Commentary Magazine

The Question of Civil Defense-A Debate

In the hope of contributing to a clarification of the whole question of civil defense by bringing into focus the precise points of disagreement between the two main contending positions, we invited Herman Kahn (perhaps the leading advocate of a more intensified civil defense effort) and Erich Fromm (who has become one of America’s most influential spokesmen for disarmament and whose collaborator in the present debate, Michael Maccoby, has been a prominent participant in the peace movement) to argue their respective cases for the readers of COMMENTARY. The two articles that follow were written independently, though the authors of course had access to each other’s previously published statements.


Herman Kahn’s On Thermonuclear War has, since its appearance in 1960, earned itself a secure place among the most controversial books of our time. Formerly on the staff of the RAND Corporation, Mr. Kahn is now director of the Hudson Institute (a non-profit research organization concerned with problems of national security and international order). His new book, Thinking About the Unthinkable, is scheduled for publication in the spring by Horizon Press. Erich Fromm, the distinguished psychoanalyst and social critic, has recently been devoting much of his time to the study of the current international crisis (which forms the subject of his latest book, May Man Prevail?). Michael Maccoby holds a Ph.D. in social psychology from Harvard and has taught both at Harvard and Chicago.



Herman Kahn: Nuclear war may be unthinkable, but it is not impossible. Obviously first priority ought to be given—and is being given—to the objective of deterring or otherwise avoiding a nuclear war. But because war is not impossible, we also have an obligation to allocate at least a modest proportion of our intellectual and material resources to the objective of alleviating the potential results of a nuclear conflict, in the event that one should occur. The fact that this is only a second-priority objective does not mean—as some people seem to think—that it is unimportant or that we have any cause to neglect it.

Of the many measures that the United States has been taking to reduce the harm that we might suffer in consequence of a nuclear war—measures intended not only to save lives and property, but also to facilitate recuperation—civil defense was until very recently underemphasized almost to the point of negligence. A much greater degree of attention has been paid to improving our active defense (antimissile and anti-aircraft systems), to strengthening our strategic forces, and to developing a more adequate system of Command and Control (including preparations to limit or terminate a war). To be sure, these measures also contribute to the double aim of saving lives and property and facilitating recovery, but they need to be complemented by an intelligent shelter program and a careful series of plans designed to cope with the medical, economic, social, and political problems that might arise in a postwar world. Indeed, if one evaluates our over-all military posture by its ability to protect people and property in the event of a nuclear war, it becomes clear that civil defense now deserves increased emphasis relative to the other things we have been doing.

In considering civil defense against nuclear weapons, we enter a field which is, in a critical sense, new: there is no adequate experience; no one has fought and survived more than a comparatively small and one-sided nuclear war. If, therefore, we wish to understand what the existence of these weapons of unprecedented destructiveness may mean for us, we have no choice but to rely on theoretical analysis and extrapolation, while trying to relate our theories as closely as possible to the known facts and lessons of the real past. So self-evident is this statement that it would hardly be worth making, except for the fact that a great many people regard nuclear war as “unthinkable” and refuse to think about it at all.

This refusal to “think about the unthinkable” takes a variety of forms, among them a primitive escapism (pretending that the terrible danger we are living under simply does not exist), the construction of a crude mirror-image (“I don’t want to hurt anyone—why should anyone want to hurt me?”), and even a deliberate overestimation of the horror of war that acts to justify a feeling of hopelessness and apathy. I must admit, a bit reluctantly, to believing that even many of the “realistic” and sophisticated objections to civil defense that have been made by its opponents stem from roughly the same escapist motives. Nevertheless these objections have been put forward with great force and intensity; they deserve to be confronted and analyzed as arguments, whatever the motives of the objectors may or may not be. On what, then, do the opponents of civil defense base their case?

According to one prominent school of thought, there can be no effective defense against a thermonuclear attack, and therefore a program of shelter-building and other preparations for survival involves a waste of money and energy that might better be spent on “waging the peace,” or, alternately, on improving our deterrent and our capability for waging limited war. A second school of thought opposes civil defense for exactly the opposite reason, arguing that our preparations for survival might become so effective that the Soviet Union would regard them as “provocative”; they would thus lead to an acceleration of the arms race or even to a Soviet attack. Nor do some opponents of civil defense rule out the possibility that we ourselves might launch a surprise attack if we were sure that we had effective protection against retaliation.

Though it would seem logically impossible to combine the view that civil defense is ineffective with the idea that it is too effective, some of its opponents reconcile the two positions by assuming the worst: either we or the Soviets will launch an attack out of the belief that civil defense makes a difference, and it will then turn out that the belief was unfounded.

Still another school of thought (mostly, but not completely, composed of certain conservative groups and some military officers) rejects civil defense because it is a form of Maginot-mindedness—that is, because it is defensive rather than offensive: “Brave men do not hide in holes.” Until recently these groups made up the most effective opposition to shelter programs. While, by and large, they have backed down somewhat (probably as a result of the Berlin crisis and through a sort of “right-wing” reaction to the extremism of the “left-wing” criticism of civil defense), they still probably constitute the most important political and bureaucratic opposition to increases in civil defense.



Before the main arguments outlined above are analyzed, one important point should be made clear: that civil defense is indeed likely to be ineffective in contributing to three major objectives. First of all, it cannot reliably hold casualties and property damage to “classical levels”; in fact, it is almost prohibitively difficult to give any great degree of protection to concentrated populations in target areas from a surprise attack directed at them. Second, civil defense cannot, by and large, contribute directly to the conduct of military operations, and third, it cannot contribute very much to deterring a Soviet attack.

If we go back to the First and Second World Wars, we find that civil defense functioned as an integral part of the war effort; the ability to mobilize men and materials after the war started was crucial to victory. Civilians, therefore, represented a second line of defense. They supplied men, materials, and morale to the fighting forces. A thermonuclear war, however, changes the situation entirely. Almost (but not all) strategists today believe that it is impossible—even by heroic achievements in civil defense—to mobilize the civilian society to contribute to a significant military effort after a thermonuclear war has started, if the enemy tries to prevent it. For one thing, a thermonuclear war is unlikely to last longer than thirty days, and it may even last as little as thirty minutes—hardly enough time for the operation of a “post-attack mobilization base.” Thus, the fact that we can give some protection to a factory worker or a machine tool or a mine or even a city would not, on the whole, make the Soviets fear the United States any more than if we could provide no protection at all.

For this reason, cities (with a few possible exceptions) cannot be considered high-priority military targets in a thermonuclear war, as they were in World War II. On the contrary, there is good reason to believe that any first strike would be aimed at the capacity to retaliate, and that cities would not only be secondary targets, but might well be deliberately spared.1 If we were ever to bomb Russian cities, we would be bombing them because we wanted to punish the Soviets; we would be bombing them for reasons of malevolence or reprisal and not because they were military targets. (The Soviet planners, in my opinion, would to some extent agree with this judgment from their side; or if not, it is likely that they will soon come to accept a similar conception.)

But the case for civil defense need not rest either on the degree to which it can contribute to the war effort or on its capacity to strengthen our ability to deter a Soviet attack. It is not the purpose of civilians to protect the military. People are ends, not means. Therefore the main question is: Can civil defense be used to give a worthwhile degree of protection to lives and property and to facilitate recuperation after a war is over? The answer to this question, under many plausible circumstances, is undoubtedly Yes. Taking adequate measures now could save the lives of forty million men, women, and children and reduce enormously the medical and genetic after-effects of a war. To be sure, in most situations forty million of the survivors’ fellow citizens would not have been saved, but this is no reason for needlessly condemning an additional forty million people to death or neglecting those measures which could certainly diminish the other harmful effects of a war. In an even grimmer situation, civil defense could save the lives of as many as a hundred million people, and under some not improbable conditions civil defense could keep casualties down to one or two million. It could also mean that the nation would recuperate in five years instead of twenty, or in twenty rather than a hundred. In this sense, civil defense can be considered at least partially effective against most forms of nuclear attack that we can now anticipate, and extremely effective in certain special cases that could very well arise if our programs for avoiding war should fail in their purpose. And since there is always a chance that these programs may fail, we have an inescapably compelling reason—which is at once moral and political—at least to examine, and very likely to take, the steps that would reduce death, suffering, and destruction as much as possible.



On what grounds do those critics of civil defense who call it ineffective base their judgment? One argument often advanced is that civil defense is ineffective because the long-range physical consequences of a thermonuclear war would make life impossible for the survivors, assuming that there were any. Now, it is true that—according to the best scientific estimates—the postwar environment would be more hostile to human life for some ten thousand years. However, objective studies also indicate that this environment would not be so hostile as to preclude, at least in the long run, decent and useful lives for the survivors and their descendants.2

People in the postwar world would for a time have to get by on a standard of living far below the one we consider necessary in the United States today. We must remember, however, that our standard today is far higher than the mere preservation of life would require.3 Even more to the point, the present standard of public health in America could be dropped without catastrophic consequences. For example, though much of the food supply in the postwar world would be contaminated by Strontium 90, much of it could still be used without necessarily causing death or even serious illness—and even more of it could be used if preparations were made in advance to combat contamination.

Under such conditions, the average individual would run somewhat greater risks of various types of disease and greater risks of having genetically deformed children, but statistically these risks would not be so much larger than those normally run today. Thus—to take the case that most horrifies people when confronted by the idea of a thermonuclear war—the percentage of children born with serious genetic defects might go up, after a war big enough to have exposed the survivors to an average dose of 250 roentgens,4 from the present level of about 4 to an estimated 5 per cent. It can, of course, be said that this is an intolerably large increase; even one more deformed infant is too many. One can hardly disagree with that proposition. Yet the fact remains that life would still go on. People, after all, have lived under far worse conditions than we are accustomed to in this country today, not only throughout most of human history, but even in vast areas of the contemporary world. To argue that an effort to save people’s lives is useless because life would not be worth living in the postwar world is tantamount to saying that people should not be saved because they would have to endure a lower standard of living and a lower standard of health. It is hard to think of any other equally preposterous proposition that serious men are still willing to back.

It is curious, moreover, that those who are most pessimistic in their estimate of the consequences of a nuclear war are generally to be found among the opponents of civil defense. If one believes that a thermonuclear war would be so horrible as to make life for the survivors more difficult than it has ever been in the whole of human history, then it would seem that one is all the more obligated to take measures that might ease the lot of the survivors rather than abandoning them to their fate.



Civil defense, then, is not rendered ineffective by the long-range physical consequences of a thermonuclear war. Nor is it likely to be rendered ineffective by the psychological consequences that some observers have claimed would follow from the traumas of such a war. One must recognize that for most people deep grief is alleviated by time, that people do recover in large part from tragedies, that life does go on. In fact, the detailed map exercises that have been made by students of nuclear war lead to the conclusion that most survivors would not go through as horrible a set of personal experiences as many Russians, Germans, Poles, Yugoslavs, Japanese, and others did in World War II. These people have been left with deep emotional scars, yet few of them now feel they would be better off dead; most are leading “normal and happy” lives.

None of this is to imply that we do not have good sound technical reasons for worrying about the effectiveness of civil defense. Any man today who says that we, as a nation, can survive a thermonuclear war is saying something very complicated. He is saying that we can handle all of the problems—military, social, political, economic, and medical—that would arise from a thermonuclear war; not merely some, but all. Furthermore, he is assuming that a civil defense program would be functioning on the day the war started and that it would not have been rendered obsolete between the time it was designed and the time it was completed. As I have pointed out above, many different kinds of thermonuclear war might conceivably occur. The worst kind—a surprise attack out of the blue directed against population—presents a virtually impossible problem of defense for those in target areas. But our weakness in the worst case (the very case on which opponents of civil defense like to dwell) does not settle the issue. For one thing, a surprise attack out of the blue directed against population is the least likely of all the possibilities. For another thing, programs which are designed to meet less ferocious and more likely wars (e.g., a “straight counterforce” attack which ignores cities and aims only or mainly at strategic capability) can still accomplish something; even in the worst case there is still a difference, after all, between 180 million dead and 90 million dead. One should not lightly condemn 90 million people to an unnecessary death by an undemonstrable supposition that nothing can be done to save them.

As for the problem of obsolescence, civil defense, like any other military system, must be improved and adapted over time. Actually, obsolescence is less serious a problem for civil defense than for many other military systems because even a civil defense system which has been outstripped by technological developments may nevertheless provide a large degree of protection (depending again on the kind of attack that is made).

Unhappily, it may soon be technologically possible to build (probably in less than ten years and at a cost of something in the neighborhood of $10 billion) what I have elsewhere called “doomsday machines”—that is, devices which could actually destroy all unprotected people, or perhaps eventually all people, for only the most elaborate measures could have any hope of protecting against such devices. However, so far as we know, doomsday machines are not now being built by either side, and there are good reasons for believing that they will not be built in the near future. This is not to say that they will never be built; I am not alone in thinking that there is a serious danger of one or more such weapons being built, in a matter of decades, if we do not first arrive at an adequate system of arms control. Of course, the weapons that already exist give us reason enough to want arms control, while the possibility of doomsday machines gives us no reason whatever to neglect civil defense against the far less destructive weapons that are now or soon will become part of the nuclear arsenal, and those which may be developed in the near future.5



Apart from getting a program inaugurated early enough so that if a war should break out we can rely on having a shelter capability in being on the day it starts, there are other problems to be considered in estimating the effectiveness of civil defense and the possibility of the nation’s recovering from a thermonuclear war. Would the social organism fall apart completely—that is, die in some sense—as a result of the tremendous shock it would receive from a large thermonuclear war? Obviously we cannot know for sure. In order to argue that society would reorganize itself after an attack, one must have faith in the ability of people to improvise, to meet emergencies with some intelligence and energy. Faith of this kind is not unreasonable. Insofar as there are historical examples to study (and some of them are close to thermonuclear wars in intensity, e.g., the devastation of Germany and the Soviet Union in World War II)—they provide evidence that people can and do rise to the occasion.

Undoubtedly we are moving here amid vast uncertainties. Many of our suppositions may prove wrong. Given these uncertainties, the advocates of civil defense should not claim too much, and neither should its opponents. It does seem, however, that if there is a reasonable possibility for the survival of society after a nuclear war, a moral obligation is imposed upon us to prepare facilities in advance that would help people to meet the emergencies they would have to face and that would improve their capability for improvising and organizing. That there is such a possibility is indicated by whatever serious studies have been made; these studies are only partial and they are certainly not infallible, but they do give us enough ground for supposing that the survivors of a nuclear war would not meet any objectively insuperable obstacles—especially if they were supported by proper preparations.

What do such preparations consist of? Among the most important are those designed to cope with immediate survival needs and to maintain or restore economic momentum. Plans must be made so that enough food, water, shelter, and clothing will be available immediately after an attack. That means having sufficient stocks on hand to last until production can begin again, and preparing schemes and facilities for distributing these stocks under the many different possible circumstances that could arise. One reason for optimism concerning U.S. recovery is that, as certain analyses indicate, enormous stocks of the highest priority items would be left after the kinds of nuclear attacks which might be launched against this country if a war should break out at some point in the early 1960’s. For example, according to the best estimates, we would not face starvation even if we were unable to get agriculture going again for a year or two.6 To take another example, studies also provide evidence that the national transportation system would continue to work adequately,7 and that therefore the preparations now being made to distribute food before an attack are not strictly necessary. But since, as I have repeatedly tried to emphasize, these studies are not infallible, it is wise to take out as much insurance as possible against the errors they may contain.

The question of political recovery is more difficult to answer. We live today in a very stable country. It is one of the few countries in the world in which the government does not worry about revolution and subversion as major problems. However, these problems might well exist in the postwar world. Even if we won the war, it is conceivable that we might no longer live in a democracy. But again, if adequate preparations were made, our democratic institutions could probably survive most kinds of thermonuclear war. For some very small wars this is almost certainly true; for others, it is a judgment based on the belief that while the lives and thoughts of all the survivors would be affected by the war, their character structure and value system would probably not be changed in any startling fashion.

On this point it has sometimes been argued that the only survivors of a thermonuclear war would be “backwoods reactionaries” who could not be expected to support the rebuilding of democratic institutions. But the argument is faulty, since it not only exaggerates the difference between urban and rural America but also fails to recognize that many or even most of our cities would survive the likeliest forms of nuclear war that we can anticipate. Yet even if it were true that the cities would all be wiped out, it would still be an indefensible moral position that implied—as this one does—that people who live in the country are less worth saving than those who live in cities.



This is by no means as complete a picture as can be drawn, and although the studies that have been made contain many gaps, there is good reason to conclude that civil defense is not completely—or even largely—ineffective. Precisely how effective civil defense can be depends on the kind of war that occurs and the kind of programs that are instituted. The recommendation by the Kennedy administration that we spend $207 million in 1962 for the identification, counting, and equipping of existing shelter space, for the improvement of air-raid warning and fallout detection systems, and for research and development, is both realistic and desirable as a first step. (In fact such a step was recommended in the RAND 1957 civil defense study.) But it is only a first step.

A reasonable program might involve a gradual build-up from about $1 billion annually to somewhere in the neighborhood of under $5 billion annually. An expenditure of that order would buy a valuable degree of protection against most forms of nuclear attack that might occur. The money would be used for the construction of various types of shelter, research and experimentation, educating civil defense cadres, and preparing plans and facilities for post-attack recuperation. But if an ambitious civil defense program should be implemented, it would be important to keep it small enough and to carry it out in a way such that it would be unlikely, by itself, to provoke an accelerated arms race.



This brings us to the second set of arguments often used by opponents of civil defense: not that civil defense is ineffective but that it is potentially too effective. The argument is this: If civilians are considered a target, then an attempt to protect them could touch off a greater effort by the Soviets to acquire the necessary power to destroy them even in their protected state. If we build an adequate shelter system, they may then go to heroic efforts to build larger missiles and a greater number of them. Or equally important, if the Soviets fear that our civil defense preparations increase the chances of our striking them in a crisis or in an emergency, they may then have to keep their forces more alert. This could, in turn, make them more accident-prone or trigger-happy. We would then be faced with what is known as a situation of “false preemption” or “anticipatory retaliation.” That is, they may strike us because they think that we are going to strike them.

Now, I agree that some of these problems might be raised by a large crash civil-defense program—say, one that was initiated at substantially more than $5 billion annually. But I do not believe that the kind of program recommended by the present administration—or even a program as large as the one advocated here—would greatly affect the arms race.8 In order to understand why, we have to consider the psychological aspect of a conflict such as the one we are involved in with the Soviet Union.

There is a widespread feeling that civil defense is a sign of war hysteria and militarism or an admission that war is inevitable, and that it will therefore bring about the very thing we fear: if we build shelters, we will have to use them. In other words, civil defense is an example of the mechanism known as “self-fulfilling prophecy.” In the present context, this term—whose original application was somewhat different from the use that is made of it by opponents of civil defense9—is invoked to refer to situations in which one side acts in a hostile manner because it suspects the other side of hostility; the other side, observing this behavior, responds by acting hostile itself, thus confirming the original estimate made by the first side, which then acts even more hostile and suspicious than before, and so on, until the spiral reaches a very high level of mutual distrust and belligerence.

I think there is no doubt that this process actually does operate both in personal and in international relations. But I also think that when the question is one of civil defense rather than of strategic force (which is a different matter), many of the people who appeal to the “self-fulfilling prophecy” are usually being less rational than superstitious: “thinking makes it so.” They object to civil defense in much the same way that some women will object to an examination for cancer or some men will object to buying life insurance. Yet it is not the examination which brings on the malignancy or the preparation for family security which causes death. On the contrary, it is an indisputable principle of everyday experience that appropriate precautions taken as a result of realistic fears ordinarily help to prevent such fears from being realized. Of course, the operative words here are “appropriate” and “realistic,” and it is on determining what constitutes an appropriate precaution and a realistic fear that we ought to concentrate our minds. Prophecies are not fulfilled magically, but only through communications and reactions that can be identified when they occur. And we may be willing to accept some small costs on the debit side of the ledger if the over-all balance is improved.

In short, the problem in launching a civil defense effort is similar to the problem that exists in all our military programs: to find the line above which we cannot go without provoking the Soviets into greatly accelerated efforts. To those of us who are not ready to accept any large degree of immediate unilateral disarmament, this quantitative question is crucial, since even maintaining our current establishment at its present level involves us in the danger of getting caught in a spiralling arms race. It is true that both sides—restrained by economic and social limitations and the knowledge that unreasonable efforts are likely to touch off similar responses on the part of the antagonist—are running neither as fast nor as hard as they could. Consequently, it is of the utmost (though still not overriding) importance to do nothing that could be interpreted as an excessive loosening of self-restraint.

So far as the over-all strategic balance is concerned, I believe that if we desired we could in the short- and medium-run obtain a meaningful and large increase in our “current superiority” over the Soviets (given their technological and economic limitations). However, it is unlikely that this could be done without making the arms race much more dangerous than it already is. Therefore, unless the Soviets force international relations to deteriorate drastically, we should be careful to avoid pursuing an undue degree of increase in the superiority of our over-all strategic force. So far as civil defense is concerned, I would—as I have already indicated—be opposed to current programs at the level of $5 or $10 billion a year, despite the fact that in my opinion such larger programs could be justified by an analysis based only on narrow military and economic considerations. The civil defense line probably should be drawn somewhere below $5 billion annually, and the U.S. should build up to that figure only gradually, so as not to involve itself in too abrupt a change of policy.



The third argument generally advanced by opponents of civil defense is that an American civil defense effort might lead to a preventive war by the Soviet Union (because such a program would convince the Soviets of our aggressive intentions). Alternately, it is asserted that a civil defense program might induce a preventive war by the United States (because by attacking at a propitious moment, we could hold casualties to a few million). Both these arguments seem almost completely incredible. They rest on the notion that unless one can guarantee total annihilation, the other side will not be deterred; that is to say, unless we can promise the Soviets that every single American will be killed in their reprisal, they will worry about a surprise attack against them out of the blue. Yet neither side has shown so great a desire to strike the other as this notion implies. In my opinion, if there were perfect protection for every citizen of both countries, and an invulnerable post-attack recuperation capability to boot, deterrence would still operate under most circumstances; after all, the empty cities—which have such immense historical and cultural significance—would still be hostages. In addition there would be uncertainties in the analysis : who would trust a hypothetical analysis enough to make him press a button in cold blood? But in the more practical case we can ignore these subtleties; no country is going to go to war lightly simply because by doing so it could reduce fatalities from 60 million in a possible future war to, say, 20 million. Twenty million is a very impressive number of people to sacrifice to an estimate of the future course of international politics.

There are, however, circumstances in which certain kinds of civil defense programs might tend to convert an especially tense crisis into a war. Harsh choices can arise. We may have to decide between the risk of immediate war and appeasement or surrender, with whatever that entails in terms of future risks. In such a situation some kinds of civil defense, in particular evacuation programs, could affect one side’s decision or the other’s, and thus would increase the risk of immediate war. But we must recognize both that the risk may have to be taken and that having a credible ability to accept that risk may deter the Soviets from deliberately creating the very situation in which it would arise. I would, in fact, conjecture that an appropriate civil defense program might even contribute to a relatively large reduction in the probability of war—by reducing the frequency and intensity of Soviet-inspired crises.

Still another argument that has been made against civil defense is that although it is ineffective, it can fool our leaders and make them more reckless, or it can fool the people who will then become more reckless themselves or allow the government to be. Since I have already set forth my reasons for believing in the effectiveness of civil defense, and since I have already indicated my objections to the notion that a danger of recklessness is involved, there is no need to discuss this argument further. However, it may be worth speculating on why people hold to such positions. The reasons, I think, range from a simple visceral desire to dismiss the whole subject of nuclear war; to a fanatic desire to concentrate all our energies—material and intellectual—on a single “approved” approach (whether this be deterrence, accommodation, or unilateral disarmament) with no insurance for contingencies; to an even more fanatic desire to construct an oversimplified “everybody red or everybody dead” argument.

In short, the critics of civil defense do not want to think about the possibility of a nuclear war actually being fought. They prefer wishing it away, ignoring it, denying its existence as a problem worthy of the most serious thought and consideration. Even professional strategists sometimes do not want to concern themselves with the details of the balance of terror—the obvious possibilities for miscalculation, unauthorized behavior, accident, or even war by calculation. They do not want to consider these possibilities seriously, in the sense of letting them affect programs. Most such strategists fear that civil defense competes financially with improved deterrent or limited war forces. Many also fear that it will weaken the morale of the civilian population by casting doubt on the capacity of SAC to deter or by making the risks of war seem more urgent and actual. Hence these strategists take refuge in the automatic balance of terror—the idea that there can be only one kind of thermonuclear war, a war inevitably involving mutual annihilation, and therefore a war that can never take place. Yet everything depends on how the war may start, how it may be fought, and how it may be terminated. And unless one understands that there is a whole range of possible situations, one cannot fully appreciate the potential effectiveness of the different kinds of civil defense programs which might be recommended.



It is not only our first-priority objectives that are essential; there is a long list of things that we cannot safely do without. We need to eat and sleep as well as to breathe, and though human biology permits of no question about which comes first, neither can be ignored indefinitely. Similarly, there is no question that it is far more important to avoid war than to find ways of reducing its damage and of recovering from its effects. But since we cannot be certain that we will succeed in preventing war, it is essential that we take moderate and prudent steps to minimize the disaster that such a failure would mean. In the event of a war, civil defense could not only save millions of lives but could also prove crucial to the continued survival in the world of Western ideals and institutions.



* * *

Erich Fromm & Michael Maccoby

Up until 1961 few Americans took the possibility of nuclear war very seriously. Many were convinced that such a war would never occur because it would be too destructive; others did not think about it at all. The change which took place in 1961 was brought about in the first instance by the acute Berlin crisis and President Kennedy’s speech of July 25. The President told the nation after his Vienna meeting with Khrushchev that the Soviet leader had threatened to execute his Berlin plans within six months at the most, and since we were resolved not to accept such an ultimatum, the inference was that a thermonuclear war was a definite possibility before the end of the year. The President added in this context that he would announce the steps a citizen could “take without delay to protect his family in the case of an attack,” and proposed a new $207 million shelter program.

The President’s speech might not have been so effective had the soil not been prepared by the most vocal and most influential spokesman for civil defense, Herman Kahn, and a number of his colleagues, especially from the RAND Corporation. Mr. Kahn’s basic approach can be described as being close to psychoanalysis. Not so much because he makes many statements about psychological matters (such as the quality and duration of grief, the discipline of people during and after a nuclear attack, the capacity to be happy in the post-attack world, etc.), but primarily because his central thesis is that, out of fear, people repress the awareness that thermonuclear war may come, and consequently they ignore the possibilities of defense. Hence in order to enable them to prepare adequately, the thing to do is to make them aware that a thermonuclear war can occur. No psychoanalyst could quarrel with this principle. The question is only whether an awareness that thermonuclear war is actually possible leads to an increased sense of realism or whether, as we believe, old illusions are replaced by even more dangerous ones. Kahn, whose good will and great ability we do not doubt, has on the whole given the opinion leaders and the political leaders of this country the impression that nuclear war need by no means be catastrophic—that, provided we take the proper steps, the country can recover, and that after ten or twenty years people can once again lead happy and prosperous lives. He has, indeed, qualified this general thesis by many “ifs,” but it is the general thesis that has taken hold, while the qualifications have become the fine print which is forgotten.

There are two conceptions of the role of civil defense and, specifically, of a shelter program. The first—stressed in Kahn’s testimony last August before the Holifield Committee10 and also in President Kennedy’s speech of May 25—sees the shelter program as “life insurance,” and argues for it on the ground that it would save many millions of lives. At present the assumption is made that fallout shelters might save not only lives in rural areas, but also in the cities—since it is calculated that the Soviets are not likely to attack our cities. As we shall try to show later, this calculation is quite unwarranted, and the probability is that our urban population would be wiped out in a thermonuclear war. However, since millions of people living in rural areas away from population centers and military installations might be saved by fallout shelters, and since nuclear war is possible or even likely to occur, who would dare to dissuade a family or a community from constructing fallout shelters? We certainly would not.

Quite different from the “insurance” idea is the conception of civil defense as part of our national strategy. Many spokesmen for civil defense contend that it will greatly improve our strategic position, that it may help to avert war. We shall try to show: (1) that there are severe limitations to the effectiveness of civil defense; (2) that it is more likely to provoke war than to deter it; (3) that even if it were optimally successful in war, it would not prevent the replacement of our democratic system by a totalitarian one. If we are right on these points, then even the justification of civil defense as “insurance” may have to be reconsidered. The true situation may be analogous to one in which a man takes out life insurance under conditions that considerably increase the likelihood of his death. In such a case, one might still not try to dissuade him if he wished to buy insurance, but certainly he and his friends would be right to have severe doubts about its usefulness.



The first and most fundamental limitation of any shelter program is that it could not save our urban population in the event of an attack against our cities. Rather than frankly accepting this fact, Kahn and other advocates of civil defense argue that if the Russians strike first they will try to destroy our missile and SAC bases, not our cities. Given such a strike, the greatest danger to those city dwellers fortunate enough to be living well away from strategic centers would be from delayed fallout, and thus fallout shelters might lower the immediate death toll from fifty to ten million people.11

On what grounds do strategists suppose that the Russians would not direct a first strike against our population centers? Kahn, in his book On Thermonuclear War, gives several reasons: (1) they have so few missiles that they would have to choose between attacking our military installations or our cities, and therefore in order to neutralize our striking power they would choose to hit our military installations; (2) they would not wish to attack our cities first, in order to hold them as hostages against our retaliatory strike; (3) “almost nobody wants to go down in history as the first man to kill 100,000,000 people.”

The weakness of Kahn’s first point lies in its shortsightedness. Since it will take no less than a year for even a minimal civil defense program to bear fruit, calculations ought to be made not on the basis of present Russian missile strength, but on the number of missiles the Soviets may have a year or two from now. In addition, even if they did decide to make our military installations their main target, the Soviets might kill an additional 50 million people by diverting only a fraction of their missile strength to our ten largest population centers. (This is the kind of attack Kahn calls “counterforce plus bonus.”) Further, we have so increased our strength in the past year that a successful attack against our bases would no longer cripple our retaliatory power. The Russians would thus be foolish to limit their attack to our military installations, knowing that their own cities would remain as hostages.

As to the second point—that they would avoid hitting our cities in order to hold them as hostages—Kahn imagines that after having destroyed our military bases, they would command us not to strike their cities on pain of having our own cities destroyed in retaliation. But is it realistic to assume that after the destruction of our military installations and the death of ten to fifteen million people, the Russians would expect our government to wait and listen to their demands rather than to use all its strength immediately in a mood of fury and revenge against the Russian cities?

And finally, as for the reluctance of a political leader to go down in history as the first man to kill 100,000,000 people, such considerations did not interfere with the decision to saturate cities with bombs during the Second World War; nor did they restrain the dropping of atomic bombs on Hiroshima and Nagasaki. Why then should we expect the force of moral considerations to be greater now when the survival of entire nations is at stake?

All this, it should be added, applies only to a first strike by the Russians. If we were to initiate a first strike—in retaliation for, say, a Soviet invasion of West Berlin—the Russians would not hit back at our empty bases; clearly, they would attack our cities.



Assuming, then, that our cities were attacked, what chance of survival would the urban population have? Almost none, given the effects of the megaton weapons. A 20-megaton groundburst leaves a crater 300 feet deep and a half mile in radius, destroying all underground shelters. Within a four-mile radius, the most heavily reinforced concrete structures are leveled. At eleven miles, the blast pressure destroys all conventional frame or brick buildings, and buries most basement fallout shelters, while winds of 160 miles per hour turn hurtling debris and human bodies into deadly missiles.

But blast is the least of the killers in thermonuclear war; fire, instant radiation, and delayed fallout would claim many more lives. Within at least a 25-mile radius of a 20-megaton blast—an area of about 2,000 square miles—any exposed person would die of burns, and raging fires would soon begin to consume the air in fallout and blast shelters.12 As long as there was fuel for these fires, they would burn on, unchecked. Even more widespread in its potential effect is the danger of blindness to those above ground at the moment of blast; anyone who glimpsed the explosion would be blinded, for it would emit a light pulse at less than 0.015 second (the time needed for a saving eye-blink).

While in the city itself almost nobody could survive, people living in the suburbs—even if they escaped these instantaneous effects of blast, flying debris, heat, blindness, and radiation—would still be threatened by delayed fallout. At 25 miles from a 20-megaton explosion one might expect doses of 3,000 roentgens per hour arriving after about twenty minutes—enough to kill a person within two minutes, unless he were in a specially constructed fallout shelter. If he could succeed in reaching a shelter with a large shielding factor within twenty-two minutes after the blast, and if he could avoid glimpsing the fireball, a hard fight for survival would then have only begun.

What about saving the urban population by evacuation? While many reject the idea of evacuation, others (including Kahn) consider it a serious possibility. It is hard to see why. If we were to evacuate our cities for every political crisis, we would probably have to leave them empty for several months practically every year. Even if this were feasible (which it is not), after one or two such evacuations no one would leave, for the warning signal would come to be considered a repeated cry of “wolf.” Further, every time we evacuated, we would give the Russians a reason for supposing that we were planning a first strike, and hence the chances of a preemptive attack would increase. As to evacuating cities after a warning, fifteen minutes would obviously not be enough time if the Russians struck first. If, on the other hand, we wanted to strike first, it would require many hours at the very least to evacuate all our major cities, and since such a move would be impossible to conceal from the Russians, they would obviously not wait for our attack, but would hasten to hit us first. At best we might secretly evacuate our leaders before striking, but how many of them would be willing to leave for safety knowing that their wives and children would soon be killed?

Finally, let us consider the possibility of protecting our urban population by vast underground shelters (which have also been proposed by certain enthusiasts of civil defense). In the case of a surprise attack it would take more than fifteen minutes to get people down to the streets from big offices and apartment buildings. The panic, and the struggle for elevators, doors, and the like could only result in the same kind of situation that arises in a theater when fire breaks out. Even if there were a shelter entrance not further than five minutes’ brisk walk from any point in the city, it takes little imagination to visualize how many people would be trampled to death before reaching the shelter, and how few—even of the strongest and most brutal—would be saved.



If everyone in the city and its suburbs is likely to be killed, what are the chances for survival beyond the 25-mile range? Here fallout would be the greatest menace. In some areas—say 50 miles from the blast—it would probably be necessary to remain in a fallout shelter for at least two weeks, and afterward it would be possible to leave it for only a few hours each week. How many people would find themselves in this situation depends on the magnitude of the attack, and on the amount of fallout generated by megaton bombs—something the experts have still not agreed upon. A problem in making any calculation is the lack of studies which outline exact fallout danger at different distances from the blast, or in accordance with various possible attacks. Neither are there sufficient data on the effects of other radioactive particles—for instance Strontium 90, which might make farming impossible anywhere in the country. There are experts who think that many farmlands might require forty replantings before becoming safe again, and no one has disproved this estimate. Norman A. Hanunian of the RAND Corporation, who prepared the statistics on heavy attacks used by Kahn in his book, concludes his testimony to the Holifield Committee by stating that “the outcomes of future attacks are anything but precisely predictable. Fallout could create overwhelming disaster, but it might not. Whether it would depends to some extent on factors we have not examined today—on wind, for example. But it depends most importantly on the kind of war that the potential combatants may be prepared to fight.” What kind of war can we expect? Ralph E. Lapp has written that present Soviet stockpiles are more than adequate for a 10,000 or 20,000 megaton attack, enough to saturate the whole nation with fallout.

Hanunian considers attacks of from 300 to 30,000 megatons, but the latter, as we have just seen, seems closer to Russian capability. In such an attack, with five-sixths of the bombs directed against military targets, Hanunian estimates that even a total fallout shelter system would be unable to prevent from 54 to 85 million people from being killed. What the experts have so far not studied are the long-range effects of 30,000 megaton bombings, which might so contaminate the countryside as to leave crops inedible, unstored water undrinkable, and food-giving animals dead from radioactivity. If a 30,000 megaton attack were to take place—as indeed is possible—what would become of Kahn’s optimistic idea of the “B country” (the rural areas and the small towns) rebuilding the “A country” (the 53 major metropolitan areas)?

In summing up our discussion of the limitations of civil defense, let us take a brief look at the possible types of shelter program and try to determine how effective each is likely to be in the event of the kind of war we consider most probable—a war involving attacks on our cities, either directly or as “bonus.”

The effectiveness of the current minimal program of marking and stocking shelter space mainly in urban areas would be most limited—except in the unlikely case that the enemy were to decide against attacking even the ten largest cities.

The next possible program—proposed by Walmar Strope of the Naval Radiological Defense Laboratories—is the $10 to $30 billion network of well-stocked communal fallout shelters built to house everyone in both rural and urban areas. In our opinion, these shelters could not protect the population of the cities. Perhaps in suburban areas 20 miles or more away from the explosions, some people could be saved (provided, of course, that they were well-organized, disciplined, and able to weather severe stress and disease). However, it does seem to be true that, depending on the type of attack, the time of day, and many other unpredictable circumstances, this program could save a large proportion of the rural population.

A final possibility is the science-fiction program that Kahn has suggested, which would cost $200 billion. This envisages underground space for factories and urban blast protection that could—as Kahn puts it in On Thermonuclear War—“probably take direct hits of ‘small’ bombs (say less than 5 MT) and [might] even take ‘near’ misses of ‘large’ bombs.” Yet even after spending so much we would still have no security: since it is infinitely cheaper to increase striking power than it is to raise the level of protection, even the most hardened shelter cannot guarantee safety. If there were a possibility of adequate shelter, it would be found only in underground cities where we would have to live permanently. Is this troglodytic life the fulfillment of the American Dream?



Assuming, with all the qualifications introduced above, that a complete and thorough shelter program could save millions of people in rural areas, what would post-attack life be like for the survivors—psychologically, morally, economically, and politically? The situation in the shelters, of course, would vary with the different types of shelter and the amount of fallout in the particular locale. Privately owned luxury shelters would be comfortable, provided they were defended successfully and would-be intruders did not retaliate by blocking the air vents. In the public shelters, the danger of over-crowding would exist, especially if the program were not completed before the attack came. Moreover, we should expect that in many communal shelters disease and meager rations would exact a psychological toll.13 People might have radios, which would boost their morale (provided transmitting stations were not destroyed), but this might also be offset by hearing that all big cities had been leveled, and that there was widespread disease. Further, unless the attack were to have occurred at night, many families would have been separated, with people not knowing whether their wives, husbands, or children had found protection. Given all these conditions, what would be the state of mind of those in the shelters, immediately after the attack and for some time to come?

Kahn’s answer to this question is very optimistic. “It is my personal belief,” he told the Holifield Committee, “speaking less as an expert than as a man who has read widely, that these problems [social, psychological, political, and moral questions] have been grossly exaggerated. Most people will not be psychologically deranged. One is not, for example, going to break up family relations by a war. The family relation is a very stable one. . . . One is not even going to obliterate the fact that people are Americans. By and large, they will be about as honest, hardworking, reliable, and responsible as they are today. While everybody’s life and thoughts will be affected by the war the character structure of the survivors is unlikely to be changed in any startling fashion.

In his testimony Kahn also speaks about “post-attack grief” and argues against the notion many people have that “because of the enormous number of casualties, all of the pleasure, all of the taste will permanently go out of life for almost everybody.” “As far as I know,” he says, “that just hasn’t happened in anything that has occurred before, and one would not expect it to happen even as a result of a large thermonuclear war.” One reason for believing that it would not happen is that “in a sense, grief is family-sized. If one loses a close relative or close friend, one will grieve. If one loses one’s family, one will grieve even more. But, in some sense, that is about as far as one can go. Most people would not mourn for a million people much more than they would mourn for their family.” Kahn also claims (in On Thermonuclear War) that a shock spaced over a few days “is good, not bad” so far as its psychological effects are concerned, because “the habits of a lifetime cannot be changed for most people in a few days.” And here, finally, is how he summarizes his conclusions as to the psychological effect of a thermonuclear war in his book: “Despite a widespread belief to the contrary, objective studies indicate that even though the amount of human tragedy would be greatly increased in the postwar world, the increase would not preclude normal and happy lives for the majority of the survivors and their descendants.”

Bearing in mind that Kahn himself states explicitly that he does not speak as an expert on such matters, and that he also says that the disaster studies which have been made are not sufficient to establish his case on solid scientific grounds, let us now consider the picture he paints of the post-attack psychological situation.



To begin with, the problem of post-attack psychic shock is not one of families breaking up. The problem is how those families that would not be broken up would react to the break-up of the whole world around them. Kahn believes that the character structure of the survivors would probably not be “changed in any startling fashion,” but he fails to say why he believes so, and his patriotic appeal to the fact that “people are Americans” is not an adequate substitute for good reasoning. Indeed, one might more plausibly assume that non-Americans accustomed to totalitarian discipline would, if anything, be less radically affected than the average American. Moreover, the notion that sudden shock is less far-reaching in its effects than prolonged suffering is totally indefensible. Vast psychiatric experience and a huge body of literature are there to show that traumatic neuroses are produced both in peace and war by sudden fright and by tension of an intensity which transcends the amount our nervous system can tolerate. Such neuroses can result in severe depression, suicidal tendencies, self-accusations, amnesia and disorientation, and states of anxiety—all of which may persist for many years. To be sure, long-lasting states of despair can also produce severe psychic damage, but to ignore (as Kahn does) the effect of sudden shocks of great intensity is only to make the picture rosier than it really is.

The experts testifying last August at the Holifield Committee hearings on civil defense tell us that no disaster study yet made reports the psychological consequences of devastation as wide and as great as would result from a thermonuclear war. We, however, would like to recommend to Kahn and his colleagues that they look into one disaster which has been studied, and which is comparable to a thermonuclear war in terms of loss of life and disruption of society: the Black Death of 1348-1349. As the distinguished historian William L. Langer writes, the Black Death was “the greatest single disaster that has ever befallen European mankind. In most localities a third or even half of the population was lost within the space of a few months. . . .”14 Particularly relevant in the context of the present discussion is the fact that the cities were the hardest hit by the Black Death. Professor Langer notes that with the Black Death the phenomenal economic progress of the 13th century came to a halt, followed by a prolonged depression, but he also feels that in some sense the economic effects were secondary to the long-range psychological consequences of those mass deaths. He writes that “the horror and confusion in many places brought general demoralization and social breakdown. Criminal elements were quick to take over, looting the deserted houses and even murdering the sick to rob them of their jewels.” The period after the crisis was marked “by a mood of misery, depression, and anxiety, and by a general sense of impending doom,” so much so that it has been suggested that people hesitated to marry and raise a family. (According to reports, this has also been true of the survivors of Hiroshima.) Langer’s summation seems to us remarkably applicable to the most likely outcome of a thermonuclear war:

It is perfectly clear that disaster and death threatening an entire community will bring on a mass emotional disturbance, based on a feeling of helpless exposure, disorientation, and common guilt. Furthermore, it seems altogether plausible to suppose that children, having experienced the terror of their parents and the panic of the community will react to succeeding crises in a similar but even more intense manner. In other words, the anxiety and fear are transmitted from one generation to another, constantly aggravated.

Which brings us to Kahn’s idea that grief is “family-sized.” The trouble with this idea is that grief is not the only problem when a thermonuclear war is in question. We must remember that the survivors would witness a sudden tearing apart of the whole fabric of society. For most people, the sense of stability, and even of their own identity, rests on the meaning society gives to their lives. What then might we expect would happen to men if everything that seemed to be certain became completely unstable within a matter of hours? Previous wars supply no precedent for such a situation. The soldier was of course exposed to great stresses, but life remained stable for him precisely because he knew that his family and the rest of society were still relatively unchanged. In thermonuclear war, however, no part of the social fabric would remain stable. Half of the population killed; most of the leaders gone; no transportation; unburied corpses; epidemics; no communications, electricity, or water supply; divided families; many months necessary to create the minimum conditions for renewing a semblance of life as it was previously known—and for what? What sense would life make? What hope would there be? How much fury would be generated in those who had fared worse than others? How many would blame themselves for being alive when others were dead? What would be the reaction of those who were just managing to get by when refugees turned up to be fed? How much rage would there be against the leaders or scapegoats who would be held responsible for having brought the war on? No, for the majority of people the problem would not only be grief, but the destruction of a way of life which had given meaning to their efforts, which had produced a sense of identity, as well as a sense of hope for the future.



If these would be the psychological effects of a thermonuclear war, what shall we say about the moral consequences of such a war? Keeping in mind the fact that morality, like psychology, is in large part socially conditioned—that individual morality existing without support from the community is rare—let us try to imagine what the postwar moral atmosphere would look like. Let us imagine a situation in which millions of innocent people have been horribly killed; in which we may have defended our own safety by letting our neighbors die; in which we may have to fight for our minimum standards of living against thousands who come into our area to be fed and sheltered; in which we are envious of those who protected themselves better than we did; in which we are frightened and resentful of those who made thermonuclear war seem palatable and possible. What sort of ethics would develop in such a situation—something similar to a belief in God, in brotherly love, and in freedom, or the ethics of the jungle and the concentration camp? The question very nearly answers itself. Is it not indicative that even now people speak of the duty to defend their shelters with guns against those who have been less “provident” (or affluent), and that at least one “man of God” has said that such actions do not contradict Christian teaching? Yet not only do they contradict Christian teaching, they even contradict the ethics of military behavior which command the individual soldier to risk his own life in order to save his fellow-soldiers. In the light of all this, it seems quite obvious—and even the experts sometimes vaguely hint at it—that post-attack life would be possible only under a military dictatorship which used force to uphold even a minimum of social responsibility. Not morality but martial law would be the basis for whatever vestige of civilized behavior might survive a nuclear war.

In talking about recuperation, Kahn occasionally draws on the experience of the last two wars. Millions of people were killed, he points out, and billions were destroyed in property value, yet only a few years later things had more or less returned to normal, and most of the survivors were again leading “happy” and moral lives. This is simply not the case. The history of man since the First World War, though still to be written, would show an increasing brutalization; it would demonstrate that brutalization, approved by society, leads to further brutalization. The slaughter of the First World War was senseless; in contrast to the belief that this was the war to end all wars and to establish democracy, it was in fact fought for territorial aggrandizement and the ambitions of the contending political leaders. In that war, for the first time in modern history, a recognized moral principle—that unarmed civilians must not be attacked—was violated by the aerial bombardment of cities on both sides. Then came the state-approved massacres of Stalin and of Hitler, which were allowed to take place with astonishingly little moral protest, except for the kind motivated by political considerations. Finally—let us at any rate hope that it is the final development—came the indiscriminate slaughter of civilians in the Second World War, first by the Nazis, then by the Allies in the mass bombing of German and Japanese cities. What we see here—to use a favorite term of the atomic strategists—is the “escalation” of brutality from 1914 to 1945; if it were not for this escalation, these same strategists would not be able to write about forty or sixty million dead being “acceptable,” nor would anyone be able to take such reasoning as “normal.” The very fact that a balance sheet of death can today be calmly drawn up is the result of the brutalizing influence of two world wars and the systems of terror that have operated in our time. Many experts are unaware of the degree to which this brutalization is contained in the very discussion of the “acceptability” of killing fifty million people on each side, and they are equally unaware of the further brutalization which a thermonuclear war would produce. Moral development, indeed, is always the moral development of a society, and when a society commands mass murder and mass suicide, only very few will be able to hold fast to Judeo-Christian or humanist ethics.



Let us next consider how a thermonuclear war would affect the economic situation of the country. Kahn’s estimates here, as in the psychological realm, are quite cheering. It is his thesis in On Thermonuclear War that “if proper preparations have been made, it would be possible for us or the Soviets to cope with all the effects of a thermonuclear war, in the sense of saving most people and restoring something close to the prewar standard of living in a relatively short time. But there is no reason to believe this will be true unless both nations investigate the problem more thoroughly than has been done so far, and then take the necessary preparations.” S. G. Winter, an economist of the RAND Corporation, in his testimony before the August hearings of the Holifield Committee, is equally optimistic. If his assumptions are valid, he says, “it turns out that capacity is back to the 470 million level in just over a decade.”

But what are the premises on which these estimates rest? In Kahn’s case, the premise (as expressed in his book) is that we succeed in holding damage down to the equivalent of something like 53 metropolitan areas destroyed, and that “seven optimistic assumptions” materialize: “1—Favorable political environment. 2—Immediate survival and patch-up. 3—Maintenance of economic momentum. 4—Specific bottlenecks alleviated. 5—‘Bourgeois’ virtues survive. 6—Workable postwar standards adopted. 7—Neglected effects unimportant.” Why these optimal conditions should all be fulfilled, neither Kahn nor Winter makes clear.

Winter arrives at his hopeful diagnosis by thus qualifying all his conclusions: “The issues are too complex to be fully understood, and consequently there is no possibility of providing answers that are beyond reasonable challenge.” “No amount of research,” he continues, “is likely to alter the fact that decisions will finally have to be based on a large measure of faith in, or skepticism about, the basic strength or resilience of the people and institutions of our Nation.” However, he goes on to admit in discussing the research concerning economic recovery: “A good deal of competent and important work has been done, but it does not really scratch the surface of this vast problem and there is in particular, a definite need for a systematic and comprehensive re-examination of the whole problem.” [Our italics.]

Such, then, is the scientific basis for the bright outlook of Kahn and Winter.

The picture grows even darker when we study the conditions Winter specifies for avoiding “complete failure in the recovery effort.” Such failure would occur if “the effectiveness of the federal government and many state governments is greatly diminished, the banking system disrupted, most surviving firms are bankrupt, electric power and water supply systems are severely damaged, and the transportation network broken in many places, and where few survivors have the responsibility, authority, and plans to do anything about it.” Is it not likely that this is precisely what would occur, even with a vast civil defense program? Another condition for recovery is a release from the necessity of spending money on rearmament; that is, the war must “produce a substantial or fairly permanent reduction in the external threat.” How could this be expected to happen? If we were to have destroyed Russian military power, China and other nations probably would still have escaped the worst of the war, and might by then have acquired nuclear weapons. Or are we to assume that after a war we would get universal disarmament, which now appears impossible?

There is yet a further condition which Winter believes important for a rapid recovery and the avoiding of bottlenecks: the willingness and ability of foreign nations to trade with us, and even to provide assistance for our reconstruction. This assumes, apparently, that the European countries and Japan would not have been involved in the war, for what other nations would trade with us or give us assistance? The Soviet Union? China? Australia and New Zealand? Or Latin America, which needs our assistance now? Eventually, Winter adds another condition to his prognosis for recovery: that all analysis of the economic problems hinges on relatively optimistic answers to the psychological problems that would arise in the post-attack situation. If this is the case, then it seems to us that the whole prognosis rests on a house of cards, many of which are themselves shaky, being made up as they are of questionable or improbable premises.

Apart from all this, both Kahn and Winter take insufficient account of the interaction of the various disasters that would result from a thermonuclear attack. To mention only a few such interactions: people would need instruction by radio, but most transmitters might be destroyed; people would need hospitals, but most would probably be demolished; certain injuries would have to be treated by extensive washing, but not enough water would be available; the dead would have to be buried in heavy fallout areas, but no one would be able to leave the shelter without getting lethal doses of radiation himself, while the bulldozers which, according to Kahn, might be necessary for mass burials would not be available. Our whole mechanized agriculture depends on gasoline, but the refineries situated near population centers (which make up two-thirds of all refineries in this country) might be destroyed; how could the remaining one-third be made sufficient to supply all agricultural and other needs? Winter, in answering questions after his testimony, was forced to admit that his optimistic calculations are based on the premise of an uninterrupted national transportation system. What value has an analysis of economic recovery that depends on such unrealistic assumptions?



If we bring some of the factors neglected by Kahn and Winter into the picture, what would the economic situation after a thermonuclear war look like? Let us assume that all urban centers, more than half (at the least) of the “survival industry,” and two-thirds of the heavy industries are bankrupt. Many of their stockholders, who had lived in cities, are dead. Their stocks and bonds have been burned. Under such circumstances how many would be able to prove property rights? What about money in banks whose books had been destroyed by fire? Winter tells us that “a number of financial institutions are microfilming their vital records . . . but this is not by any means universal.” Others, according to the same testimony, would be able to microfilm and store records in a safe place “if they had a day or two of warning.” But what about the records of small firms and of individual wills, all of which would be destroyed in the population centers? We must expect that a disappearance of paper from our highly complex economy would wreak havoc with private property.

But even if it turned out to be possible to find records of ownership, another and much more serious threat to the free enterprise system would arise. Some parts of the country would be less devastated than others and would have to help those in worse condition. This could be done only if the state took over the economy and divided goods according to need. Even accepting Winter’s optimistic estimate of recovery on the basis of a 25 per cent rate of growth in order to build up destroyed industries, the state would have to control capital investment, and manufacturing would have to be centrally directed in order to secure the production of the most vital necessities. Even if half the population and half the industrial plant were not destroyed, much of our capitalist economy would have to be replaced by a state-directed, centrally managed industrial system. Whether this system would be managed by a small group of industrialists or by the state, and whether individuals would to some extent retain nominal ownership while the state took over ownership of a large chunk of the national wealth, of one thing we can be sure: even the most favorable possible outcome of a thermonuclear war would lead to a centrally controlled managerialism. We have to consider, in addition, that the severe sacrifices and discipline necessary to recovery would make it imperative to introduce a system of total control not only of production, but of the population. To be sure, such a system might be imposed in the name of freedom rather than by martial law, but it would be totalitarianism all the same. The fact is, then, that even a “successful” thermonuclear war would leave the survivors with a political and economic system not too different from the one we are supposed to be fighting. The alternative, in other words, would not be “better dead than red” but “better red than ‘red.’”



If no civil defense program can save our cities or the fabric of our society from the ravages of a thermonuclear war, can an extensive system of shelters nevertheless serve as a deterrent to the enemy? There are two very different forms of deterrence: deterrence against attack, and deterrence against political provocation (Deterrence I and II in Kahn’s terminology). No one claims that shelters would deter an enemy from attacking us first, and President Kennedy has stated explicitly that civil defense “cannot deter a nuclear attack”; according to most experts, the only attack deterrence is a powerful second-strike (or retaliatory) capability. But what about political deterrence, the attempt to restrain the enemy from political provocations by threatening to strike him first?

According to Kahn and others, civil defense makes it more credible to the enemy that we might strike first; if we cannot protect ourselves against a retaliatory strike, the enemy might call our bluff, discounting the possibility that we would risk our population merely to defend, for example, access routes to Berlin.

Though—as we have seen—civil defense would not save our cities and would even in the “best” case leave us with 50 to 70 million dead, it can be admitted that a shelter program does increase our first-strike credibility and thus improves our political deterrence. However, the enemy must still become convinced of our willingness to make such a sacrifice, and we are therefore forced to gamble on whether he would believe that a particular political aim was important enough for us to accept destruction of these dimensions. (As Kahn himself points out, most of our leaders would not start a war if they expected to lose more than 60 million American lives.)

But in addition to this, civil defense tends to provoke war precisely because it improves our political deterrence. The more credible we make our resolve to strike first, the more the Russians will expect us to attack during a crisis, and hence the more they will be likely to launch a preemptive strike. (Kahn himself writes in his book: “The one circumstance under which almost all Soviet experts agree the Russians might strike is the one in which they anticipate a strike from us.”) Our own fear of Russian preemption will in turn make us more prone to strike first, and so on up the spiral. Thus, to the extent that our first-strike capacity becomes more credible, Russian preemption becomes more likely; in balance-sheet language, what we gain in political deterrence we pay for in an increased probability of war.

Aside from the war-provoking aspect of first-strike credibility, a large civil defense program tends to indicate that we are getting ready for war, and this might start a vicious circle of preparations, counter weapons, and counter preparations—which would have the combined effect of hastening the onset of war. Even if we could build a system of civil defense so perfect that it would reduce fatalities to the 3-5 million range—a prospect that at the moment seems no more than a dream—is it likely that the Russians would sit back while we were making ourselves invulnerable enough to force them into any concessions we chose? Notwithstanding our second-strike capability, they might attack before we had gotten very far in building our fortress. For, as Kahn points out, they tend to view strategy more in terms of chess than poker—and in chess one exchanges queens in order to maintain a tactical position that will otherwise deteriorate.15



Another way in which even a national fallout shelter program could increase the chances of war is by lulling the nation into a false sense of security. We are even now being led to believe the claims of Life magazine or Dr. Edward Teller, that with some fallout protection most people could survive a thermonuclear war, while the probable fate of our cities is hardly publicized. To support this illusion of safety, all the tricks of modern advertising are being drawn upon: gay pictures of teenagers chatting in shelters, survival statistics based on minimal attacks against military installations, claims of overwhelming military superiority on our side, and even appeals to individuality and the spirit of the old frontier, as though winning a thermonuclear war were a matter of showing manly courage. Thus Kahn says that “We are in a position much like the pioneer. He had to carry a gun because the Indians might attack him.” (This analogy makes sense only if one substitutes “neighbors” for “Indians.”) Under the spell of this false sense of security the American people may become more willing to support an adventurist military policy rather than more fervent in demanding disarmament negotiations, just as our leaders may become less hesitant about pushing their terrible buttons.

The belief that thermonuclear war need not be catastrophic increases the possibilities of thermonuclear war. As Walter Millis wrote in reviewing On Thermonuclear War, “Unless thermonuclear war can be re-established in the official mind as something which it is possible both to fight and to survive, it is unlikely that there will be a thermonuclear war.” We share with Millis the opinion that thermonuclear war has been avoided until now because neither side believed that it was possible to survive such a war. Once it is accepted that thermonuclear war is essentially no more catastrophic than past “conventional” wars, a major restraint will be gone. It is precisely for this reason that we consider it so dangerous to underemphasize the fantastic damage that a thermonuclear war would surely bring about.



Suppose, however, that we are mistaken in our arguments and that Kahn’s most optimistic estimates are right. What would the future look like in that case? We would have a totally effective civil defense program, and so—it must be assumed—would the Russians. The stabilized deterrent would work and war would be avoided for the next ten years, or, if a war came, only 3 to 5 per cent of the population of each side would be killed. But where are we then? Each new generation of weapons becomes more destructive; people get more frightened; the protection that shelters may give today will be useless against the much more destructive weapons of 1970. The shelter idea, adopted, may logically lead to building our cities underground, or—as has been seriously suggested—to selecting small numbers of people to live in such deeply buried shelters that they would be sure to survive; thus a new nation might be built up from a few thousand survivors. Against even this last hope for survival, Kahn admits that ten years from now we, or another nation, might develop a “doomsday machine” which could literally wipe out all life on this globe. What use, then, is even a good shelter program now if it will not halt the march toward doomsday? Or are we to believe that after a thermonuclear war in which only 3 per cent of the population has been killed, the leaders will gain enough wisdom to decide on complete disarmament?

To sum up: in debating the position of Kahn and other experts on civil defense we have been forced to accept their data on technical problems, even though we suspect that their data are themselves not free from the bias which quite naturally comes from their intention to prove that thermonuclear war is not only possible but also “acceptable.” Not being physicists ourselves, and not having the facilities of big research organizations at our disposal, we can at least examine the logic and mode of thought on which the studies of Kahn and the others are based. Our main criticism is that their approach to the question of the survival of our nation, and even of civilization, is the approach of a gamble. That is, they tend to accept the idea that thermonuclear war need not be catastrophic on the basis of (1) complete uncertainty in many important areas of investigation; (2) an optimistic slant which leans more heavily on estimates of better rather than worse conditions; (3) unproven or definitely wrong assumptions; (4) neglect of various important factors as well as the interactions between factors.

At best, what these experts are giving the American public is a piece of optimistic guesswork governed by the logic of a gamble. With the life of a nation and perhaps of all mankind as the stake, it is neither wise nor sane to gamble. Herman Kahn’s arguments leave us convinced that there is only one moral and rational way out of the grim predicament we are in: universal disarmament, combined with a political settlement in which neither side tries to upset the other’s present position.




1 I have discussed the different kinds of thermonuclear war at length in my book On Thermonuclear War and in my testimony (August 6, 1961) before the Holifield Committee. One of the most important distinctions depends on the target system the enemy chooses for his first strike. These can be characterized very roughly as:

  1. Countervalue (attacks against people and property);
  2. Countervalue + counterforce (attacks against people and property and also against retaliatory forces; i.e., bombers and missiles);
  3. Straight counterforce (SAC, ICBM's, Polarises, etc., are the only objective);
  4. Counterforce + “bonus” (people and property are included as secondary targets whenever they can be hit without distracting from the primary military objectives);
  5. Counterforce + avoidance (people and property are carefully avoided where possible and only military objectives are aimed at).

Only numbers 3 and 5 seem to be rational for the attacker (who would then try to use the cities as hostages either to intimidate or to try to force his opponent to negotiate), but any of these attacks could occur. Even the simplest civil defense programs would be extremely effective in the last three cases (keeping casualties in the 1—25 million range instead of the 10—100 million). Quite elaborate programs, however, might fail to protect much more than half the population in the first two cases.

2 Indeed what evidence there is suggests that relatively normal and happy lives would not be impossible even under the relatively harsh conditions that might prevail after a nuclear war, and in spite of the personal and social traumas that would have been experienced. This assertion seems to arouse the utmost hostility in many people, yet it is as well documented a conclusion as many that we have in this uncertain field, and substantially better documented than the opposite view. (See for example Air War and Emotional Stress by I. L. Janis, or the series of studies on disasters and their aftereffects by the Disaster Research Group of the National Academy of Sciences—National Research Council.) It should be clear that to say that suffering would be great but not unlimited, or even to say that one would recover in some sense, is neither to advocate such suffering nor to view suffering callously.

3 Studies that have been done (mainly by RAND, the Stanford Research Institute, and OCDM) indicate that after the first year or so, assuming there is a successful reorganization, the standard of living (including life expectancy and probability of a normal birth) would be higher than the standards prevalent in the U.S. between 1900 and 1930.

4 This is a reasonable estimate of what might happen, given modest precautions against fallout, in the most likely kinds of war that could occur in the early 60's.

5 Actually, because of the desire on both sides to procure weapon systems that are relatively invulnerable, the current technological trend is toward less destructive systems. For example, a Polaris submarine carries sixteen missiles with a total yield in the five to twenty megaton region. A B-52 squadron (fifteen planes)—which under current conditions may be a less reliable deterrent—might be capable of carrying ten to fifty times as much yield. The Minuteman missile which replaces the Atlas-Titan missiles is also likely to carry much less yield per missile than its predecessor.

6 See “Report on a Study of Non-Military Defense” (RAND Report R-322-RC) and “Post-Attack Farm Problems” (Stanford Research Institute, Part I, December 1960; Part II, October 1961).

7 “A System Analysis of the Effects of Nuclear Attack on Railroad Transportation in the Continental United States” (Stanford Research Institute, April 1960) and “Effects of Nuclear Attack on Rail Activity Centers” (Stanford Research Institute, July 1961).

8 It is very difficult to see how any “aggressive” implications can plausibly be read into the current program. Far from contributing to deterrence, a program of concentrating much of our population in fallout shelters that are vulnerable to blast and thermal effects, makes people ideal hostages for an enemy who deliberately spares them on his first strike. However, very few people who have considered these issues would take the position that the probability of war in the absence of this program is so low that the hypothetical decrease in deterrence is enough to outweigh the savings in life and property in the event of a war.
But there is a more legitimate objection to the $207 million program currently being carried out. Even though it emphasizes using existing structures as community shelters, there is still a heavy reliance on individual and local community efforts. This leads to all kinds of exhortatory and sometimes inflammatory speeches or exaggerated claims by civil defense enthusiasts, and raises the “Shoot Thy Neighbor” and other divisive problems. A great country ought not to proceed permanently in such a fashion with the business of defending its citizens against the potential hazards of nuclear war.

9 The phrase, I believe, was first given wide currency by Dr. Robert K. Merton in an article published in 1948. Merton's concern was with how the treatment of ethnic out-groups by ethnic in-groups sometimes acts to bring about the very characteristics condemned in the original stereotypes. He tried to outline the different conditions under which prejudiced prophecies might be defeated rather than fulfilled, and to emphasize that “deliberate institutional controls” could prevent such fears from being translated into reality. Certainly the expectation of hostility from other nations that underlies national defense can have both a self-fulfilling (arms race) and a self-defeating (deterrent) aspect. Which aspect becomes more important depends entirely on our goals (unilateral disarmament, deterrence unless and until better arrangements can be worked out, bargaining from a position of strength, etc.), on the specific situation, and not least on the type (first-strike capability, second-strike capability, active, civil, etc.) and quantity of defense which is under consideration. How much is often as important as what kind.

10 “Hearings before a Sub-Committee of the Committee of Government Operations,” House of Representatives, August 1961. All quotations in this article, unless otherwise designated, are from the record of these hearings.

11 This calculation assumes a strike with a yield no larger than 3,000 megatons, in spite of the fact that we have the power to launch over 40,000 megatons, and there is no proof that the Russians are behind us.

12 In the fire-bombing of Hamburg during World War II, the firestorm caused a ground temperature of 1,400 degrees Fahrenheit in which all exposed humans were incinerated, while those in shelters were asphyxiated or burned.

13 This expectation runs contrary to the experiment of the Naval Radiological Defense Laboratories showing “positive” reactions and “considerable satisfaction in the communal experience” after two weeks. But the NRDL's study has limited applicability; the subjects were volunteer prison convicts who could leave at any time, and are therefore not comparable to survivors, fearful of surfacing to death, chaos, and despair, yet stifling in cramped quarters while they remain sheltered.

14 “The next assignment,” The American Historical Review, January 1958.

15 Kahn himself comprehends the war-provoking aspect of a large civil defense program, and suggests that even a nationwide program of community fallout shelters should be constructed slowly, so as not to “seriously perturb our own people, our allies, and the Russians.”

About the Author

Pin It on Pinterest

Welcome to Commentary Magazine.
We hope you enjoy your visit.
As a visitor to our site, you are allowed 8 free articles this month.
This is your first of 8 free articles.

If you are already a digital subscriber, log in here »

Print subscriber? For free access to the website and iPad, register here »

To subscribe, click here to see our subscription offers »

Please note this is an advertisement skip this ad
Clearly, you have a passion for ideas.
Subscribe today for unlimited digital access to the publication that shapes the minds of the people who shape our world.
Get for just
Welcome to Commentary Magazine.
We hope you enjoy your visit.
As a visitor, you are allowed 8 free articles.
This is your first article.
You have read of 8 free articles this month.
for full access to
Digital subscriber?
Print subscriber? Get free access »
Call to subscribe: 1-800-829-6270
You can also subscribe
on your computer at
Don't have a log in?
Enter you email address and password below. A confirmation email will be sent to the email address that you provide.