In response to:

How Not to Think About Nuclear War from the July 15, 1982 issue

To the Editors:

Theodore Draper [NYR, July 15] makes many good points about nuclear weapons, but he misses one aspect of an argument, and one of his conclusions needs more discussion.

  1. A “no-first-use declaration” would be more than an empty gesture, because it would compel NATO to abandon its strategy relying on the use of tactical nuclear weapons in the event of an attack by the Warsaw Pact conventional forces which could not be stopped by conventional means. Those of us who agree with the late Lord Mountbatten and with Lord Zuckerman that this strategy is unwise and dangerous, would therefore welcome the no-first-use declaration. However, if it is indeed true that NATO conventional forces are inferior to their counterparts, such a declaration will not be made, until this deficiency is rectified. This might include measures which would be politically unpopular.
  2. The present grotesque overkill capacity of each of the superpowers is defended by their hawks as necessary for an effective deterrent because a “first strike” by the enemy could destroy all their missiles. To be sure of deterring such an attack, each side must therefore, it is claimed, have as great a destructive power as the potential enemy, or preferably greater.

This, as Draper rightly points out, ignores the submarine-launched and cruise missiles, which in the present state of technology, would survive a first strike and be available as an effective deterrent after such a counterforce strike. It is argued, however, that there is no guarantee that one day effective means of submarine detection and defence against cruise missiles may not be developed, and while the experts regard this as unlikely, one should plan for the “worst case.”

The question turns, therefore, on whether one believes that, even without submarines and cruise missiles, a counterforce first strike is a realistic possibility. Would any government stake the survival of their country on a highly complex operation functioning with high efficiency and with nearly perfect timing? Failures of missiles are not uncommon, in spite of careful preparations. Even with space shots, prepared and supervised by a collection of the most highly qualified experts, delays due to malfunction of vital components are frequent. One knows that in the heat of battle the efficiency of equipment and of the humans who operate them, is very different from what can be done in the laboratory, or in computer war games.

This is the problem on which a realistic assessment by experts is needed. I believe the answer would show that the counterforce first strike belongs to science fiction. If so, Draper is right about the dangerous waste of the present stockpiles.

Rudolf Peierls

Oxford, England

To the Editors:

In his otherwise lucid and useful article on thinking about nuclear weapons Theodore Draper fails to understand that no-first-use, negotiated arms reductions, and deterrence are interdependent. Draper rightly identifies the culprit—weapons designers who are constantly improving their product, making them more accurate and easier to use—but in my opinion he misses the point, which is that better weapons threatens deterrence.

The problem is that deterrence cools and shrinks. In the 1950s the near-monopoly on nuclear weapons of the United States effectively deterred a whole list of undesirable things—not just a Soviet attack on the United States, but war in Europe or adventurism elsewhere as well. The outcome of the Cuban missile crisis was the result of the nuclear superiority which allowed us considerable leeway in pressing the Soviets to back down. The Soviet did not miss the significance of this broad and robust deterrence. Vasily Kuznetsov told John McCloy at the time that we would never be able to do this to them again. Shortly thereafter the Soviets embarked upon the long strategic buildup which has resulted in their current status of rough parity with the United States.

This parity is not without practical result. The list of things we can deter has cooled and shrunk. Even a decade ago American strategic superiority allowed us to threaten direct attack on the Soviet Union to deter war in Europe. This is no longer the case. Such an attack would bring a devastating response. We know it, and the Soviets know we know it. Our threat to risk a big intercontinental nuclear war in defense of Europe is simply not credible. We don’t believe it ourselves. In their roundabout way American military leaders have been trying to confess this for the last decade. A threat to use nuclear weapons on a large scale still deters, but it only deters one thing—a big attack. Anything lesser would bring a lesser response, or no response at all.

When James Schlesinger reintroduced the idea of flexible or limited use of nuclear weapons in 1974, he was trying to address this problem. At that time a limited execution of the Single Integrated Operating Plan—the SIOP—called for a minimum initial salvo of about 2500 nuclear warheads, pointed at Soviet nuclear and conventional military targets. That was our idea of a warning shot. The Soviets could hardly have responded to attack on such a scale with anything less than all-out war. In short, we had painted ourselves into an all-or-nothing corner. The current talk of “protracted war” theories in Washington is an attempt to get out of this corner. In my opinion these theories have a fatal weakness of their own—the nearuniversal belief, even in military circles, that limited use of nuclear weapons would escalate into all-out use—but even so I think it fair to grant these theories are a rational attempt, at least, to deal with the shrinking of deterrence, and to get out of the all-or-nothing corner.

Draper is right to be skeptical of our ability to change the world in dramatic ways, and right to emphasize that about the best we can hope from nuclear weapons is deterrence. But we can never go back to the old robust deterrence. The threat of a big nuclear response can deter a big nuclear attack, but that is all it can deter. If we want to deter limited nuclear attacks, or conventional war, we have got to do it with something other than a threat to destroy the world which not even we take seriously.


This is where I think Draper goes wrong. If we are to deter a conventional war without running the risk of nuclear war, we have got to deter it in a conventional way. A no-first-use pledge would be an important step in this direction, not because the Soviets would believe it, but because we would have to reshape our military plans accordingly. I think Draper underestimates the effect of such a pledge on the military planning process. We can be sure the Pentagon would make itself heard if we renounced plans to use tactical nuclear weapons in the event of war, without making conventional provision to take up the slack. As things stand now, conventional war in Europe means nuclear war.

That leaves the problem of deterring limited nuclear attack. This can no longer be achieved by threatening all-out nuclear response. The consequences of such a response would be too great for it to be credible. This pretty much means that the threat of a limited attack can only be deterred by the threat of a limited response—“protracted war” under one name or another. But limited nuclear war is only a hair short of being as dangerous as all-out nuclear war, something just about everybody seems to recognize. The move to pure minimum deterrence which Draper seems to favor, and which makes a lot of sense if we agree with Draper that the total abolition of nuclear weapons is beyond us, requires two difficult steps—a sharp reduction in the numbers of nuclear weapons on both sides, and some degree of change in the weapons themselves to make them less threatening to each other. Both steps clearly depends on successful negotiations with the Soviets. In their absence we are stuck with some sort of limited nuclear war-fighting capability as the only practical deterrent to limited nuclear attack.

I stress these points because Draper too easily dismisses the proposals for no-first-use and for negotiated reductions—both important—and because he seems to feel the United States can unilaterally return to a posture of robust deterrence based on a single threat of all-out response to attack. Things have gone too far for that. The weapons designers have put paid to such an approach. The Soviets may not have to contend with public agitation in the Western style, but they do not need it. They can calculate the threat posed by vast numbers of versatile weapons as well as we can. It is the scale of that threat—persuasively described by Jonathan Schell—which explains our current and thoroughly justified alarm. If anything is to be done about it, it must be done by both sides together.

Thomas Powers

South Royalton, Vermont

To the Editors:

Your recent article by Theodore Draper is indeed a demonstration of how not to think about nuclear war. Draper’s contentious approach allows him to score points, but in doing so he discounts the authors’ efforts to confront the fears that fuel the arms race. He emphasizes weak points in the books under review and tends to disregard the strengths of their arguments. This approach to such a serious matter as the nuclear arms race reveals a sad lack of comprehension of Jonathan Schell’s major message. The Fate of the Earth is certainly often hyperbolic and vague, nevertheless it communicates, at times brilliantly, the conviction that none of the objectives for which wars have been fought throughout history can possibly be achieved if nuclear weapons are used. Therefore our thinking about wars and conflicts between nations has to change.

This change will have to include our approach to the Soviet Union. It is unrealistic to discuss the lasting trauma of the memories of the second world war in Western Europe without considering the similar trauma in the Soviet Union. As George Kennan and others who know the Russian people have attested, their sentiment for peace, although it cannot be organized, is nevertheless both strong and important politically.

Undoubtedly the fears arising from the experiences of WWII, including the specter of Munich, have been exploited by leaders of both sides to justify the ever-increasing expansion of the arms race. The advantage of proposals such as No-First-Use, Nuclear-Free Zones and the Nuclear Freeze lies in their attempt to defuse the root fears which drive the arms race, while at the same time halting the ever more dangerous arms build-up.

In our opinion, the no-first-use proposal is crucial. Contrary to what Draper argues, there is an enormous difference between a no-use pledge and a no-first-use pledge. Unfortunately, at this time, the former is not credible. But a no-first-use pledge is believable, because it makes it clear that the only use of these weapons is for deterrence. Since it works to make nuclear war less likely, it is in the self-interest of all parties.


The 1925 Geneva protocol banning the use of poison gas has been in fact interpreted as a “no-first-use pledge,” since the major signatory countries stockpiled poison gases for deterrence, and trained their troops in poison gas techniques. Nevertheless the gases were not used in combat, even in as severe a conflict as WWII. The fact that this ban held under those conditions gives one hope that an analogous no-first-use pledge for nuclear weapons would hold in the future. In the present situation, this pledge, in combination with a freeze on nuclear weapons development and the establishment of nuclear-free zones, would improve the political atmosphere by reducing the legitimate fears in both the East and West.

Aron M. Bernstein

Massachusetts Institute of Technology

Cambridge, Massachusetts

Jeanne Theis Whitaker

Wheaton College

Norton, Massachusetts

To the Editors:

Theodore Draper argues that the United States should follow a nuclear arms policy based on the principle of minimum deterrence. His argument rests on three points. For the foreseeable future, nuclear arms cannot be completely gotten rid of. The Soviets will not launch a first strike if they believe we will be able to retaliate. And, building more weapons than the minimum required for deterrence is potentially destabilizing and needlessly costly. To me, at least, his argument is convincing.

But we passed the point of minimum deterrence quite some time ago. Until the Carter administration, US policy makers espoused the principle of minimum deterrence publicly, while continuing to overstock nuclear weapons. The Carter administration began to talk about fighting “limited” nuclear wars. Now the Reagan administration is talking about winning these wars—whatever that means.

As Mr. Draper points out, the Reagan administration has expanded the concept of deterrence in a way that is internally inconsistent. It has asserted that the threat of retaliation alone is no longer adequate to deter a Soviet attack. Now, to convince the Soviets it is not in their interests to launch nuclear warheads at us, we must make them believe that we can outgun them in a protracted nuclear war. I think Mr. Draper clearly demonstrates the absurdity of this proposition.

What bothers me about Mr. Draper’s article is what it omits in this connection. Why has the Reagan administration adopted this new line? To say that “there are some people in high places who are interested in fighting as well as—or even more than—deterring a nuclear war,” is not to say enough. Why did previous administrations—those that did not speak of fighting nuclear wars—continually overstock nuclear weapons? This, as Mr. Draper himself points out, is not a new policy.

Mr. Draper, with Solly Zuckerman, contends that scientists in the weapons developing bureaucracy are major proponents of overdevelopment. But I would guess that they are not alone. Defense contractors have probably exerted a bit of pressure too. So have members of the military.

Mr. Draper’s argument in support of minimum deterrence is logical, articulate, and I think correct. Unfortunately, although his article hovers around the question of why we began overstocking nuclear weapons in the first place, it does not directly address it. But it is only by answering this question that we will come to understand the interest politics associated with nuclear weapons within the United States. And it is only by manipulating those politics that we can hope to work toward a policy of minimum deterrence.

Daniel B. Kaplan

Brooklyn, New York

Theodore Draper replies:

Let me start with the problem of no-first-use, which is raised by Sir Rudolf Peierls, Thomas Powers, and Aron M. Bernstein-Jeanne Theis Whitaker.

I do not disagree with Professor Peierls in one respect. It would be best for all concerned if an attack by the Warsaw Pact conventional forces could be “stopped” by conventional means. I wrote in my article: “If Europeans have to choose between a conventional war and a nuclear war, they would be mad to choose the latter. A conventional defense is really the only kind of defense rationally open to them in the event of an actual war; a nuclear defense can be justified only as a way to prevent all war conventional and nuclear.”

But then Professor Peierls add a necessary condition to no-first-use—that NATO conventional forces should be adequate for the task of “stopping” a conventional attack. The implication is that tactical nuclear weapons should not be used if a conventional defense is adequate, but that tactical nuclear weapons might be used if a conventional defense is inadequate. Everything, then, hinges on the adequacy of a Western conventional defense—an adequacy which no Western nation is prepared to guarantee, and the measures for which may be so “unpopular,” as Professor Peierls suggests, that no Western nation could actually impose them.

In any case, I had already pointed out: “It is important to note that no-first-use is only half of the new equation proposed by the four [American] authors. The other half ‘would require’ a greater conventional defense force in Europe, including a strengthening of the American conventional forces there. No-first-use, then, does not stand by itself; it depends on what happens to the conventional forces of the alliance.”

Here we come to a paradox which is not sufficiently confronted. It is one thing to have tactical nuclear weapons in place, but pledge no-first-use of them as long as a conventional attack can be “stopped” by a conventional defense. It is quite another thing to remove and do without tactical nuclear weapons altogether because a no-first-use policy has been adopted. The former is implicitly the position of Professor Peierls, for he would not make such a declaration if a conventional defense proved to be inadequate. No one will ever know whether a conventional defense is adequate until it is tried and has failed, whereupon it would be too late to put tactical nuclear weapons in place. Nor is it clear what “stopping” an attack by Warsaw Pact conventional forces entails. If it means stopping it from ever getting started, that too is one thing. But if it means stopping it from achieving its objective and repulsing it by means of a clash of conventional forces, that is again something else. Such a clash, conventional though it may be, would be so destructive in itself, if less so than a nuclear exchange, that Europeans can hardly be expected to find it acceptable. It is paradoxical to have weapons, which might make the difference between deterrence and defeat, and promise not to use them as long as the other side is content to use other weapons to its own advantage.

The only way to escape from the paradox, as far as I can see, is to think in terms of deterring a conventional as well as a nuclear war in Europe rather than of setting up conditions for a supposedly acceptable conventional war or for planning a protracted nuclear war. Such deterrence in the present circumstances can only be conventional plus nuclear, because it must rest on mutual destructiveness. There is no equality of destructiveness if one side must resign itself to a defense waged on its own territory, while the attacking side is safe from retribution on its territory. A conventional Western force given the sole mission of “stopping” a Warsaw Pact attack in Western Europe, whatever that means, has nowhere to go but backward. There is no way of getting around the oppressive reality that anything which weakens nuclear deterrence makes a conventional war more feasible. That is not a fact of nuclear life that Europeans or Americans should regard with equanimity.

I do not find in Professor Peierls’s letter, perhaps because it is too short, a consideration of these painful choices. I would like to agree with him, but I am not altogether sure what I would be agreeing with. I would certainly like to know more about how one could tell in advance whether a conventional attack could be “stopped” by conventional means and what “stopped” really means in this context. I wholly share his interest in getting a more realistic assessment of the so-called counterforce first strike.

Thomas Powers raises some of the same issues in his letter. Among other things, he charges that I fail to understand the interdependence of no-first-use, negotiated arms reductions, and deterrence. If they are interdependent, as he thinks, we are in a very bad way. Interdependence implies that they must stand or fall together. If so, deterrence is dependent on two very weak links, so weak that the chain can hardly be said to exist.

No-first-use is merely a declaration of intention, with nothing to back it up. Negotiated arms reductions that will really reduce the ability of both superpowers to annihilate each other are as far away as ever. If deterrence is interdependent with them, we might just as well give up the idea.

Deterrence rests on stronger grounds and on its own. It is simply based on the irrationality and inexorability of a mutually devastating nuclear war. There is no necessary connection between the irrationality and inexorability of such a war and no-first-use or arms reductions. We have not had no-first-use or meaningful arms reductions for decades, but we have had some assurance of deterrence by virtue of the inability to use nuclear arms without self-destruction.

The distinction between “robust” and “shrunken” deterrence is equally misconceived. There was nothing “robust” about the first American advantage in atomic and then nuclear arms, if “robust” means a large stockpile; the advantage was simply unilateral. That type of nuclear deterrence was also unilateral. When the Soviets developed their own nuclear weapons, deterrence became mutual. The real distinction is between unilateral and mutual, not robust and shrunken.

The body of Mr. Power’s argument in behalf of “protracted war” strikes me as hopelessly confused and self-defeating. We are told that all those theories leading to protracted war have “a fatal weakness.” If so, why are they “rational”? What is rational about fatally weak theories?

In his seventh paragraph, Mr. Powers uses “protracted war” to refer to a conventional war. I was concerned in my article with the plan for a protracted nuclear war. By confusing the two, Mr. Powers has the protracted war that I abominated, but it is not the same kind of war. This American advocacy of a protracted conventional war in Europe is anathema to most thoughtful Europeans. As the four distinguished German writers—Karl Kaiser, Georg Leber, Alois Mertes, and Franz-Josef Schulze—put it in the summer 1982 Foreign Affairs: “What matters most is to concentrate not only on the prevention of nuclear war, but on how to prevent any war, conventional war as well.” That point of view pervaded my article.

If, as Mr. Powers says, we must deter a conventional war in a conventional way, then the Western alliance is going to have to fight a conventional war in Europe, not in the United States or in the Soviet Union. With the present destructive capability of conventional weapons, that is not a prospect as alluring to Europeans as to Americans and Russians. Nuclear deterrence rests on the prospect of mutual Soviet, European, and American devastation—let us put it as plainly as that. Conventional “deterrence” cannot hold out to the Soviets—or the Americans—the same sort of retribution as that in store for Europe. Again, the viewpoint of the four German writers above should be taken to heart by Americans: “One cannot help concluding that the Soviet Union would thereby [through the renunciation of first use of nuclear weapons] be put in a position where it could, once again, calculate its risk and thus be able to wage war in Europe. It would no longer have to fear that nuclear weapons would inflict unacceptable damage on its own territory. We therefore fear that a credible renunciation of the first use of nuclear weapons would, once again, make war more probable.”

Ironically, the main opponents of no-first-use are in responsible positions in Europe, not in the United States. It is doubtful whether the military alliance could be held together at all if the United States gave up first-use. That is the side of the problem stressed in my article and not addressed at all in these letters. The Europeans abhor the idea of nuclear war but cannot give up the ultimate threat of nuclear retaliation to prevent both nuclear and conventional wars in Europe. This infernal paradox bedevils the entire subject and should make one wary of attributing first-use to the machinations of the Pentagon.

It is bootless to make first-use into a bogy, as Mr. Powers does with: “As things stand now, conventional war in Europe means nuclear war.” As things stand now, conventional war in Europe may mean anything from conventional war to nuclear war, depending on circumstances. The Western alliance adopted a policy of “flexible response” which Mr. Powers chooses to make inflexible. This flexibility extends to the possible use of nuclear weapons, without which it could no longer be considered flexible. The Europeans, and most of all the Germans, who are on the front line, insist on such flexibility precisely because they wish to discourage both conventional and nuclear wars on their territory by holding out the threat that the attacker could not count on keeping his own territory as a privileged sanctuary.

Mr. Power’s conclusion baffles me. He attributes to me the notion that “the United States can unilaterally return to a posture of robust deterrence based on a single threat of all-out response to attack.” My article was entirely based on the premise that nuclear war had become mutually destructive and, therefore, had brought about a state of mutual deterrence. A return to unilateral American deterrence was the last thing I had in mind or suggested, though it may be recalled that the United States did not know what to do with a unilateral advantage when it had it. There was nothing original about my position: I merely protested against giving it up in favor of denying the value of deterrence à la Schell or of perverting it to mean protracted nuclear war fare à la Weinberges.

The Bernstein-Whitaker view of no-first-use is exceptionally unpersuasive. They distinguish between no-use and no-first-use on the ground that no-first-use “makes it clear that the only use of these weapons is for deterrence.” How can that be? If we promise no-first-use in any circumstances, we give up its deterrent value, whatever that may be. We cannot deter with a weapon we will never use except as retaliation for its prior use by the other side. Instead of deterring nuclear war, we would already be in it, thanks to first-use by others. The illogic of this argument should be clear even to professors at MIT and Wheaton College.

As for the Russian people’s sentiment for peace, I cannot see how it can be both strong and important politically if it cannot be organized or, indeed, take any other political form. Would the American or the European people’s sentiment for peace be strong and important politically if it could not be organized? Of course not. This sort of academic apologetics for the Soviet system is sickening.

The question asked by Mr. Kaplan is not one that I can safely answer. If I had to guess, it would be that the very existence of an entire range of nuclear weapons provides a temptation to find a rationale for them. Vested interests in such weapons must also play a part, though I doubt whether it is the major factor. Sheer human folly, even in high places, must never be underestimated. I am baffled by people who think that we must be ready and willing to fight a “limited” nuclear war, but who also believe that such limited use would escalate into an all-out nuclear war. They might just as well believe in the latter as in the former. But then, I am baffled by all sorts of things, so that I am not as mystified by wrongheadedness as Mr. Kaplan seems to be.

This Issue

September 23, 1982