A LearnWorld Text

Race to Oblivion

A Participant's View of the Arms Race

Herbert F. York




Table of Contents57Missile-Gap Mania125
INTRODUCTION78The McNamara Era147
Prologue: Eisenhower's Other Warning9PART TWO: UNBALANCING THE BALANCE OF TERROR171
1The Arms Race and I159MIRV: The Multiple Menace173
2The Race Begins: Nuclear Weapons and Overkill2711Other Lessons from the ABM Debate213
3The Bomber Bonanza4912The Ultimate Absurdity228
4The Elusive Nuclear Airplane60A Glossary of Acroyms241
5Rockets and Missiles75Index245




What are the basic forces behind ABM deployment and what do they really demand? A partial answer to this question may be found in the testimony of Donald Hornig, Lyndon Johnson's Special Assistant for Science and Technology and, in accord with custom, also Chairman of the President's Science Advisory Committee (PSAC). Hornig served in these capacities for five years, from January, 1964, through December, 1968, and before that he was one of the members of PSAC. He therefore had had the opportunity to watch the parade of the various versions of ABM from a grandstand seat for as long a time as anyone. Like all the other former Special Assistants to the President for Science and Technology, he opposed the deployment of the Safeguard system. He said:

If I were convinced that the protection of a credible deterrent were indeed the eventual goal and that Safeguard was the best way to protect that deterrent, I would support it. But the uneasy feeling persists that although Presidents may change, Secretaries of Defense may come and go, the philosophies enunciated by political leaders may change, the design of our ABM system hardly changes at all. It includes


the same radars, the same rockets, and largely the same deployment which was contemplated for the "heavy" defense. Safeguard continues to look like a first step toward a much bigger, more expensive and still ineffective system.

To many of the rest of us, too, the ABM appeared to have all the characteristics of a solution in search of a problem. As I have already remarked, this characteristic shows up all too often in our defense research and development programs, especially in the field of nuclear weaponry. I suggest the fundamental reason the ABM decision came up in 1969 was that ten years earlier Secretary of Defense McElroy, in dividing up the space and missile roles and missions among the three services, assigned the ABM to the Army as its only large sophisticated missile program. This decision created a situation in which for many years the lives and careers of many able persons have been closely entwined with the life and fate of one single program: the Army's ABM. This includes not only the civilians employed in the program office and by the main contractors, it also includes uniformed personnel and, probably just as importantly, a whole host of part-time advisers at all levels. If; in fact, we examine closely the testimony given by persons who are part-time advisers to the defense establishment and who were also in favor of the deployment of the present ABM system, we find that with only very few exceptions they favor Safeguard not as an end in itself, not for the purposes which the President laid down, but rather as a prototype of something else much bigger and much more complex and enormously more expensive. They want a grand system which they hope could protect not only the deterrent but also the rest of what goes to make up the United States of America. In short, they want to do a job which almost certainly cannot be done even if, by some magic,, it could be done as the "last move" in the arms race. Equally certainly, it


would result in a reaction by the Soviets which would more than offset even the theoretical capability of such a system, and, again equally certainly, it would cost vastly more money than the sums anyone is now talking about. Looking at the problem this way, we can see why it is that as a practical matter it is almost impossible for the United States to build a "thin" ABM system.

Let us turn now to a somewhat more general question: Can any ABM system be made to work? If one of the main weaknesses in Safeguard is that it is too thin and hence easily defeated by simple exhaustion, how about a thicker system? The answer to such questions has to depend in part on what "work,, means, and that definition is not always as simple as it might seem. In fact, the question "Will it work?" has three quite distinct meanings.

First, as discussed earlier, it may mean, Will it work on the test range? That is, will it regularly and reliably intercept the best mockups we know how to make of enemy warheads accompanied by enemy penetration aids and tactics? A favorable answer to this question is a necessary though not a sufficient condition for having any confidence in the system in question. And studying this question is an essential part of the design and development of any system. Experimentally verifying a favorable answer to this question would normally precede a decision to deploy, though under certain emergency conditions a high promise that such will be obtained might temporarily suffice. An ABM system that will work in this sense may well be possible.

Second, we may mean, Will it really work when the "balloon goes up"? That is, if a real enemy fires a real enemy warhead accompanied by real enemy penetration aids at a time of his choosing in an atmosphere of incredulity, will an operational unit manned by regular troops intercept it? Here we are dealing again with the difference between the test range and the


real world. It is, of course, always impossible to be absolutely certain of the answer to this type of "Will it work?" question until after the war is over, so for all new military systems we have to rely on theoretical predictions in dealing with it. However, such predictions are much more difficult to make about missile-defense systems than about offensive systems. The reason for this is to be found in certain intrinsic differences between offensive and defensive missiles. Once an offensive warhead is finally on its way, it need not "do,, anything until it contacts the ground or arrives at some predetermined height, at which time it has only to explode. A defensive system must, in a matter literally of only some seconds, puzzle its way through the deceptive devices and tactics of the total offensive payload and then explode its warhead at precisely the correct time and place, neither of which can be predetermined before the battle starts. This latter is intrinsically a more complex problem, and hence the subtle differences between the test range and the real world matter much more in the case of defensive missiles than in the case of offensive missiles. I have never heard of an ABM proposal which would lead me to have much confidence in a positive answer to this second type of "Will it work?" question.

The third meaning of the "Will it work?" question is largely political rather than technological in nature, and, as a result, it is by far the most important. A general way of phrasing this question is, Will it help its possessor to achieve some particular political objective? More specific ways of phrasing this third question might be, Will it contribute to deterring war? or even Will it make a preemptive attack more promising? The answers to these questions, unlike the answers to the first two, do not depend directly on what the technological facts really are, but rather on what the political and military (not the scientific) leadership on each side thinks they may be. Presumably, someone planning a preemptive attack would make a most serious attempt to estimate conservatively just how


much damage the retaliation that such an attack inevitably provoked would wreak on his own country. Clearly, a thick ABM system without serious soft spots such as Safeguard has could be just enough to make the difference in a go-no-go judgment by the decision-makers.

For example, if only the side considering making a surprise attack had such a system, and the system defended that side's population, it would make such an attack more promising. On the other hand, if only the side to be attacked had such a system, and if the system defended that side's retaliatory forces, then it would make a surprise attack look less promising. However, if each side possessed such a system, then these things would tend to balance each other off, and we would, as usual, be back more or less where we started, with neither of us having achieved any net gain.

During the 1969 ABM debates, several ways for making a more effective system were discussed. These usually involved both making the system thicker (i.e., composed of much larger amounts of all kinds of equipment) and removing the most serious soft spots. Some suggested that this last be done by duplicating or triplicating the large missile-site radars. Others suggested that the radar part of the system be completely redesigned with the objective of replacing the current large single radar with a large number of small cheap radars. I believe that a combination of such ideas could lead to a system that would be effective in the political or strategic sense. However, I also believe such a system would be very expensive, certainly costing more than fifty billion dollars, and it would make the difference only in the extremely unlikely circumstance that everything else looked favorable to a preemptive attack.

Thus, under certain fortunately very unlikely circumstances, possession of an ABM system could produce just enough false confidence to make a holocaust more likely. This is not something dreamed up late at night after a meal of pickles and apple pie: I have heard a number of high American general


officers and civilian officials, including Nelson Rockefeller, make remarks to the effect that defenses (including shelters) were needed in order to "stiffen the backbone of the American people." In other words, many people want such defenses because they want the possibility of nuclear war to be less unthinkable and hence less of a deterrent to other types of foreign political adventures. 1 am pleased to be able to report that I never heard a United States Secretary of Defense make such a remark, even though I was, for a number of years, often present in discussions of the type that would have provoked such a remark if it had been an expression of real belief.

Other interesting side issues came up during the ABM debates. One of the most important had to do with the proper role of the expert in making decisions of the kind then under discussion. Practically all of the nongovernment witnesses heard by the two principal Senate committees were experts at something relevant to the discussions usually, but not always, technology. These experts frequently disagreed flatly with each other. To some people, these disagreements among experts were something new. They expressed dismay, and plaintively asked if the experts cannot agree among themselves, how is a poor layman to decide?

Looking more deeply into the matter, we find that the experts not only disagreed on the answers, they also disagreed over what the important issues were. Further, the disagreements were not over technical facts, but rather over judgments and predictions: What will the Soviet technological response to our ABM or MIRV be? Will it "work" against a real attack seven or ten years from now? How will their (or our) assessment of that question influence their (or our) decision-makers? How will spending so much money this way affect other elements of our military power? How will it affect the domestic scene? These are all basically nontechnological questions. To be sure, technological facts, to a degree, are relevant to all


of them, in some cases very much so. But they do not have numerical answers and they generally transcend the kind of questions that do.

Much of the confusion and dispute that have arisen over this matter of the role of the expert derives from a failure to recognize the very fundamental difference between the kind of expertise and knowledge needed to design a weapons system and the entirely different kind of knowledge and wisdom needed to judge one. To take an easier case, consider the lunar landing If you want to know how to get to the moon, ask a rocket expert; if you want to know who should pilot the spacecraft, ask experts in space medicine and psychology; if you want to know what ought to be done after getting there, ask a geophysicist. But if you want to know whether someone should go there in the first place, ask any sensitive informed person. A politically literate housewife has at least as much business answering that last question as any of the experts mentioned, unless, perhaps, her husband makes his living in the space program.

The same reasoning applies to the ABM argument. If you want to know what a Soviet warhead looks like, ask the intelligence experts; if you want to know what radar wavelength to use to detect that warhead behind a cloud of ionized air produced by a previous nuclear burst, ask a physicist; if you want to know what the options are regarding the nuclear explosives to be used in its interception, ask the nuclear-weapons designers. If you want to know how a Soviet operations analyst would react to a given move, ask a U. S. operations analyst. But if you want to know whether or not to deploy Safeguard, there is no better place to go to than the usual political arena. Compared, for example, to the question of what to do about urban problems and racism, the question of what to do about the Safeguard ABM is certainly simpler and more straightforward and probably less important.

The same thing, of course, applies much more broadly: all


major weapons systems and defense policies should receive the kind of political and public scrutiny that the ABM received in 1969 and 1970. We must hope that the precedents established in those hearings will be used in other similar cases.

The man who claims, in a political argument, to have special understanding of some abstruse or arcane theory involved in weapons design and the man who claims to know a secret you don't know may indeed be privy to certain facts not generally available, but nearly always they are using such claims as a means to avoid arguing about the issues that really count. Such persons may not be doing so deliberately; most often, I think, they are themselves deceived and they simply do not understand the difference between designing something and judging something. They get caught up in an argument that says, "If it's feasible it must be useful; if it's useful and involves defense it must be essential; and since only I really understand how it works, my judgment is paramount." It is easy to be deceived by details and secrets. A man who spends a long time and a great deal of his personal psychic energy on solving a particular specialized problem can very easily believe that he has solved more than just that problem when he finally does succeed; and the man who knows a long list of secret details, every one of which may be iffy and uncertain, can easily lose sight of the bigger issues in which they are imbedded. As the political columnist Joseph Kraft said in discussing the Vietnam War, "Whom the gods would destroy in Vietnam, they first fill full of detailed information."

The arguments over the kind and degree of expertise needed in such matters was epitomized for me during the ABM hearings in the remarks of two quite different men: Donald Brennan, a mathematician and one-time president and still a member of the Hudson Institute (the "think tank" founded by Herman Kahn), and Senator Henry M. Jackson, a very able lawyer with long experience in defense matters. It happened that they both held that Safeguard should be deployed, but the


remarks I have in mind (which they did not make to each other or in reference to each other) bear sharply on this question of the "expert" role from opposite points of view.

Brennan, in a discussion of whether it was appropriate for the Senate to decide whether Safeguard should be deployed, said:

It is possible to conceive that the detailed scale and timing of expenditures for protecting offensive forces could properly come under Congressional scrutiny, though this would represent a new thing and I am not sure I should favor it. It would, in any event, have to be done in executive sessions, with sensitive information. It does not seem at all feasible to me to have such scrutiny of the decision as to the best means of providing the protection. The judgment about the best means should be based on a complex of factors that can scarcely be grasped whole by a full-time Secretary of Defense. That a committee of the Congress could meaningfully penetrate such a judgment seems to me most unlikely.

Subsequent testimony and later debates made it clear that Brennan had no doubts about his own ability to scrutinize such decisions. But in the course of the same hearings, and at the end of an unusually long day of testimony, Senator Jackson expressed the opposite point of view, as the following excerpts from the transcript of these hearings show.

SENATOR JACKSON: Mr. Chairman, I should just like to make a brief observation. In the fall of 1949--I was a Member of the House at the time--I was appointed by the late Senator Brien McMahon to the so-called Hydrogen Bomb Subcommittee. It was our task to make a recommendation whether this


country ought to go ahead with the development of the hydrogen bomb....

Senator Jackson then described how the subcommittee decided, against the advice of most experts, to go ahead with the H-bomb, and he continued:

I mention all of this, Mr. Chairman, to emphasize that we have to weigh the arguments . . . that we have to look at all points of view and in the end use our judgment. I do not think we laymen need be embarrassed in trying to weigh scientific judgments.* My own experience, which goes back a number of years in the House and Senate, has never left me with an inferiority complex in trying to deal with these problems that involve great scientific expertise . . . in 1949 every member of the blue-ribbon Science Advisory Committee of the Atomic Energy Commission, except Dr. Glenn Seaborg, who was not present, was opposed to going ahead with the hydrogen bomb, for the various reasons that have been alluded to here. That Advisory Committee was a Who's Who of science.

The brilliant physicist and weapons expert W. K. H. Panofsky, who was a witness at the hearings, responded at this point:

I just wanted to add I agree very much with Senator Jackson's remarks that decisions have to be made by the nonprofessionals, but they should be taken with the best information the professionals can provide. Also I am sure, as you have found out today, that the testimony which professional witnesses

* My italics. [H. York]


can give is limited because the future is involved and thus judgment and knowledge are both involved.

When I discussed the nuclear airplane (ANP) I mentioned a quite different kind of intervention by politicians in a technical program. Then the politicians persisted in concerning themselves with how to go about developing the power plant for a nuclear aircraft. In particular, they tried to insist on a particular sequence of developmental steps (all of which would be, to be sure, ultimately necessary). The result was a mess, and the nuclear airplane was never built. The present situation is very different. The Senate did not try to determine how an ABM system should be built; rather, the debate concerned whether it should be built. As I've already said, in my view there is no superior forum for a discussion of this kind of question.

What about shelters, another device often proposed for making nuclear war less "unthinkable?" Shelters are usually spoken of as being "passive defenses," in contrast to ABM systems, which constitute one kind of "active" defense.

Passive defenses suffer from the same fundamental failings as active defenses. Those which are inexpensive enough and unobtrusive enough to be acceptable will neither contribute to the credibility of our deterrent nor substantially reduce casualties if nuclear war occurs. Those which could reduce casualties (and conceivably contribute to the deterrent by so doing) are so extremely expensive and would so disrupt and change our whole way of life that they are completely unacceptable to most persons who have considered them seriously, including me.

In any speculation about the kind of attack to which this country might be exposed, it is useful to note where the military targets are located. Many of the missile bases are, in fact, far from the largest cities. Other key military installations,


however, are not so located. Boston, New York, Philadelphia, Seattle, San Francisco, Los Angeles (Long Beach), and San Diego all have important naval bases. Essential command and control centers are located in and near Denver, Omaha, and Washington, D. C. The roll call could be extended to include other major cities containing military installations that would have to be attacked in any major assault on this country. The list does not stop with these; it is only prudent to suppose still other cities would come under attack, because there is no way to know in advance what the strategy may be.

The only kind of shelter that has been seriously considered for other than certain key military installations is the fallout shelter. By definition, fallout shelters offer protection against nothing but fallout and provide virtually no protection against blast, fire storms and other direct effects such as would occur in all the cities named above and others like them. Some people have tried to calculate the percentage of the population that Would be saved by fallout shelters in the event of a massive attack, but since the form is unknowable the calculations are nonsensical. Even for the people who really would be protected by fallout shelters the big problem is not a problem in the physical theory of gamma-ray attenuation, which can be neatly computed, but rather the sociological problem of the sudden initiation of general chaos, which is not subject to numerical analysis.

So far, for these and other reasons, the government and the people of the U. S. have always rejected proposals for major fallout-shelter programs. But suppose, in spite of all this, the country were to take fallout shelters seriously and build them in every city and town. The people living in metropolitan areas that qualify as targets because they contain essential military installations and the people living in metropolitan areas that might be targeted as a matter of deliberate policy would soon recognize that relatively lightly constructed fallout shelters could not be adequate to protect them from direct blast effects. Then the same logic that led to the construction of


fallout shelters in the first place would lead inevitably to the more heavily constructed, much more expensive blast shelter.

After large numbers of blast shelters were built and evenly distributed throughout the metropolitan community, people would soon realize that shelters alone are not enough. Accidental alarms, even in tautly disciplined military installations, have shown that people do not always take early warnings seriously. Even if they did, a fifteen- to thirty-minute "early" warning provides less than enough time to seal the population into shelters. Accordingly, the logical next step is the live-in and work-in blast shelter, leading to still further disruption and distortion of civilization. There is no logical termination of the line of reasoning that starts with belief in the usefulness of fallout shelters; the logic of this attempt to solve the problem of national security leads to a diverging series of ever more grotesque measures. This is to say, in so many words, that if the arms race continues its current tempo and if we turn seriously to civil defense as a "solution," 1984 is more than just a date on the calendar fourteen years hence.

The preceding four chapters of this book have dealt with the deployment and development of strategic weapons in the period 1961-70. They could be summarized as follows:

1. During the first half of the sixties, the U. S. research and development program relating to strategic-weapons systems was largely one of continuation and consolidation of the programs initiated in the fifties. Thus, the program in this area was characterized by fewer false starts and less unnecessary duplication than formerly. (In some other areas, such as development related to tactical warfare, the opposite may have been true, but that is outside the scope of this book.)

2. The most important of the few new development programs that were initiated during the sixties was MIRV. It threatened to have a very destabilizing effect on the arms race, more so than was at first realized.

3. ABM development continued throughout the sixties. As


in the fifties, proposals to deploy an ABM system continued to be rejected, until late in 1967. At that time it was claimed that the possibility of a Chinese missile threat seemed to "marginally" justify a thin deployment, and this was authorized without much argument. In 1969, after a long public debate, deployment of a different system, the Safeguard ABM, was only narrowly authorized by the Senate. The principal objective of the Safeguard program was said to be protecting the deterrent and maintaining its credibility. Such additional protection is not now needed, and if it were, Safeguard would not provide it. Its deployment may well turn out to be the most costly and most futile of all our excesses in strategic arms deployment.

4. ABM and MIRV both epitomize the accelerating trend toward complexity and a concomitant need for a more highly automated, quicker response. The main dangers in this trend are three: it may result in placing too high a premium on going first; it probably increases the danger of accidental war; and it surely accelerates the trend in the seemingly inexorable transfer of authority over certain life-and-death decisions from statesmen and politicians to soldiers and technicians, from high levels to low levels, and from human beings to machines.

5. The growth of the stockpiles of nuclear weapons in terms of their total explosive yield pretty much stopped in the United States in the mid- sixties. It may still be continuing in the U.S.S.R., which has remained somewhat behind in this area.

6. The deployments of the various types of strategic missiles in terms of numbers were all finally determined in the United States in the early sixties. In the merciless glare of hindsight, I think they were all quite excessive; at least, they ran far ahead of the corresponding deployments in the Soviet Union. Soviet deployments of ICBMs did not catch up with ours until the very end of the sixties; Soviet deployments of submarine- launched missiles similar to our Polaris or Poseidon are still far behind ours in 1970, but could catch up in a few more


years if present building rates continue. Soviet intercontinental-bomber deployments are still lagging and show no sign of catching up. In more qualitative terms, the total explosive power deliverable by Soviet missiles is greater than the corresponding United States total, but U. S. bomber capabilities more than make up for this difference, and U. S. MIRV multiplicities are much larger than anything yet shown by the U.S.S.R.

Nuclear Designs: Great Britain, France, and China in the Global Governance of Nuclear Arms
[Transaction Publishers, 1996]

British SSBNs

French SNLEs
Other Sites
"Comprehensive Test Ban" [28 February 1996] and a 21 June 1996 addendum on China's CTB policy. The Acheson-Lilienthal Report [16 March 1946]: Report on the International Control of Atomic Energy. Re CTB

LearnWorld Explained FAQ Bosnia War Theory Templates

Our Home Page Our Home

Please direct comments and corrections by email to Bruce D. Larkin, who has prepared this page.