCLEAR TESTING

The recess of the Geneva Conference on the Discontinuance of Nuclear Tests with a stated date (April 13) for reconvening, requires the United States Government to re-examine its policy positions with respect to these negotiations.

This memorandum deals with my personal summary of the technical considerations which have a bearing on these policy questions. In discussing the technical factors, I recognize that they probably are of secondary importance to political and policy objectives.

Trends in Military Technology

Let me first point out how current trends in military technology emphasize the urgent importance of arms limitation of some kind. While deeply convinced that we must be unremittingly alert to keep ahead in our military technology, I also must conclude, in the light of factors mentioned below, that I see little opportunity to simplify the complexity of our military technology, to reduce the burden of defense, or to achieve a stable condition by means of military technology alone. On the contrary, the trend of technological measures and countermeasures will steadily complicate our defense, augment instability, and increase the cost of maintaining the relative strength we need.

We see this trend clearly revealed in our needs to make our strategic striking power, both aircraft and missiles, less vulnerable to surprise by dispersal, hardening, and shorter reaction times (including ground alert and possibly some degree of air alert). We see it clearly in the changing requirements of air and missile defense as typified by the problems associated with the Sage system and Bomarc, with Nike-Zeus, ballistic missile early warning systems, antisubmarine warfare systems, and with requirements for more comprehensive communications systems.

We see further evidence in the effort to improve the yield-to-weight ratio of atomic weapons, and in the developing program for using space and space vehicles for military purposes. The possibilities for military technology in modifying the geophysical environment of the earth are illustrated in the possibilities (remote) now envisaged for weather modification and by the results of Argus and other high-altitude experiments.

As military technology moves toward more systems operating on a global scale, and as it begins to make use of extraterrestrial space and effects, and as the time-scale becomes more condensed, we face the requirement for weapons systems, offensive and defensive, of increasing performance and complexity. Because of rapid technological change, we see the new systems overlapping with quickly obsolescing older ones and we see the consequent possibilities of systems being piled upon systems.

As complexity increases, the chances for error or aberrations on the part of humans and machines grow greater, and the consequences of such errors and aberrations (e.g., accidental war) become enormous. The shortened time-scale also reduces the opportunity for careful judgment by increasing the rewards for spontaneous response.

The overriding technological fact, however, is the continued build-up of improved high-performance nuclear weapons on both sides to make possible catastrophic effects if they are used in massive attacks.

Another technical factor involving uncertainties is the problem of fallout. The biological effects of radiation involve uncertainties, particularly in the genetic area, and we may possibly face a growing body of sober scientific judgment that the fallout hazard is greater than we now believe.

The profound over-all effect of these trends points to the great urgency and importance of our diligently and creatively seeking methods of arms limitation—limitation which will not weaken our position relative to the Soviets.

Some Technical Aspects of Arms Limitation

If progress is to be made in any kind of arms limitation (such as reducing the hazard of surprise attack or limiting the production of fissionable material or any other form of limitation), the agreements for such limitation will have to be accompanied by some form of monitoring to assure both sides that agreements are being carried out. We have so far held steadfastly to the principle that arms limitations agreements must be monitored. If we are to achieve such agreements and preserve this principle, then we will certainly be faced with the design of monitoring and inspection systems that will have to fulfill technical, military and political requirements.

While the discontinuance or limitation of nuclear tests are not, except in a limited way, disarmament measures, they do have a special importance in establishing the principle and techniques and practice of monitoring agreements. On the technical side, and perhaps on the military and political sides too, we have given more study and thought to monitoring and inspection systems for the control of nuclear tests than we have for any aspect of arms limitation. We probably have a better chance for arriving at some acceptable arrangement for monitoring and inspection of nuclear tests than we have for any other form of arms limitation. This was illustrated by the difficulties encountered in the Geneva Conference on Surprise Attack. A monitoring and inspection system for nuclear tests is likely to be simpler and more achievable than a monitoring and inspection system for surprise attack. We are not only further advanced in the technical aspects of the problem, but we have large pressures of world opinion favoring the achievement of such an agreement.

What Do We Do About the Geneva Negotiations?

Let me now turn specifically to the problems associated with our negotiations on the limitation or discontinuance of nuclear tests. We must recognize that new data secured in the Hardtack II tests, together with conclusions reached in recent studies on underground testing and on testing in outer space, leave us in a position where the system agreed to last summer at the Conference of Experts involves more substantial and significant uncertainties than believed at the time. While the Panel on Seismic Improvement has concluded that the Geneva system can be substantially restored to its originally-conceived capability, they also concluded that by deliberate concealment it would be possible to reduce the signal from an underground explosion by a factor of 10 or more and that, in theory at least, the signal might be reduced by a much larger factor than this. It is, therefore, impossible—without further tests—to give any firm estimate of the capability of the Geneva system for underground tests. This does not mean that such a firm estimate may not be achievable in the foreseeable future.

The Panel on High Altitude Detection indicated the technical feasibility of testing in outer space, as well as the technical feasibility of a system to detect these tests. Such a system would have a detection threshold which would be as high as a few hundred kilotons if the violator of the agreement resorted to very expensive measures to try to achieve concealment.

The reports of these two panels represent as good a discussion of the technical possibilities as could be achieved in the time available. They emphasize that continuing studies and experiments on a reasonable scale are needed for the further understanding of test detection and for the understanding of concealment possibilities. It was clear from the findings of both panels that attempts to conceal tests, either underground or in outer space, would be very costly.

The Concept of Probability or Calculated Risk

In our policy-making, we should recognize that no technological system is going to be perfect or absolute in its performance. We must consider the effect of such technological inspection systems in terms of probability and consider their restraining value on the basis of an estimated probability to detect within stipulated limits.

In a world of rapidly-changing technology, it may be impossible to devise fool-proof monitoring systems—either for nuclear test cessation or for other forms of agreed arms limitation. The most that these systems may accomplish will be to make evasion very costly and very uncertain. These may be the principal functions of monitoring systems. Hence, if they are to be of maximum use to us, they would have to be supplemented by highly-developed intelligence systems of our own and with appropriate military measures.

Political-Technical Factors in Geneva Negotiations

The current Geneva negotiations indicate that there are very great difficulties in reaching a satisfactory agreement with the Soviets on the critical questions of voting, staffing, and inspection. The present Soviet position on these questions is clearly unacceptable since it would eliminate even the deterrent effect of the Geneva System. We should not seek to find our way out of the present impasse in Geneva by relaxing our requirements on the critical political aspects of the control organization.

It would, therefore, seem desirable at this stage to draw back from our original efforts to achieve a system as comprehensive as that discussed by the Geneva Conference of Experts last summer and to settle for some more limited form of test agreement that would simplify both the technical and political requirements of the control organization. The most obvious possible alternative approach would be to seek an agreement which would provide for no testing in the atmosphere but would permit testing underground and at high altitudes. The timing and method of introducing this alternative is not dealt with here.

It would seem important that if we seek to agree to exclude atmospheric testing that we propose doing so by formal agreement (rather than by unilateral action) which involves some system of monitoring and which thus preserves the principle of monitoring and, perhaps, inspection and provides experience with such a system.

Such an agreement might include specific provisions for a phased, evolutionary extension of the test ban to include coverage of testing underground and at high altitudes when controls adequate to detect such tests become technically available. Research to develop this more extensive control organization might well be made a responsibility of the control organization itself.

If our decision is to proceed in this direction, it is important that we make an early technical analysis to determine what we mean by the “atmosphere.” In view of the present discussion about fallout from the stratosphere, it seems clear that a system to limit testing to the troposphere will neither solve the fallout problem nor alleviate public concern about this problem. Some radioactivity will return to earth from tests conducted out to a distance of many thousands of miles. The limit of the “atmosphere” will probably be difficult to establish.

It is also possible that an agreement not to test in the atmosphere might be accompanied by an agreement to conduct explosions in outer space only under internationally-supervised conditions, or alternatively, it is possible to define somewhat sharply the outer limits of the atmosphere and assume that unilaterally-planned testing will take place beyond that altitude. There are other kinds of technical problems that would need to be resolved, such as the conduct of explosions lightly covered with earth, or explosions under water.

It is reasonably certain, however, that these technical matters can be satisfactorily resolved from the U.S. point of view and that it would be possible to suggest a monitoring system that would be relatively simple and that would greatly reduce, if not eliminate, the requirement for inspection teams and the fear that they would be used for purposes of espionage.

However, before proposing such an agreement, I believe we should know, more clearly than we do now, the effect of this type of agreement on our own weapons program. Specifically, estimates should be made as to the additional costs involved and the usefulness of data obtained if our test program were to continue at its present rate and all tests were conducted either underground or in outer space. In addition, we should examine carefully the effect of an atmosphere test ban on our anti-intercontinental ballistic missile weapon system testing.

Sustained Disarmament Studies

The complexities inherent in understanding the monitoring of arms limitation agreements and the great importance of our achieving a thorough understanding of what is to our advantage or disadvantage suggest the importance to the United States of undertaking systematic and sustained studies of the technical, military, and political aspects of arms limitation. Our experience in the Geneva Conference on Reducing the Hazards of Surprise Attack pointed up the deficiencies of ad hoc and hurried preparation for such negotiations.

Even though an agreement for limited nuclear test cessation remains our first goal, we should look beyond this. We might well direct our studies and planning toward a possible reduction in the advanced means or delivery of nuclear weapons. This may be a more practical objective than control of nuclear weapons stockpiles and production.

A small but important beginning was made in this direction by the Geneva Conference on Surprise Attack. The failure of this conference should not deter further efforts. It seems especially urgent that we proceed with further studies of ways to reduce the hazards of surprise attack.

We must also undertake studies and experiments to improve our capacity to detect tests other than in the atmosphere (as, for example, in outer space), even if we achieve no agreements. The research and experiments recommended by the Panel on Seismic Improvement and the Panel on High Altitude Detection should be carefully considered and followed up as a part of our long-term effort to advance the technology of detection.

Summary

The trends in military technology, together with the threat of catastrophic war, in the continuing development of nuclear weapons systems emphasize the overwhelming importance of seeking sound ways of limiting armaments. The possibilities of uncertainties in scientific estimates of biological effects of radiation hazards should be kept in mind in formulating policy.

Since any kind of arms limitation will probably have to be accompanied by a monitoring system, it is important to establish the principle of monitoring and inspection and to achieve an agreement which will give us experience in monitoring. The monitoring of nuclear tests has received more study and is more thoroughly understood than any other arms limitation monitoring. This is an added reason for seeking an agreement for the limitation of atomic tests.

It would seem technically feasible to achieve a sound agreement that would involve the stoppage of tests in the atmosphere, this stoppage to be subject to monitoring through an agreement. Such an agreement might provide for the evolutionary development of improved detection systems for underground and outer-space testing. If we are to seek this kind of modified test limitation agreement, we should at once clarify the technical premises for such an agreement.

We also need to organize and pursue on a sustained basis creative efforts to understand the technical, military, and political aspects of arms limitation of other kinds than nuclear tests cessation.

D.E.

1. Source: Technical factors relating to arms limitation and to the Geneva conference on nuclear test cessation. Secret. 11 pp. Eisenhower Library, Whitman File, Administrative Series, Killian, James R., 1957.