Foreign Relations of the United States, 1961–1963, Volume VII, Arms Control and Disarmament

47. Letter From Secretary of Defense McNamara to the President’s Adviser on Disarmament (McCloy)⁰

Washington, July 28, 1961.

Dear Mr. McCloy: I recommend the Committee of Principals propose to the President that the United States initiate preparations to resume nuclear weapons testing.

The Department of Defense supports the recent series of U.S. actions to explore all reasonable means of reaching an acceptable agreement in the Geneva Conference on the Discontinuance of Nuclear Weapon Tests, and is in accord with continuing the negotiations. It now seems evident, however, that the Soviet Union is not interested in a treaty to discontinue nuclear weapons tests except on terms unacceptable to the U.S. This was made clear in the Soviet aide-mémoire of 4 June¹ and subsequently by the Delegate in the Conference. It therefore is timely to consider making a judgment on the resumption of testing, as the President indicated in his public statement of June 8.

In developing a recommendation, three sets of considerations must be examined:

A. The payoffs which the U.S. and U.S.S.R. may expect from testing.

B. The U.S.-U.S.S.R. comparative positions before and after U.S. resumption of tests.

C. The political implications of resumption of testing.

A. The payoffs which the U.S. and U.S.S.R. may expect from testing.

The following objectives of nuclear weapons testing are considered to be most critical for the U.S.:

1. Pure fusion weapons.

[less than 1 line of source text not declassified] pure fusion weapons, which would be particularly applicable to limited warfare, but which with further development would have general application to the whole spectrum of offensive and defensive warfare. [13 lines of source text not declassified]

2. Weight-yield ratio.

Testing would result in a spectrum of reduced size and weight weapons, without undue sacrifice of yield, for particular application to strategic missile systems with improved survivability. [1-1/2 lines of source text not declassified] They offer the prospect of low cost lightweight strategic missile warheads for delivery by mobile systems (land, sea, and air) and for cluster warheads to improve penetrability, thereby plugging possible soft spots in second strike systems and improving our deterrent posture. In turn this could improve the capability to limit and control nuclear conflict. This field of development has been one of the most rapidly changing areas of weaponry.

3. Weapons effects.

Tests would result in gaining more information on weapons effects, especially to determine the lethal ranges and characteristics of kill mechanisms to be used against ballistic missiles. Knowledge of this kind can be of critical importance both for examining possibilities for defense against ballistic missile attack and for developing penetration capabilities for use should the Soviets deploy an AICBM system. [1 line of source text not declassified] important to determine the kill mechanism for AICBM’s and in deciding how our ballistic missile attacks must be planned including the nature of the re-entry vehicle, degree of saturation and the necessary separation between warheads. Other important objectives of effects tests are: The vulnerability of our own warheads and carriers to nuclear bursts; the effects of nuclear bursts on electronic equipment and radio and radar propagation; the acquisition of design data for hardening underground and other structures; and the kill radii and safe delivery ranges of our weapons.

4. Proof testing.

To assure that new weapons entering stockpiles to meet critical needs will function properly. Testing by functional tests underground at fractional yields will provide assurance that weapons not tested as finally designed will function properly, e.g., warheads for Polaris, Minuteman and Skybolt. Presently planned warheads have all been certified by the Atomic Energy Commission as having the same assurance of functioning as those which have been tested; further developments in the absence of testing could result in loss of confidence in the stockpile.

B. The comparative U.S.-U.S.S.R. positions before and after U.S. resumption of tests.

The situations before and after U.S. resumption of tests are difficult to compare because the U.S. does not know whether the U.S.S.R. is or is not now testing. Three situations, therefore, must be examined:

1. The proposed situation.

It is assumed that if the U.S. resumes tests the U.S.S.R. will also test. With both sides testing, there is no reason to expect U.S. weapons development to proceed on terms which are worse than even. Because of the asymmetry of the strategic positions of the two countries, it is likely that the advances made will be more important to the U.S. than to the U.S.S.R.

2. One view of the present situation: Neither side tests.

If neither side tests, weapons development in both countries will be hampered. The Soviets would be denied some developments useful to them (e.g., smaller high-yield warheads which could make a more mobile deterrent force, to the extent that they are interested in deterrence). The U.S. would be denied the developments listed in A above, which the DOD considers to be of great importance to U.S. strategic and tactical capabilities. Moreover, because of the differences in conditions of secrecy prevailing in the two nations, the Soviets would over time have an advantage in acquiring U.S. secrets. And, since the line between “laboratory experiments” and “testing” is not a clear one, the Soviets are less likely to refrain from significant “experiments” which the U.S. might call “tests”. These factors could lead to weapons advantages for the Soviets.

3. A second view of the present situation: The Soviets test clandestinely while the U.S. abstains.

If the U.S. refrains from testing while the Soviets conduct clandestine tests underground and in space, the U.S.S.R. will make significant advances not available to the U.S. in the critical areas of weapons development mentioned in Part A above. The results of Soviet tests could be important and varied. If the U.S.S.R. achieved a major breakthrough, especially on the area of AICBM (where there have been indications of Soviet interest and activity), the U.S. might have its deterrent missile force emasculated with little warning. Similarly, a composite of developments of the kind which the Soviets might be reasonably expected to achieve through a program of clandestine testing could, in the absence of parallel development by the U.S., degrade or possibly even negate U.S. strategic capability based on present weapons technology. Further, a major even though not decisive advance by the U.S.S.R. could be a strong political asset and could give them a formidable advantage at the conference table. It may be argued that advances which the Soviets could make clandestinely in a few months or a year cannot unbalance the strategic situation. However, this question should be considered in the light of the almost three years during which the Soviets have been able to test clandestinely if they wished. In three years, one or even two new generations of weapons development have occurred in the past. In any event, this situation—in which the U.S.S.R. tests while the U.S. abstains—has grave implications for U.S. national security and for the security of our allies.

C. The political implications of resumption of testing.

The political issues involved in a decision to resume testing, while not the province of Defense, are clearly serious. The following political considerations deserve special mention:

1. The adverse reaction of some nations may be expected and must be evaluated and taken into account.

2. The resumption of tests will probably not have a significant effect on the Nth-country problem. At this time it appears that even the unlikely situation of the successful conclusion by the U.S., U.K., and U.S.S.R. of a test ban agreement will not lead to the adherence of France and Communist China, unless these countries have previously achieved a nuclear weapons capability satisfactory to them. How many other countries would adhere that are otherwise likely to develop nuclear weapons is uncertain, but the number is probably small. On the other hand, in the absence of a treaty, it is doubtful that resumption of testing by the U.S. would cause proliferation of testing, especially if the U.S. maintains a consistent policy of discouraging diffusion.

3. Continuation of a voluntary and uncontrolled moratorium on tests puts the U.S. in a weak position to press for adequate controls in the disarmament negotiations and in connection with other arms control measures; it sets a bad precedent.

4. If there is any remaining chance of obtaining a satisfactory test ban agreement, resuming tests is more likely to help than hinder the chance. A continuing uncontrolled moratorium removes any incentive for the Soviet Union to conclude a treaty or to negotiate seriously on an acceptable control system.

5. Most important—especially in view of the impending Berlin crisis—is the role played by resumption of tests as a demonstration of U.S. resoluteness. The U.S. has stated that it would not continue indefinitely under an uncontrolled moratorium. This is an appropriate time to demonstrate that the U.S. means what it says.

D. Conclusions.

If the U.S. could be sure that the U.S.S.R. were not testing (and if all political considerations were set aside), the justification for resumption of testing would be weaker. However, the crucial fact is that the U.S. does not know whether or not the U.S.S.R. is testing. The U.S. does know:

• —that the motive to test exists (it may be assumed that the possibilities for advancements along some or all of the lines mentioned in Part A above are open to and known to the U.S.S.R.);
• —that there have been unexplained activities which could be indicative of testing at Soviet sites suitable for underground testing, and that there has been a continuing association between weapons development laboratories and Soviet organizations known to have been associated with testing programs prior to the moratorium;
• —that underground tests could have been performed in the U.S.S.R. with next to no risk of detection; and
• —that the Soviets have consistently rejected proposals for arrangements which would give reasonable assurance that they are not testing.

The first judgment therefore is that the U.S.S.R. may now be testing or may be planning to test clandestinely. The second is that the chances are substantial that such one-sided testing will lead to grave consequences for the security of the U.S. and her allies. And the third is that the risks involved in the resumption of tests are less than the risks involved in continuation of the unpoliced moratorium.

After consideration of all aspects of the problem, including the political factors, I recommend that the U.S. complete preparations to resume nuclear weapons testing, and collaterally prepare for a Vela program of underground shots for seismic research. The latter are essential if we ever are to achieve an improved underground detection and identification system, either national or international.

Assuming that the decision is made to resume testing, the timing of announcements, and of the actual shot schedule, will be significant. The AEC and Defense have developed a proposed program of tests, based on priority of need for the data and taking into account practical readiness dates for various shots. This is attached² and discussed in more detail in Par. E-4.

E. Recommended sequence of events.

This proposed sequence consists briefly of the following steps:

1. An early announcement that preparations for nuclear tests will start at once, but that the decision to conduct tests will depend on the results of test ban negotiations over the next five or six months.

2. Physical preparations for underground tests at the Nevada Test Site and continuation of test ban negotiations.

3. If treaty negotiations are unsuccessful, announcement that tests are being resumed.

4. Early resumption of underground tests.

Each of these steps is discussed more fully below.

1. Announcement that preparations for tests have been ordered.

There should be an announcement, on, or shortly after, August 1, that the President has instructed the AEC and Defense to get nuclear test sites in readiness, as a decision to conduct underground tests may be required for reasons of national security.

The ground work for this announcement was laid in the U.S. reply (released June 17) to the Soviet aide-mémoire.

It is recognized that any announcement of preparations for resumption of testing should be accompanied by a continued intensive program by State and USIA to gain U.S. and world support. The announcement explaining U.S. preparations may well be short and to the point, along the following lines:

“The U.S. has done everything in its power to negotiate a treaty banning nuclear testing with adequate safeguards. The U.S.S.R. does not want a test-ban treaty and has not negotiated in good faith. They consistently reject proposals for arrangements which would give reasonable assurance they are not testing, and they may be testing. If the U.S. resumes testing, it will be because the irreducible minimum security needs of the free world require it. The AEC and the DOD have been instructed to make preparations for nuclear weapons tests as soon as possible. The first important tests cannot be conducted for five or six months, and the U.S. does not want to test except as national and free world security dictates. In the interim, the U.S. will continue to negotiate in Geneva towards reaching agreement on a treaty banning nuclear weapons tests. Should a satisfactory agreement be reached, preparations for testing will be discontinued. The U.S. is preparing sites for underground tests involving no fall-out hazard. The decision whether testing will actually be conducted will depend on events in the next few months.”

The preparations to test should be announced publicly rather than kept secret for two reasons:

a. It is consistent with past policy of disclosure (and the news could probably not be suppressed anyway).

b. Such an announcement may be consistent with other steps being taken in connection with the impending Berlin crisis.

The separation of the decision to make preparations to test and the decision to test correctly implies that the latter decision is genuinely left open. The decision to test will depend on conditions at the later time.

2. Preparation for tests.

Detailed preparations for a schedule of tests along the lines of the attached program would of course immediately follow the announcement. The program by State and USIA to gain world-wide public support should be continued and intensified during the interval between announcement of preparations and resumption of testing.

The six months preparation period will cost the AEC about $25 to $30 million and the Defense Atomic Support Agency about $6 million, while the whole two-year series of tests, including preparation, would cost the AEC about $315 million and the DOD about $150 million, not including the cost of nuclear material in the various devices. More than half of the DOD’s $150 million estimate represents the cost of high altitude and space tests, including the necessary boosters and instrumentation.

The first two weapons effects tests (Nos. 4 and 15 in the attached table) which are the responsibility of the DOD, have been studied in considerable detail. The Hard Hat shot (No. 4) will give engineering data on the design of hardened underground structures, on which we are now spending hundreds of millions of dollars without sufficient knowledge to insure that the engineering designs are adequate, and provide seismic data to replace one of the scheduled Vela shots. About $2 million has already been spent on Hard Hat in constructing a series of tunnels in hard rock, lined with different types of materials to be subjected to very high blast forces. Instrument and cable placement can begin about August 1st and will require several months’ time and an additional $1 million. The data could be obtained by exploding 1,250,000 pounds of NE, but this would cost an additional $1 million and would not satisfy the Vela requirements.

The second effects shot (No. 15) must be nuclear, as it will give simulated data on X-ray and other nuclear kill mechanisms against ICBM warheads above the atmosphere, provide data on the kill radius of Nike Zeus, and test the response of nose cone and other materials to nuclear bursts. Many of the questions about ICBM effectiveness and vulnerability should be answered or greatly simplified by this and the preceding effects test. The tunnel and shaft are already constructed. An 800 ft. steel vacuum pipe should be ordered immediately, followed by design, construction and placement of the pipe, targets and instruments. An additional $5 million will be required.

No detailed estimates of the necessary preparations have been made by the AEC for their development shots, pending a decision to prepare for nuclear testing, but a total of about $25 to $30 million will be required during a six months preparation period. This would cover the digging of additional shafts and tunnels, the installation of cable, diagnostic firing and timing equipment and other preparations of a detailed nature that are now under study.

3. Announcement that tests are being resumed.

Assuming that conditions have not materially changed, there should be an announcement shortly before the first test is to be made, of the resumption of tests. It should be explained that the tests contemplated in the first series are underground and involve no fall-out hazard.

The announcement would be similar to the one made when preparations were begun. This time, however, it should conclude in the following vein: “… Time has run out. We said we could not abide an uncontrolled moratorium. And we will not. Underground tests will be resumed____.”

4. Resumption of tests.

The first of the series of tests should be made very soon after the announcement, and a demonstrably significant series of tests should be carried out. The attached tables show such a proposed series of 44 development, effects and proof tests prepared by DASA in cooperation with the Division of Military Application of the AEC, and representing the best current judgment of what could and should be done over a period of about two years following a six month period of preparation. Certain changes may have to be made as the results of preceding tests become available. In addition, 10 underground safety, or “one point detonation” tests, are proposed which would produce little or no nuclear yield, and could therefore be carried out concurrently with the others, as convenient.

The first six months of actual testing would be carried out underground, with essentially no contamination of the surface or the atmosphere. Beyond that period, 10 tests (marked with an asterisk in the table) might be made in and above the atmosphere, on the surface of land or water and underwater, but designed to produce a minimum of radioactive contamination. [less than 1 line of source text not declassified] These 10 shots are considered of lower priority than the others, and would be omitted if political considerations so dictate, but have been inserted in the list in logical time sequence. The last 6 shots would be either at very high altitude or in outer space. They are most important [1-1/2 lines of source text not declassified] and also to obtain information on radar and radio black-out caused by nuclear bursts at very high altitudes. The only two high yield shots of this nature, fired in the last test series, demonstrated the need for much additional information about these important and potentially serious phenomena. They have been scheduled last only because of the extensive preparations required for testing at very high altitudes or in outer space.

There is also attached the present schedule of Vela underground shots. The proposed weapons tests would delay the conclusion of the Vela Uniform program by 6 months to a year, and increase its cost by possibly $4 million. However, a considerable amount of seismic data could be obtained from the weapons shots. The DASA Hard Hat shot could eliminate the need for the Vela Lollipop event. In fact it is possible that during such a series of tests electromagnetic or other phenomena may be found that would prove superior to seismic effects as a means of detecting underground nuclear explosions. Alternative schedules involving a mixture of the two programs can also be carried out.

[2-1/2 lines of source text not declassified] No specific test [less than 1 line of source text not declassified] is listed because it could not be built until more data are available, but it would be tested as soon as possible.

Three other tests in this first six months series of 15 shots would be devoted to securing a higher yield-to-weight ratio in weapons producing [less than 1 line of source text not declassified] KT. The remaining seven shots would verify performance and prove the yield of new weapons going into stockpile where there is some doubt about the final design or where the yield is particularly important for small tactical weapons such as Davy Crockett and the air delivered missiles, Falcon, Terrier and Bullpup, where safe firing distances must be determined.

The remaining tests in the list would continue with advanced development and proof tests in about the same ratio as during the first six months, to answer questions that have arisen so far. Obviously other questions will become critical as the tests proceed.

While it is not contemplated that any tests in the first six months period would be in the atmosphere or underwater, the U.S. should not commit itself in any public statements to refrain indefinitely from testing in any environment. We should retain freedom of action, should the Soviet Union test in those environments, or should the military value of such tests later be determined to outweigh the possible adverse political effects of producing a small amount of local radioactive contamination.

The United States should continue to reaffirm the U.S. desire for an effective test ban treaty and should endeavor to keep negotiations going in Geneva.

I am sending copies of this letter to the Secretary of State, Chairman, AEC, Special Assistant to the President for Science and Technology, Director, CIA, Special Assistant to the President for National Security Affairs, Director, USIA, and Chairman, JCS.

Sincerely,

Robert S. McNamara³

0. Source: Kennedy Library, National Security Files, Departments and Agencies Series, ACDA, Committee of Principals, 7/27/61-7/28/61. Top Secret; Restricted Data.
1. See footnote 1, Document 31.
2. The enclosure, entitled “Proposed Test Program” is not printed.
3. Printed from a copy that indicates Secretary McNamara signed the original.