Foreign Relations of the United States, 1949, National Security Affairs, Foreign Economic Policy, Volume I

PM Files

The Chairman of the Joint Committee on Atomic Energy (McMahon) to President Truman

top secret

Los Angeles, November 21, 1949.

Dear Mr. President: The profundity of the atomic crisis which has now overtaken us cannot, in my judgment, be exaggerated. The specific decision that you must make regarding the super bomb is one of the gravest ever to confront an American president. This letter, reflecting ideas stimulated by a number of recent conferences which I attended at Washington, Los Alamos, Argonne, Hanford, and elsewhere, is written in sincere hopes of being helpful to you. It is easily the longest letter that you have ever received from me, and I apologize in advance for its length, although the gravity of the subject justifies extended discussion.

Those who oppose an all-out “crash” effort on the super impress me as being so horrified at the path down which the world is traveling that they have lost contact with common sense and reality. Of course I can understand and share in their horror. In a moment I will try to show the good that may come of the revulsion which any moral individual must experience at the prospect of bringing forth a weapon such as the super. But first it may be useful to attempt an exposé of what I regard as the false, horror-inspired logic put forward by those who recommend against this project.

They stress that the super is unique and that it differs in kind, not merely in degree, from ordinary atomic bombs; and yet they argue that its military worth is dubious and that ordinary atomic bombs could retaliate adequately against an enemy who used it upon the United States. Here is a fundamental inconsistency. If the super would accomplish no more than weapons already in our arsenal, why single it out for special objection? If, on the other hand, the super represents a wholly new order of destructive magnitude—as I think it obviously does—then its military role would seem to be decisive. Consider that about 23 current-type fission bombs would be needed to duplicate the effect of one super which destroyed 150 square miles; about 143 fission bombs would be needed to equal the effect of one super that destroyed 1,000 square miles; and the ratio increases still further assuming, as the scientists suppose, that the super is unlimited in potentiality and might take out far more than 1,000 square miles. If an enemy employed a few supers against us, our entire stockpile of ordinary atomic bombs might fall pitifully short of inflicting an equal damage area upon him. Moreover, the expense of developing this weapon is estimated at only $200 or $300 millions—less than a sixth of what we spent upon the wartime Manhattan project—and unit cost, as was true of the fission bomb, may be expected to decline markedly when production and design improvements are achieved.

I am not a professional strategist, but various military possibilities inherent in the super seem clear even to a layman. An attacker who dropped a half-dozen supers upon the opponent’s largest industrial areas would free the hundreds of fission bombs otherwise needed to do the same job for other purposes. These fission bombs could therefore be used simultaneously against air and sea bases, as well as smaller industrial objectives. Thus the attacker would compress within a few hours or days several times the total punishment administered to Germany throughout the entire span of World War II. Equally important, the fission bomb strikes upon air and sea bases—made possible because supers were available for use against wide-area targets-would severely reduce the opponent’s capacity to retaliate. The attacker might escape with comparatively little injury to himself. Lacking supers, however, he could not inflict nearly so much damage using fission bombs alone; and such damage as he did inflict would require more time—giving the opponent opportunity to adjust, to recover from shock, to tighten defenses, and to retaliate in force.

I am aware that numerous TNT bombs, accurately distributed and equal in cumulative blast effect to one fission bomb, actually create greater havoc than a single atomic explosion. Similarly, 25 or 100 or 200 fission bombs, properly placed, might exceed the effect of one super bomb. But to stress this point it [is] to overlook the shock and demoralization, psychological and otherwise, that follow from concentrating an offensive within the shortest possible space of time. Also a fission bomb must usually detonate a mile or half-mile or even less distance from the target to be effective, whereas a super might miss its target by ten miles or more and still serve the purpose intended. The problem of deliverability is not an easy one, but it can be overcome through sufficient effort—just as the once difficult problem of delivering the Hiroshima and Nagasaki bombs was overcome. One estimate given me as to how much a super might weigh is less than the maximum pay load of the present B–36 airplane. In all probability the delivery of a super would be easier and tax our resources less than delivery of an equivalent number of conventional atomic weapons. Therefore, if it was militarily worthwhile to build the fission bombs dropped upon Japan, even though a large fleet of planes carrying fission bombs might do somewhat comparable damage. [sic]

I am likewise aware that Moscow and Leningrad, the two chief cities of the sole potential enemy in sight, are only said to occupy areas of about 120 and 110 square miles respectively. Thus, runs the anti-super argument, there are scarcely more than two or three urban targets in all Russia which measure up to the tremendous destructive power of this explosive, and they could be thoroughly attacked with ordinary atomic bombs. But the argument admits that at least two or three urban targets tailored to the super do in fact exist—and rapid Soviet industrialization may soon increase the list. Air bases and isolated factories surrounding these same targets might succumb to a super although they would not succumb to a fission bomb attack. Air bases and isolated factories in the vicinity of other, smaller cities—as well as medium-sized towns or built-up areas located within ten, twenty, or thirty miles of one another—might also supply objectives against which the use of a super would be economical. There are additional targets, many of them tactical, which only the super might successfully destroy: a fleet, troops in the field, forces preparing for amphibious invasion, guerrilla fighters, an underground storage site containing fissionable material, key segments of a radar warning network, the Baku and Caucasian oil fields, a string of air strips, submarine pens, missile launching sites, fortifications around a narrow water passage such as the Dardanelles, or atomic production facilities grouped in the same general area but too far apart for efficient employment of fission explosives. Even further, the Hiroshima and Nagasaki bombs did not themselves defeat Japan, but they accelerated recognition by her that she could no longer resist—and supers might perform a like function in the future.

It is claimed that the super would inevitably exterminate hosts of civilians and therefore constitute a pure weapon of genocide. But aside from the exclusively tactical purposes which it might serve, its use against cities could be preceded by a warning to the inhabitants that they must either evacuate or suffer the consequences. If the warning were heeded, bomb damage would be confined to physical buildings and plants; and the disruption generated by evacuees moving from city to country might hamper the enemy’s war effort more seriously even than mass casualties. The basic question, however, is this: what happens if supers are aimed at New York, Chicago, Los Angeles, and Washington? Will we possess our own supers, ready to retaliate in kind and to throttle the attack at its source—or will we lack such weapons and suffer defeat and perhaps utter annihilation as the result?

There is no moral dividing line that I can see between a big explosion which causes heavy damage and many smaller explosions causing equal or still greater damage. Where is the valid ethical distinction between the several Hamburg raids that produced 135,000 fatalities, the single Tokyo “fire” raid that produced 85,000 fatalities, and the Hiroshima bomb that produced 65,000 fatalities? What, then, is the distinction between the 1,000 square miles which one super might scorch and the 1,000 square miles which 143 fission bombs might equally destory? Is a given weapon to be adjudged moral or immoral depending upon whether it requires hours, days or weeks to take its toll? … Modern warfare, even if waged with pre-atomic weapons only, is the real instrument of genocide—not a single agent like the super. The havoc which Germany visited upon Russia and Western Europe and which the Allies visited upon Germany and Japan during World War II probably surpasses the destructiveness of a dozen supers.

The contention is made that a war involving the super would leave behind such chaos and vengefulness as to create a worse situation, with a darker outlook for lasting peace, than the one existing at present. Perhaps this is so. Yet our first duty consists in doing what is necessary to win. We know that because of manpower limitations and the oceans that separate us from Eurasia, we could not use surface forces to invade and occupy Russia. The only choice left open is heavy reliance upon strategic air power, despite our own immense vulnerability to nuclear weapons. The super should end all debate as to whether or not strategic air power could win a war. Without American victory—which supers alone might render feasible—there would be no post-war existence for our country, much less post-war problems. I might add that, to my mind, almost nothing could be worse than the current atomic armaments race and that victory in a future war, whatever its sequel in other respects, would at least assure effective international control over weapons of mass destruction.

A number of reasons for opposing the super are advanced that impress me as make-weight arguments. Radioactive contamination resulting from a super explosion is one such reason. The advices I have received are to the effect that contamination would merely correspond to the aftermath of ordinary fission bombs and that, in any event, it might be reduced through the use of appropriate “casing” materials. The super is said to promise little or no progress in the field of peaceful applications—even though this contention flouts all past experience and even though, to the best of my knowledge, there never has been research and development of any type or character that failed to increase human capacity for constructively harnessing the forces of nature. It is argued that if we build a half-dozen or more new reactors and use them to manufacture the tritium necessary for supers, we would forego the plutonium otherwise obtainable from those same reactors—thereby losing about as much as we gained. Yet the best figures which I can assemble show.… Realistically speaking, moreover, the half-dozen new reactors would not be built rapidly without a super program to spur along their construction—so that, absent the program, we would not soon acquire much additional plutonium anyhow.

Opponents of the super imply that as much as five years would elapse before an all-out effort produced results and that success is only a little more likely than failure. This, I fear, represents an attempt to dampen feelings of urgency. My information, in contrast, is that Teller,¹ Oppenheimer, and Bethe² made the first calculations respecting a super bomb in 1942; that the matter has been under study ever since; that no reason for anticipating failure has emerged from such study, that, on the contrary, success seems highly probable; and that intensive work short of wartime methods might reach fruition in early 1952.

The Atomic Energy Commission weapons program, as formulated more than a year ago, envisaged achievement of a super by about 1958. No one in authority protested against the program; no one argued that problems affecting the super should be left unexplored. Now that Russia has broken our atomic monoply three years sooner than we had expected—giving us compelling reason to speed up the rate of American development—the anti-super counsellors adopt a reverse logic. Instead of agreeing that the new situation pressures us to accomplish in two years what all along we planned to accomplish in eight, they would either mark time or abandon the project together. Their advice, if followed, would also have the effect of placing a ceiling upon our military advancement; for I do not know how the Los Alamos Laboratory would occupy itself, after a few years had passed, unless it ventured far into the thermo-nuclear field.

Equally surprising is the suggestion that we refrain from pressing the super if Russia pledges herself (without any inspection or control) to follow a similar policy. Since mid-1946, we have urged the world to accept our view that atomic weapons are so unique and dangerous that only far-reaching international measures will afford protection. The need for a tight control system of the kind approved by the UN General Assembly is now intensified, since the secret manufacture of just one fission bomb might furnish an initiator for a super and since the necessary tritium might be produced in dozens of small reactors difficult to locate through inspection. Yet—with more cause than ever before to insist upon strict safeguards—it is proposed that, as regards supers, we repudiate the position taken during the past three years and depend upon a mere paper pledge.

… If an actual, full-fledged super were tested, the gigantic blast effect could hardly help but make itself known to us. But I am told that tritium and deuterium, after a thermo-nuclear reaction, would not register upon distant monitoring instruments.… Apart from these points, I fail to see the special advantage of knowing when Russia has nearly acquired supers if we were so far behind that she would achieve them first regardless.

To me the notion that our possession of this weapon would harm our moral position makes no sense, provided that we offered to relinquish it in exchange for a just and enforceable system of control. Only the nation which rejected such an offer would occupy an indefensible moral position. Any idea that American renunciation of the super would inspire hope in the world or that “disarmament by example” would earn us respect is so suggestive of an appeasement psychology and so at variance with the bitter lessons learned before, during, and after two recent wars that I will comment no further.

But I do think it important to challenge the complacent attitude toward Soviet progress that pervades the thinking of those against the super. Some of them are the very ones who preached, from 1945 onward, that Russia would soon achieve the atomic bomb through her own independent effort; and yet they now ignore the logic which led them to this correct conclusion, saying that Soviet achievement of the super may well be a decade away, if not longer. They speak of our taking the initiative in super development—just as though such a weapon has never occurred to Russian scientists, just as though we dare assume that the Soviets are not working toward it with all haste, and just as though American and British intelligence had not lately underestimated Russia to the extent of missing by three years the date of her first fission bomb test.

Only about one-fortieth of our total military spending since 1945 has been devoted to atomic weapons. In fiscal 1950, funds for that purpose will comprise less than one per cent of the national budget. These amounts might impress us as inadequate, to say the least, if we could know the precise extent of Soviet commitments.… German scientists imported into Russia have long studied the heavy-water reactors well suited to tritium manufacture; Kapitza, the great Soviet physicist, is a specialist in low-temperature problems associated with liquid deuterium; densely populated American cities are made-to-order for an explosive that levels hundreds of square miles; communism suffered a prestige loss when “decadent capitalists” completed the earliest atomic bombs, but this loss would be more than recouped should Russia complete the earliest super; and if the Kremlin believes that it cannot out-produce us in ordinary fission weapons, then its logical strategy is to excell in the thermo-nuclear field. All such factors as these warn that complacency could be fatal. As one noted American scientist has said, we should imagine a force of Soviet planes and submarines, each carrying a super and each poised to effect delivery at our inland and coastal cities—and with that picture in the forefront of our minds move heaven and earth to gain the super first.

Although any other decision would almost guarantee disaster for our nation, in my opinion, I thoroughly agree with the opposite school of thought that armaments races lead to war. I agree that ours are the cities most threatened by the super. I agree that if war comes—and if it is postponed until Russia accumulates a stockpile of supers and fission bombs, or even fission bombs alone—Western civilization may well crumble whether we win or lose. I consider the most significant single aspect of the present situation to be this: with each day, week, and month that passes, the Kremlin acquires an added supply of fissionable material. My thesis, however, is that if we let Russia get the super first, catastrophe becomes all but certain—whereas, if we get it first, there exists a chance of saving ourselves.

That chance can best be grasped, as I see it, by immediately taking the entire problem of the super to the people of the United States and the world. The existence of an all-out American effort to build this weapon could not be kept secret in any case: the purchase of lithium and other key materials in great quantity, the construction of special new reactors, the presence of Teller and, I hope, Fermi, Bethe, and equally famous scientists at Los Alamos, the extensive literature already published about the super, the question of Congressional appropriations, and the need for British and Canadian assistance would inevitably alert Moscow to the true facts. But even if our effort could be successfully concealed, the people of New York are entitled to know that an innocent-appearing merchant vessel registered under the Dominican flag might introduce a Russian super into their harbor and destroy, not just a few square miles, but their entire metropolis. The people of Chicago are entitled to know that a robot-controlled bomber, located only a few hours away in Kamchatka, might visit their city at night and leave it a wasteland. The people of Washington—when they discuss civil defense against atomic warfare—are entitled to know that organization of a disaster corps, dispersion of hospitals, and suchlike measures would be futile in coping with the super weapon which Stalin may own within three years, two years, or even less. What is more, the people of Russia are entitled to know of the suicide terminal point toward which the refusal of their rulers to accept international control is leading them.

Since this issue, involving as it does the survival or extinction of whole populations, transcends all others in importance, it should be treated in the most important possible manner. If I may make the suggestion, you have an opportunity to deliver an address before the current United Nations Assembly that might alter the destiny of mankind. You could explain to the world’s peoples, through the UN forum, that we are able to build the super and that Russia possesses a like capability. You could point out that the possibility of ravaging 1,000 square miles at a single blow does not decrease the danger from ordinary fission bombs but that it dramatizes and renders still more urgent the need for effective international control. Thus the horror and revulsion which the super inspires in moral beings might be harnessed and made to generate a world-wide pressure of public opinion upon the Kremlin to accept a sane and worthwhile control plan.

Sincerely yours,

[File copy not signed]

1. Dr. Edward Teller, Professor of Nuclear Physics, University of Chicago; participated in the atomic bomb development program at Los Alamos Laboratory, 1941–1946.
2. Dr. Hans A. Bethe, Professor of Nuclear Physics, Cornell University; participated in the atomic bomb development program at Los Alamos Laboratory, 1943–1946.