Caspar Weinberger
Caspar Weinberger; drawing by David Levine

As the United States faces a series of tangled and potentially divisive choices in its policy toward nuclear weapons, these books by young British scholars make serious contributions to our understanding of central elements of the problem. Lawrence Freedman tells us the history of our good and bad thoughts about nuclear strategy, and David Holloway describes more briefly the history of nuclear strategy on the Soviet side. Each has mastered a formidable and often opaque “literature.” Both have been respectful of the record but bold in judging it, and they write with a clarity and fairness that are distressingly rare in American debates. Neither is peddling a panacea, and citizens trying to find a sensible course between unilateral disarmament and the miscellaneous nuclear expansionism of the Reagan administration will find solid help from both. Moreover, each author is meticulous in giving credit to those on whose works he builds, so that each book is not only a free-standing work of analysis but also an excellent guide to further study.

The central lesson of Mr. Freedman’s magisterial survey is that the realities of nuclear weapons have defied the best efforts of the nuclear strategists. No one in the West has found a good or even an acceptable way to fight a nuclear war against an opponent with thousands of nuclear weapons of his own that are “survivable,” i.e., that can still be used after a first strike. “Those who have responsibility for unleashing nuclear arsenals live by the motto that if they ever had to do so they would have failed. Remarkably, up to now they have succeeded. C’est magnifique, mais ce n’est pas la stratégie.”

The absence of a workable strategy for nuclear war-fighting is not the result of inattention or stupidity. Freedman has much to say about the remarkable men who have tried and failed to tame the beast: Vannevar Bush and P.M.S. Blackett claimed that the weapon was not all that destructive, but their arguments were shattered by the hydrogen bomb and the long-range missile. Robert Oppenheimer and Henry Kissinger were attracted in quite different ways by the possibility that nuclear battlefield weapons could be deployed to keep the peace through preparation for a less than apocalyptic war but eventually they were persuaded otherwise. Robert McNamara (for a short while), James Schlesinger, and Jimmy Carter—a strangely assorted trio—also tried to design “limited” uses for nuclear weapons that might somehow be “remotely plausible”; but none of them came close to demonstrating that any such “limited” war could in fact be kept under control. Finally Freedman discusses the much more crude and still less credible dabblers in ideas of meaningful victory, from the theorists of triumphant air power in the 1940s to the brash young civilians still arguing in 1980 that “victory is possible.” It is somehow fitting that Ronald Reagan himself, fearing that he might be supposed to think otherwise, has given that proposition its definitive denial: “A nuclear war cannot be won and must never be fought.” Whether everyone in the Pentagon has heard the president may be another question.

What was learned and learned again in this long effort was what so many now understand in their bones: once they are many, big, and fast, these weapons do not lend themselves to any carefully planned or limited use of any sort, because no one can tell what will happen after even one of them is used in anger. There has never been a war game that offered comfort to either side. No one knows how any “limited” use would be answered. The most accurate long-range warheads on both sides would kill millions, by fire and fallout, even if, improbably, both sides tried not to hit a single town. Of course, no one can prove that any first use of nuclear weapons will lead to a general conflagration. But what is decisive is that no one can come close to proving it will not. Freedman demonstrates that this hard lesson is the product of a generation of efforts to disprove it.

Did the nuclear strategists teach us nothing, then? Far from it. It was the first and best of them all, Bernard Brodie, who understood at the beginning that the basic rules had changed forever. “Thus far,” he wrote, “the chief purpose of a military establishment has been to win wars. From now on its chief purpose must be to avert them.” To remember that—to aim at deterrence—was the key to good choices.

Moreover, strategists had a major part in winning recognition of the single most important element in effective deterrence: the survivability of deliverable weapons. Brodie was the first to argue this as well, but probably Albert Wohlstetter had the greatest effect, excepting only Dwight Eisenhower (whose early recognition of the problem of preventing surprise attack Freedman notes, but not as sharply as I would). Wohlstetter defined with clarity the decisive difference between first-strike and second-strike weapons—what you can do only by attacking first and what you can do after absorbing an attack—and underlined the absolutely critical importance of adequate second-strike capacity. The requirement for survivable second-strike nuclear strength governed the excellent weapons choices of the later years of the Eisenhower administration, and was respected by its successor. No other single requirement is so important to maintaining deterrence today. Of no small significance is the fact that the first people to insist on this basic analytical distinction were not the organized military, but outside civilian advisers and a president who had enough self-confidence to seek his own counsel and impose his own judgment.


The distinction between first-strike and second-strike forces is as valid today as it was twenty-five years ago. Failure to respect it is the underlying weakness of the long and so far unsuccessful campaign to persuade the Congress and the country to build the MX missile. There are many strange elements in the story of the MX, but what is fundamentally wrong with it is that it was not designed with its own survivability as the governing requirement. It is big—ninety-six tons—partly because the defenders of SALT II wanted to be able to say that under their treaty we could have missiles as destructive as those the Soviets already had, but its enormous size makes it hard to hide and move. The missile has ten warheads mounted on a single launcher, partly because SALT II limits launchers more tightly than warheads. (An MX with ten warheads counts as one launcher and so does the existing single-warhead Minuteman II, which MX was initially designed to replace.) But many warheads on a single missile make a tempting target.

To see why things went wrong here we must look deeper at what really governed the initial drive for MX. It became the dream weapon of the Air Force because it was designed to serve a governing purpose quite different from that of survivability. It was designed to be a counterforce weapon, one that could be used to kill Soviet missiles in their holes—destroying more than one such missile with each MX—and also to kill other hard targets essential to Soviet nuclear war-fighting: command posts, communications centers, and the like. In explicit doctrine MX is not a first-strike weapon; our doctrine does not say we will ever go first in an all-out strategic exchange. But MX is a war-fighting weapon, and to its ultimate sponsors in the Air Force its capacity to kill has always been more important than its capacity to survive. In that sense it is a weapon that reflects not the hard-won wisdom of Brodie, Wohlstetter, and Eisenhower but the primeval commitment to the offense of a much earlier tradition.

If, like the Air Force, you put first the war-fighting requirement of being able to hit enemy missile sites as fast as possible, MX is a technical marvel. But if you accept the imperatives of longstanding strategic thinking and put first the requirement of survivability, it is a technical monstrosity. Advertised as the answer to the emerging vulnerability of the existing Minuteman missiles, it is itself less survivable, warhead for warhead, than they are. It is not a weapon that respects the first rule of deterrence.

Having shared the desire to defend SALT II, and recognizing how this and other secondary considerations came to have the undue weight that they did, I do not criticize MX in any spirit of selfsatisfaction. Moreover, I respect the understandable institutional commitment of the Air Force to maintaining its traditionally central position in strategic deterrence. It cannot be easy or pleasant for Air Force officers to recognize that for some kinds of weapons today missiles based at sea are better than those based on land. But the interests of the military services do not come before the national interest, and “counterforce capability” is not of the same order of importance as survivability. The grave and self-destructive mistake that has been made through four administrations by the backers of the MX would never have occurred if the requirement of survivability had been put first, where it belongs, and not merely tacked on as an afterthought for designers of basing modes.

The primacy of deterrence over war fighting and of survivability over counterforce strategy are the beginning but not the end of nuclear wisdom. It is also necessary to decide how much is enough. Here again Freedman has much to tell us. He recounts fairly the debate over “mutual assured destruction” as a criterion for a sufficient force, noting, as its American critics seldom do, that the attack on cities as such was never a strategy that the US planners preferred. Indeed the original sponsors of this criterion were also the first to develop options for sparing cities. But Freedman also notes, and Holloway conclusively demonstrates, that there was and is no symmetry between any American calculation of how much potential destruction of Soviet military and industrial targets would be “enough” to deter an attack and what the Soviets have chosen to build. Neither in doctrine nor in procurement have the Soviets been governed by criteria of assured destruction; they have built weapon systems much too big to fit that standard. Nor have they been influenced by notions of limited nuclear war, which they have regularly denounced. About the realities of Soviet belief and behavior, neither our theorists nor our military officers have been notably clear-headed; analysts of all persuasions have been inclined to make the easy and erroneous assumption that what we think is what the Soviets will think when they pay attention.


Holloway knows better. Using a wide range of Soviet and Western sources, and basing his analysis on careful study of “the historical experience, the policy objectives and the institutions that have sustained the Soviet military effort,” he has produced a remarkably acute assessment of what drives Soviet military policy, especially in nuclear matters. The Soviet decision makers he describes, for their own strong reasons, have been businesslike and determined about nuclear weapons. After Hiroshima they wanted them just as fast as they could get them; they were equally quick and energetic in their pursuit of thermonuclear weapons a few years later. They were slow, if only marginally slower than the Americans, to see the advantages of arms-control agreements. They were wary of large bombers and were early believers in missiles, which they call rockets. They believe in deterrence, but talk about it as part of wider political aims, such as preventing “imperialist war” and strengthening “peace forces.”

According to their doctrine, only imperialists might unleash a nuclear war, but if such a war is thrust upon the Soviet Union, they will do their utmost to win—or, in Pentagonese, “to prevail.” They like their nuclear weapons big and numerous, and the targets they have in mind, including both military installations and industrial centers, are those that will contribute to defeating the enemy. The ability to fight, not just the ability to destroy, is the objective of Soviet military policy, and Soviet weapons systems have been developed and deployed in numbers and magnitudes that reflect this objective.

Soviet military thought is not, in this respect, deeply different from our own, but military attitudes there face far fewer constraints. “Soviet military power is not something separate from the Soviet state, but forms part and parcel of it…. Its domestic roots—ideological, political, social and economic—are very strong.” Such doctrines and such institutional power, made to seem more threatening by some vainglorious military pronouncements, mostly from earlier years, have been swallowed whole by some of the Western ideologues who proclaim that “the Russians think they could fight and win a nuclear war.” Such evidence of Soviet military capacities and attitudes, real enough as far as it goes, is treated much too casually by those on the other side of our debates who believe that if only our own misbehavior could be corrected, progress toward general nuclear disarmament would be swift. Holloway is polite but firm in his advice to Western peace movements: “It is intellectually unsatisfying and politically dangerous to ignore Soviet military power and the threat it can pose to Western countries.”

But while Holloway ducks none of the evidence on Soviet nuclear strategy, he does put it in perspective: from his study of Soviet military politics he concludes not only that the basic objective of all this planning and investment is to prevent nuclear war—to deter—but also that it coexists with a strong and increasingly evident political recognition that the Soviet relationship with the United States “is in reality one of mutual vulnerability to devastating nuclear strikes…. There is little evidence to suggest that [the Soviets] think victory in a global nuclear war would be anything other than catastrophic.” Holloway has collected authoritative quotations which demonstrate that Soviet leaders have little to learn from Jonathan Schell in understanding the undesirability of general nuclear war. Brezhnev said it succinctly and repeatedly: “To expect to win a nuclear war is dangerous madness.”

If this analysis is correct, and I think it is, we are provided with one rough but useful standard for judging how much is enough. What we have now is at least enough, since it is what we have that forces the Soviet Union to reach sensible conclusions. As long as we ensure that a sufficient number of strategic weapons will survive attack, we do not need more. To those who believe that we have too much already, this conclusion will seem mild, but for our present political debates it is not trivial. What it means is that the complex argument about “counterforce” and “prompt hard-target kill capability” and the like are wholly secondary to the prevailing reality concerning deterrence: that the Soviet government has no more desire for general nuclear war than we do, and that the persistence of this sensible attitude is amply assured by what our current forces could do should the USSR launch an attack.

Holloway provides an excellent short summary of the current Soviet-American balance which persuasively underpins the conclusion that nuclear deterrence today is stable and strong. Each side has thousands of survivable thermonuclear weapons—in our case primarily though not solely on submarines at sea—which are more than ample for replies of all sorts against both military and industrial targets. Holloway does not commit the deceptive error, to which both Reagan and Weinberger are addicted, of comparing only the landbased missiles on both sides; they make up three-fourths of the whole Soviet force but only one-fourth of ours.

Both Holloway and Freedman convincingly refute the nightmare of “Minuteman vulnerability”—the notion that the growing accuracy of Soviet missiles creates a grave present danger of early Soviet preemptive attack, or threat of attack, on these US land-based missiles. Not only is the success of any such attack deeply uncertain, but its consequences are wholly unpredictable. In Freedman’s words it “would be an immoderate act, that could not deserve a moderate reaction,” and many such reactions are possible. This inventive nightmare was fit only for zealots in search of office, and it is good to see it fading away now as they tell us instead of the persuasive deterrent force of Ronald Reagan’s will. Freedman is right in his comment on the situation at the height of this debate in 1980: “The fact that ICBM vulnerability provided the greatest cause for concern could be taken as a symptom of the underlying stability of the strategic balance.”

The reality, as policy and technology have combined to create it, is that we are still living in the situation of sturdy balanced terror that both Churchill and Eisenhower foresaw nearly thirty years ago. Moreover, both sides have learned that effective superiority is unobtainable, although illusions on this point may have lasted longer in the United States than they have in the Soviet Union. One of Holloway’s most persuasive conclusions is that parity, not usable superiority, is now both Soviet policy and Soviet expectation. There is much room for argument on just what parity means, between forces as different in detail as they are alike in their immoderate size. Soviet policy makers need not be expected to give up their persistent search for marginal advantages whether by deployment or negotiation. Nothing in Holloway’s analysis of the Soviets shows us men who are interested in solving our American problems—only men who know that the largest interest of all, avoiding general thermonuclear war, is one that both sides share.

The strategic balance today is stable; it will not be threatened except as the general capacity of nuclear forces on both sides to survive an attack might over a long period be threatened; and the primacy of this test of survivability is reinforced, not undermined, by what the Soviet government believes about nuclear war.

The strategists who first emphasized survivability took pains to note that achieving it would not be easy, and that warning is worth new emphasis today. That the Soviets themselves hold that their deterrent forces must be capable of fighting a nuclear war, as Holloway shows, complicates our task. Not only must we have survivability against heavy and increasingly accurate missiles; we must have it against a possible effort—if war comes—to strike early and hard against our whole system of command and control, both by eliminating our leaders in charge and by blocking their electronic communications. There would be an element of madness, we may think, in such a choice, but we must take our own measures against it. For survivable deterrence (and for ending a war quickly, if deterrence ever fails) we do need to make sure the Soviets never believe they could eliminate our capacity for robust command and control.

Such survivable capacity for fundamental decisions is quite different from the will-o’-the wisp of carefully managed “escalation control.” Nonetheless the matter needs attention, and the most sensible single element in Mr. Weinberger’s program for strategic modernization is his emphasis on improving the survivability of our command and control systems. Both sides would do better, plainly, to leave each other’s capacity for making rational decisions untouched, and this mutual interest deserves high priority in negotiations for building confidence. But before we complain too much about Soviet attitudes here, we should remember, with Freedman, how it was thought clever, in the Carter White House, to talk about targeting what “the Soviet leadership values most—its military forces and its ability to maintain control after a war starts.”

There remains the troubling matter of the nuclear defense of Western Europe. What can these books tell us about deterrence there, and especially about the future of the current plan to deploy 572 new American missiles in Europe unless they can be traded away for adequate Soviet concessions? Both authors address the general problem, Freedman noting that the role of American nuclear forces in the NATO alliance has been a matter of debate almost from the beginning, and Holloway addressing more directly the current effort to reach an arms-control agreement.

Freedman reminds us that historically almost any change in the emphasis of American strategy has been troubling in Europe, at least initially. Even when we have had a good case, as in moving from “massive retaliation” to “flexible response,” we have found it hard to express ourselves with sensitivity to European opinion. When the US stresses the need for greater strength of European conventional forces as the most credible and reassuring deterrent to a conventional attack, US arguments are easily perverted into assertions by Europeans that it is backing away from its commitment to protect Europe with nuclear weapons.

During the last year, in urging the consideration of a further step in arms control—a policy of “no-first-use” of nuclear weapons—a number of us have encountered similar reactions from European politicians. Thus it is not surprising that in dealing with the plans mooted during the late 1970s for new “intermediate” missiles that would counter the Soviet SS-20s, many Europeans were initially suspicious of US reluctance to go ahead with them and now object to our seemingly rigid insistence on all or nothing—i.e., either on President Reagan’s “zero option” or on deploying all the Pershing II and cruise missiles that were planned.

But the central problem is not American insensitivity, unless we make it so. The proper questions concern what Europeans want and what the Soviets will concede. The United States does not need 572 new weapons in Europe to hold Soviet forces “at risk.” There is nothing they can do that could not be done as well by one or more of the many other systems we have on hand or on order, notably the Poseidon submarines already assigned to NATO (their “capabilities” may not be precisely tailored for war fighting, but they are ample for deterrence) and cruise missiles on strategic bombers (often praised in other contexts for their versatility).

Nor does the Soviet government, as distinct from the Soviet military establishment, have any overwhelming need for an SS-20 missile. The Soviets already have two other modern systems at least partly targeted against Europe, the Backfire bomber and the large SS-19 missile. Either one alone is ample for deterrence, and the incalculable risks of war-fighting, with any of these systems, are all the same. In short there are other systems, available to both the president and the Politburo that make the missiles now under negotiation entirely marginal. Holloway, understanding the weight of the Soviet military in the Soviet government, believes that the attachment of the Soviet military leaders to the SS-20 may turn out to be controlling. He observes that the Soviets may also prefer an effort to divide the NATO alliance to any reasonable agreement.

Our task must be to frustrate that Soviet purpose, and the only way to do so is to make it wholly clear that our firmness and flexibility in Geneva will be the firmness and flexibility desired by our allies—and mainly by the Germans. We already know what a bargain might look like. It would not be an agreement on no intermediate missiles at all—the administration surely knows its first position cannot be its last. It would be an agreement for fifty or seventy-five or one hundred of some kind of launchers for both sides—that is what our negotiator Paul Nitze has already discussed informally with his Soviet counterpart Yuli Kvitsinsky, and it is the answer independently reached by most of the experienced experts. I have talked to or heard from in Europe and the US. The issue, as we have seen, is not a matter of military strength. But it has unfortunately become a matter of political symbolism—of leaders’ not wanting to appear before their various constituencies as too flexible or inflexible, too subject to American or Soviet demands or too intransigent in the face of them. The military reality is that any numbers that can be agreed on will serve.

But Holloway is right: the Russians may prefer an impasse. If so, let us be sure their preference is exposed. To ensure that result, a better offer must be made. It is an open secret that such compromise may meet passionate opposition from single-minded men in the Pentagon. The hardest test of the wisdom of George Shultz and the responsiveness of Ronald Reagan will come when they have to decide how and when to overrule such men.

A particularly obvious candidate for the discard heap, in any such compromise, is the Pershing II, a ballistic missile whose range Mr. Weinberger has publicly announced as 1;800 kilometers. The Soviets dislike the new Pershing, and their reasons are understandable. What we see as a weapon that balances their new threat to Europe they see as a new American threat to decapitate their deterrent, even to hit Moscow in minutes. We have here a splendid illustration of the way one side can cause the other side undesired fears. Our planners insist that this thing is not designed to reach Moscow—indeed that it is designed not to reach Moscow. But the Soviet leaders know as well as we do that it is only about 1,800 kilometers from Moscow to the closet point in West Germany—1,803 to be precise. If the situation were reversed, would any American planner find comfort in a three-kilometer margin of safety? We have here a case where what the planners have designed is just what the politicians need—to trade away.

Both books have much more in them than I have suggested here, and neither is written for the purpose of providing directly a prescription for present problems. But we too often suppose, in our hot arguments on today’s questions, that all nuclear problems are new. What Freedman and Holloway remind us is that the nuclear age has a substantial history, and that this history can be a good teacher. As Bernard Brodie wrote on the last page of Strategy in the Missile Age:

What we have done must convince us that Thucydides was right, that peace is better than war not only in being more agreeable but also in being very much more predictable. A plan and policy which offers a good promise of deterring war is therefore by orders of magnitude better in every way than one which depreciates the objective of deterrence in order to improve somewhat the chances of winning.

This Issue

March 17, 1983