Race to Oblivion

PROLOGUE:
EISENHOWER'S OTHER WARNING

President Eisenhower's farewell address is justly famed for the twin warnings it left with the American people. The first of these concerned the possibility of "the acquisition of unwarranted influence, whether sought or unsought, by the military-industrial complex." Arms-control advocates sometimes use it as if it came from a sacred litany. Military-power advocates and arms developers feel compelled to explain it away by emphasizing other passages in the same speech. Scholars and politicians contend it is incomplete, and speak of the military-industrial-X complex, where X can stand for such other institutions as labor unions or universities or Congressional committees or combinations of them.

The other warning which directly followed the first--and which he placed on the same footing--is much less widely known, seldom quoted, and often poorly understood. After noting that research played an increasingly crucial role in our society and that the ways in which it was conducted had changed radically in recent years, Eisenhower said, "Yet in holding scientific research and discovery in respect, as we should, we must also be alert to the equal and opposite danger that public policy could itself become the captive of a scientific-technological elite."

Because this warning is the more subtle, it is the more diffi-

cult to interpret; for the same reason it is the more important. President Nixon and high defense officials have suggested that the opposition by some scientists to the deployment of ABM missiles was what President Eisenhower had in mind. The physicists' vote against the ABM, taken after the spring 1969 meeting of the American Physical Society was described as an example of an attempt by this elite to exercise illicit power. However, I think that what President Eisenhower really did have in mind is something very different from and much more significant than the type of thing suggested by these recent interpretations.

I worked fairly closely with Dwight D. Eisenhower during the last three years of his Presidency, first as a member of the Science Advisory Committee he set up immediately after Sputnik under the chairmanship of James R. Killian, Jr., and second as the first Director of Defense Research and Engineering, a new position created by the Defense Reorganization Act of 1958. In these jobs, I was directly concerned with precisely those scientific and technological programs in which the President himself was most involved, and I think, therefore, I have a good feeling for the context in which his thinking on the subject took place.

It also happened that after leaving the Presidency Eisenhower spent his winters in Palm Desert, California, a town less than one hundred miles from my home, and I called on him there on several occasions to pay my respects. Much of our conversation on those visits was devoted to the two warnings. He told me quite specifically that he had just two purposes in mind in making his farewell address. One was to say goodbye to the American people, whom he deeply loved. The other was to bring before the people precisely these warnings. The rest of the speech, he said, perhaps with some exaggeration, was there simply to fill it out and make the whole thing the appropriate length for a farewell address. I asked him to explain more fully what he meant by the warnings, but he declined to P>

do so, saying he didn't mean anything more detailed than what he had said at the time. I knew him well enough to understand what he meant: these warnings were not the result of a careful, methodical analysis; rather, they were the product of a remarkable intuition whose power has generally been underestimated.

What, then, was the context of these remarks? What annoyed and irritated him? Just whom are we to be wary of?

The context spanned the forty months from the launching of Sputnik to the end of his administration. The people who irritated him were the hard-sell technologists who tried to exploit Sputnik and the missile-gap psychosis it engendered. We were to be wary of accepting their claims, believing their analyses, and buying their wares.

The hard-sell technologists and their sycophants invented the term "missile gap," and they embellished that simple phrase with ornate horror stones about imminent threats to our very existence as a nation. They then promptly offered a thousand and one technical delights for remedying the situation. Most were expensive, most were complicated and baroque, and most were loaded more with engineering virtuosity than with good sense. Anyone who did not immediately agree with their assessments of the situation and who failed to recognize the necessity of proceeding forthwith on the development and production of their solutions was said to be unable to understand the situation, technically backward, and trying to put the budget ahead of survival.

The claims of such people that they could solve the problem if only someone would unleash them carried a lot of weight with the public and with some segments of the Congress and the press. Other scientists and technologists had performed seeming miracles in the recent past, and it was not unnatural to suppose that they could do it again. It seemed that radar had saved Britain, that the A-bomb had ended the war, and that the H-bomb had come along in the nick of time to save

us from the Russian A-bomb. On the home front, antibiotics had saved our children from the scourges of earlier times, machines had uplifted us from drudgery, airplanes and electronics could carry us and our words great distances in short times. Scientists and technologists had acquired the reputation of being magicians who were privy to some special source of information and wisdom out of reach of the rest of mankind. A large part of the public was therefore more than ready to accept the hard-sell technologist's view of the world and to urge that the government support him in the manner to which he wanted to become accustomed. It seemed as if the pursuit of expensive and complicated technology as an end in itself might very well become an accepted part of America's way of life.

But it was not only the general public that believed the technologists understood something the rest of the world could not. Many of the technologists themselves believed that only they understood the problem. As a consequence, many of them believed it was their patriotic duty to save the rest of us whether or not we wanted them to. They looked at what the Soviets had done. They used their own narrow way of viewing things to figure out what the Russians ought to have done next. They decided then that since the Russians were rational (about these things anyway), what they ought to have done next was what they must now be doing, and they then determined to save us from the consequences of this next Russian technological threat. The Eisenhower Administration was able to deal successfully and sensibly with most of the resulting rush of wild ideas, phony intelligence, and hard sell. But some of these ideas did get through, at least for a while. Beyond that, dealing with self-righteous extremists who have all the answers--and there were many among the aerospace scientists and technologists at the time--is always annoying and irritating.

As we now know, the Byzantine technological ideas urged on us in those years were in fact a portent of things to come.

Weapons systems have become still more complex in the years since Eisenhower's farewell address. And this complexity is creating new and serious problems of the general kind that Eisenhower warned us about. Worse, this complexity involves not only the inner workings of any device but its use and application as well. By 1969, on the problem of what to do about defending our land-based missiles, one expert felt obliged to say, "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 be most unlikely." I do not believe that this statement was true insofar as the judgment concerning the specific question at issue was concerned, but Donald Brennan, who made the statement, is a long-time serious student of military technology and we must therefore take it as at least a warning of what may become true. And when such a statement can be correctly made about a large issue like protecting land-based missiles, then indeed "public policy" will have been, as Eisenhower warned, "captured by a scientific-technological elite." To allow this to happen would be to accept an absurdity: the transfer of control over our destinies from ourselves and the statesmen and politicians we select into the eager hands of strategic analysts, technologists, and other experts.

We will see in this book how the over-all complexity of systems is already leading us to a situation in which the response to a hypothetical future attack will be so complicated and the time in which to decide what to do will be so short that it will be necessary to turn to automatic computing machines for the purpose. If we continue with the present style of technological approach to defense problems, the inclusion of human beings in the decision-making loop will seriously degrade the performance of the system. Thus, here too the power to make life-and-death decisions is passing from the hands of statesmen and politicians to lower-level officers and ultimately to com-

puting machines and the technicians who program them. This trend too, if allowed to continue, will result in the capture of public policy by a scientific-technological elite. Eisenhower's warnings, while based largely on intuition, have pointed up a very real and extremely serious problem.

THE ARMS RACE AND I

Germany invaded Poland and started World War II just two weeks before I entered the University of Rochester in September, 1939. Ever since then, my professional life has been completely dominated by the nuclear-arms race.

During my sophomore and junior years my physics professors, including Lee DuBridge and Victor Weisskopf, began to disappear one by one into secret war laboratories. After my own accelerated graduation in May, 1943, I too left for the University of California Radiation Laboratory in Berkeley, where we were engaged in the separation of uranium isotopes. The director of the laboratory was Ernest O. Lawrence, and my immediate boss was Frank Oppenheimer. Frank's older brother, J. Robert Oppenheimer, was at that time establishing a laboratory at Los Alamos, New Mexico, where the isotopes we were separating would, two years later, be fabricated into history's first atomic weapon.

After spending the rest of 1943 at the laboratory in Berkeley, I went with a number of others to Oak Ridge, Tennessee, where we assisted in starting up the great Y-12 manufacturing plant, whose processing machinery was a simple direct multiplication of the isotope separation devices we had been working on in California. Finally, on August 6, 1945, the uranium isotopes we had separated with a late assist from the K-25

plant also located at Oak Ridge were detonated over Hiroshima, in history's first military use of nuclear explosives. On August 9, another A-bomb, made from a different nuclear explosive (plutonium), produced by an entirely different process, was exploded over Nagasaki.

The very next day the Emperor of Japan announced to the world his desire to surrender. We had every reason) then to believe that our bombs had caused him to do so.

A few weeks after, my father, who was a Railway Express messenger, came down to Oak Ridge on his pass to visit me. As we sat on a hill overlooking the great Oak Ridge isotope separation plant, I explained to him how we had finally succeeded in making war so terrible that it could not happen again. I really believed it. I was just twenty- three and I did not yet know that the same thing had been said many times before. I think I would have believed it anyway, if only because this latest step up in the horror of war was such a huge one. It was only much later, after still further enormous increases in the potential destructiveness of war, that I came to realize that we had made perhaps no change at all in the probability of war.

I returned to Berkeley in 1945 as a student and embarked on a career in pure science, or so I thought at the time. Four years later, in June, 1949, I received my Ph.D. Two months after that the Russians exploded their first atomic bomb. This explosion, which happened much sooner than most experts had believed possible, was followed only ten months later, in June, 1950, by the sudden outbreak of the Korean War. As with many other Americans, my life was radically changed again by these two events and by my view of them. Stalin was very much alive, the Sino-Soviet alliance had just been forged, and the future looked gloomy once again.

Edward Teller began to promote the idea that the United States, in response to the new situation, should initiate a top priority program to develop a hydrogen bomb. He, with others at the Los Alamos Scientific Laboratory, designed an

experimental device which they believed could establish the conditions necessary to bring about the nuclear ignition of the appropriate forms or isotopes of hydrogen. Hugh Bradner and I joined a number of others in establishing at Berkeley a group that later set up at Eniwetok an experimental apparatus to study the environmental conditions within which this hydrogen reaction was supposed to take place. In May, 1951, in operation Greenhouse at Eniwetok, this hydrogen burn, or fusion, did come about more- or less as expected and our experiment helped to confirm the possibility of turning it into a weapon.

Shortly after, I was appointed an assistant professor of physics at the University of California, Berkeley, and the next fall I began once again, I thought, a research and teaching career. I was just getting settled into that job when Ernest Lawrence asked me to take the lead in establishing a branch of the Radiation Laboratory to undertake research and development in the fields of nuclear explosives and controlled fusion reactions. Edward Teller, Harold Brown, John S. Foster, Jr., and I, along with two hundred others altogether, opened shop in September, 1952, in a grape growing valley at Livermore, California. As was his own special style, Lawrence at first simply told me to "run the place." Later, after we had been going for a little more than a year, he told me to "start calling yourself the director," and I did. Our first attempts to design nuclear explosives worked out poorly. The first two shots at the Nevada test site in the spring of 1953 gave less than the expected explosive yields. In the following year, at Bikini, our first attempt produced an outright fizzle, and we then canceled what would have been our second shot in that test series after painfully deducing that it would turn out the same way. Even so, we all kept trying, and eventually the Livermore Laboratory, now known as the Lawrence Radiation Laboratory/Livermore, succeeded in developing its share of America's nuclear weapons.

Toward the end of my tenure at Livermore, we began to get

involved in a small way with the group President Eisenhower had set up under Harold Stassen for the purpose of trying to find and promote steps leading to arms control and disarmament. It was at that time that I first came to realize that there was another side to the arms race and that there might really be some other more promising and less dangerous road to national security. I came to this realization later than some older and more perceptive scientists, but, even so, earlier than many others.

As director of the Livermore Laboratory, I was invited in l 954 by John von Neumann to join a special committee which he chaired. This committee had as its purposes, first, reviewing various proposals for the development of huge rockets capable of delivering nuclear warheads with high accuracy a quarter of the way around the world, and, second, advising the Air Force (and later the Secretary of Defense) as to what should be done to overcome the Soviets' earlier start in this v new field of weaponry. (By 1953 our intelligence confirmed that the Soviet Union, building on the earlier programs of the Germans in World War II, was engaged in a major effort to develop long-range rockets capable of delivering nuclear warheads.)

The Von Neumann Committee was remarkable in style as well as output. In its earliest meetings, the committee worked directly and intimately with Assistant Secretary of Defense Donald Quarles, Air Force Brigadier General Bernard A. Schriever, and Trevor Gardner, Special Assistant to the Secretary of the Air Force for Research and Development. It was instrumental in establishing the Ramo-Wooldridge Corporation in its role of what is now called "general systems engineering and technical direction" (GSETD) of all Air Force long-range-missile programs. It also established the basic design parameters for all the large rockets that the United States developed in the ensuing years. Other members of this remarkable committee included Charles A. Lindbergh and Clark B.

Millikan--who became chairman after the untimely death of Von Neumann as well as George B. Kistiakowsky and Jerome B. Wiesner, both of whom later became Special Assistants for Science and Technology to Presidents of the United States.

On October 4, 1957, the Soviets launched Sputnik, the first artificial earth satellite, and my professional life underwent another sudden and drastic change. In response to this and to other Soviet space achievements that quickly followed, President Eisenhower appointed James R. Killian, Jr., then president of the Massachusetts Institute of Technology, to be his Special Assistant for Science and Technology and to establish and chair the President's Science Advisory Committee. Killian and I had met previously in connection with the work of the Gaither panel, and I was most fortunate to be among those he invited to join the committee. I took a temporary leave (I thought) from my post as director of the Livermore Laboratory, and began in December, 1957, to spend full time working with the committee (called PSAC, pronounced pea- sack) in the White House annex. I thus had a front seat from which to watch our efforts to "do something" about the Soviets' "firsts," and I was able to participate in the making of many of the decisions involved. PSAC reaffirmed some of the old priorities and established some new ones in the United States' space and missile programs. PSAC encouraged and then endorsed the efforts of the Department of Defense to get its house in order in this area, and was the most important factor in transforming NACA (the National Advisory Committee for Aeronautics, an organization conducting research in aeronautics) into NASA (the National Aeronautics and Space Administration, the organization responsible for all the non-military aspects of the United States' space program). But, in addition, PSAC even then found time to devote much of it. attention to the problem of containing the arms race. In general PSAC developed ideas of how best to organize for arm' control; in particular, it did the lion's share of the work in-

volved in promoting the nuclear-test moratorium of 1958. I did not have a major role in that part of PSAC's work at the time, but I followed it closely and was very much impressed by it. What I learned there greatly influenced my own direct activities in arms control and disarmament, which began somewhat later.

I never did go back to Livermore. In the spring of 1958 I was invited by Secretary of Defense Neil H. McElroy to become the chief scientist of the Advanced Research Projects Agency. ARPA, as the agency is usually called, was one of the Department of Defense's responses to the shock of the series of Soviet firsts in space. Each of the three military services was determined to take advantage of the frantic Congressional and public reaction to the spectacular Soviet successes, and each began to lay its own claims to the various new roles and missions it foresaw in this new environment.

The Navy was already the agent for the Vanguard program that was to have launched America's first satellite; the Air Force then already had the responsibility and the money for the development of the largest of the long- range rockets; the Army had Wernher von Braun. ARPA, therefore, was established in large part to try to bring this very confusing and politically explosive situation under control. Roy W. Johnson, a General Electric executive, was named Director of the new agency, Rear Admiral John Clark was named its Deputy Director, and 1 became its Chief Scientist. We opened for business in late February, 1958, and we did help to produce some order in the space and missile efforts of the Department of Defense.

In December, 1958, I was appointed the first Director of Defense Research and Engineering by President Eisenhower, on nomination by Secretary McElroy and with the advice and consent of the Senate. As set out in the Defense Reorganization Act of 1958 (another American response to Sputnik), the DDRE (as the incumbent of the position is labeled) had

authority over all research, development, tests, and evaluation programs of the Department of Defense and its component parts. In addition, he had the responsibility to advise the Secretary of Defense with regard to certain other matters, including the deployment of advanced types of weapons systems. I continued to serve as the DDRE under Secretary Thomas S. Gates, and briefly also under Secretary Robert S. McNamara after being reappointed by President Kennedy. It was an exceptionally interesting time to be in a job of that sort; the general form of nearly all of our present missile and space programs was jelled at that time. It was the time when the Von Braun group was transferred from the Army to NASA and when the responsibility for the very large rockets was also transferred from ARPA to NASA. It was also the time of the first rejection of the proposal to deploy an ABM (antiballistic-missile) system, and it was the time of the U-2. It was during my tenure in that job as scientific administrator of the United States side of the arms race that I formed the views I still hold about the futility of the race and the absolute need to find some alternative course.

In January, 1961, I had the opportunity to discuss these matters with John J. McCloy, who was President-elect Kennedy's personal and principal adviser on matters of arms control and disarmament. I communicated to Mr. McCloy the substance of what I stated publicly before the Senate Foreign Relations Committee in 1963:

Ever since shortly after World War II, the military power of the United States has been steadily increasing; over the same period the national security of the United States has been rapidly and inexorably diminishing.... It is my view that the problem posed to both sides by this dilemma of steadily increasing military power and steadily decreasing national security has no technical solution. If we continue to

look for solutions in the area of military science and technology only, the result will be a steady and inexorable worsening of this situation. I am optimistic that there is a solution to this dilemma; I am pessimistic only insofar as I believe there is absolutely no solution to;be found within the areas of science and technology.

After the inauguration of President Kennedy, McCloy became responsible for developing and promoting the legislation which led to establishment of the United States Arms Control and Disarmament Agency. When the agency was formed, he became Chairman of its General Advisory Committee, and I became one of the committee members.

Some months after Robert S. McNamara became Secretary of Defense I resigned as DDRE so that I could return, as I had long been planning, to the University of California. At his request, I gave Secretary McNamara a list of thirteen names of persons who could, I believed, easily replace me. (I consulted Killian, Kistiakowsky, Wiesner, Glenn T. Seaborg and many others in making up my list.) The name McNamara picked was that of my good friend Harold Brown, who at the time was director of the Livermore Laboratory, as I had been earlier. When Brown later became Secretary of the Air Force, he was in turn succeeded as DDRE by another mutual friend, John S. Foster, Jr., who had followed Brown as director of the Livermore Laboratory.

After returning to the University of California, first as chancellor of the San Diego campus and later as professor of physics, I kept posted on the arms race and the efforts to bring it under control, through continuing service as a member of the General Advisory Committee of the Arms Control and Disarmament Agency (from 1963 to 1969), as a member of the President's Science Advisory Committee for a second term (1964-68), and also as a trustee of two nonprofit corporations: the Aerospace Corporation (the organization now

charged with doing general systems engineering and technical direction for most of the Air Force's space and missile programs) and the Institute for Defense Analyses (which, among other things, performs analyses and makes studies of weapons systems for the Joint Chiefs of Staff, the Secretary of Defense, and other high government officials).

After finishing my second term on PSAC, I began my fourth attempt to establish a serious career in basic science, this time by teaching physics and doing a modest amount of laboratory work. Fifteen months later, in the spring of 1969, when the matter of the deployment of the ABM suddenly broke from behind the curtain of secrecy to become a public issue, I welcomed the opportunity to testify before the Congress on three occasions in opposition to the deployment of what President Nixon called the Safeguard system.

By no means am I the only Department of Defense official who has come to realize the dilemma of an ever-increasing military power accompanied by an ever-decreasing national security Nor am I the only defense official to realize that the dilemma cannot be resolved by the development and deployment of ever more complex and more costly machines. Harold Brown said after serving more than four years as DDRE and nearly four years as Secretary of the Air Force:

Those who have served as civilian officials in the Department of Defense at the level of Presidential appointment . . . have recognized the severely limited utility of military power, and the great risks in its use, as well as the sad necessity of its possession . . . [The] higher their position and, hence, their responsibility, the more they have come to the conclusion that we must seek national security through other than strictly military means . . . and urgently.

In the process of participating in the arms race itself as well as in some of the attempts to bring it under control, I have

PART ONE: TOWARD A BALANCE OF TERROR

had the opportunity to watch most of the key men and events from a front-row-center seat. In Part I of this book I will outline the main elements of the arms race, and I will describe in some detail a number of specific instances where I was personally involved. Through them, I shall try to expose some of the principal factors that have driven it forward. I hope to make it clear why I believe that excessive prudence and technological hard-sell have led to an unnecessary overreaction in case after case. And I hope to convince the reader that while these overreactions have produced so far a fairly stable balance of terror between the superpowers, this balance is at a very much higher level of armaments and peril to mankind than would have been the case had a more reasonable approach been followed.

In Part II, I will describe some important current technological trends and programs, including ABM and MIRV'. I will try to show why I believe that these threaten to unbalance the balance of terror and that they hold no promise whatever as a means for achieving national security. And lastly, I will share with the reader the reasons for my growing conviction that the end result of it all has been to produce a situation which is at once wondrously absurd and exceedingly dangerous, and which no one,, neither the most sanguine weapons fancier nor the most cerebral strategic analyst, ever intended.

PART ONE

Toward a Balance of Terror

THE RACE BEGINS: NUCLEAR WEAPONS AND OVERKILL

The first atomic bomb was dropped on the Japanese city of Hiroshima on August 6, 1945. At least 66,000 people died immediately as a direct result of the explosion and the fire storm that followed. Tens of thousands more died in the aftermath. About eighty percent of the homes and buildings in Hiroshima were destroyed, and most of the rest were damaged. The bomb weighed nine thousand pounds and was ten feet long and twenty-eight inches in diameter. The explosive it contained was uranium metal highly enriched in the rare isotope U 235. The physical processes of the nuclear explosion took place in less than one one-millionth of a second, and the amount of energy released is estimated to have been about equal to that which would be released in the explosion of fourteen thousand tons of TNT. The bomb was delivered by a B-29 aircraft from the island of Tinian in the Marianas, about fifteen hundred miles from the target. It was exploded several thousands of feet above the ground in order to cover the greatest possible area with high-pressure blast waves and intense heat radiation.

The second atomic bomb was dropped on the Japanese city of Nagasaki on August 9, 1945. At least forty thousand people died immediately from the explosion and the fire storm; many thousands more died in the aftermath. About forty

percent of the city's structures were destroyed or very severely damaged. The devastation at Nagasaki was less than that at Hiroshima because of the hilly nature of the terrain and the smaller size of the city. This bomb weighed ten thousand pounds; it was ten feet, eight inches long and five feet in diameter. It used a different explosive, the new, chemical element number 94, plutonium. The explosive equivalent was twenty thousand tons of TNT, or, as this is more commonly described, twenty kilotons (1 kiloton = 1,000 tons).

The Emperor of Japan announced his desire to surrender the next day, August 10, 1945. Soon after, the planned invasion of the Japanese home islands later that same year by a force of millions of men was canceled. It seemed perfectly clear at the time that the two A-bombs had directly resulted in ending the war in the Pacific and that their use had saved, in the net, a great number of lives. The explosion of these bombs was widely celebrated because of this.

Subsequent reviews of what was really going on at that time in Japanese ruling circles have indicated that the situation was more complex than it then seemed. We now know that a peace party, centered around the Emperor himself, had been growing and that the first tentative attempts of this party to end the war predated by many months the detonation of history's first nuclear weapon. The general course of the war in the South Pacific plus the ever-increasing high-explosive-bombing and fire-bombing of Japanese cities in 1945 had steadily strengthened the hand of the peace party. It does still seem most likely that the two atomic bombs were of major importance in finally pushing the whole process over the hump, but this assumption cannot be proved.

These two atomic bombs, plus the test device detonated in the first atomic explosion at Alamogordo, New Mexico, on July 16, 1945, were the product of the Manhattan Project. This project, unique in purpose, style and magnitude, was a child of World War II and of the political and scientific history of the years immediately preceding that conflict.

In 1938, Otto Hahn and Fritz Strassmann, working in Germany and following up on earlier work by Enrico Fermi and his co-workers in Italy, discovered the process known as nuclear fission. When struck by neutrons, the nuclei of certain heavy atoms split roughly in two and, it was later found, released enormous amounts of energy and a few new neutrons in the process. This experimental discovery was soon followed by theoretical explanations, first by Lise Meitner, a German refugee then living and working in Scandinavia, and later by Niels Bohr (a Dane then visiting Princeton) and John A. Wheeler. Enrico Fermi, Leo Szilard (both European refugees then in the United States) and others suggested and soon after confirmed that the fission event took place in such a way as to lead to further fissions under certain special conditions. It was immediately realized that a chain reaction was probably possible and would lead either to enormous explosions or to almost unlimited power production. Szilard, jointly with Eugene Wigner, and later assisted by Edward Teller (all European refugees) then persuaded Albert Einstein (still another) to write a letter to President Franklin D. Roosevelt, bringing all of these possibilities to his attention.

This Einstein letter, dated August 2, 1939, resulted eventually in the creation of the Manhattan Project and in the initiation of the nuclear-arms race. To all of these refugees, and to virtually all other Americans, certainly including me, the polities of the situation were stark and simple: Western civilization was endangered by the twin menaces of Nazism and Fascism--these two unmitigated evils would do anything they could to win and destroy us. Anything that would prevent them from winning had to be good. Winston Churchill said in those days that to stop Hitler he would make an alliance with the Devil.

The Manhattan Project eventually cost two billion dollars, most of it spent in the last two years of the war. In terms of rate and scale and compared with the available national resources then, it was and still remains the largest scientific project of all

time, not excepting the Apollo program to land a man on the moon. It was initiated, directed and managed very largely by physical scientists, although the over-all director was a military man, General Leslie Groves, and the production plants were eventually run by professional industrial managers.

The over-all project consisted of a number of subprojects, each directed by physical scientists. J. Robert Oppenheimer directed the Los Alamos Scientific Laboratory in a then secret location on a mesa in New Mexico. There the bombs themselves were designed and fabricated. Ernest O. Lawrence directed the University of California Radiation Laboratory, now named after him. There the electromagnetic process for separating uranium isotopes was perfected; the "red lab" served as the pilot plant for the huge Y-12 plant at Oak Ridge where the bulk of the uranium exploded over Hiroshima was processed. Harold C. Urey and John Dunning directed the Columbia University project that led to the huge K-25 gaseous-diffusion plant at Oak Ridge for separating uranium isotopes. This plant processed a part of the uranium that went into the Hiroshima bomb, and soon after the war it completely replaced the Y-12 plant and method as a means for producing U 235. Arthur H. Compton, Fermi, Wigner and others directed the work, first at the Metallurgical Laboratory of the University of Chicago and later at the X-l0 Laboratory in Oak Ridge that laid the basis for the design and construction of the large production reactors at Hanford, Washington. Ultimately these reactors produced the plutonium for the Nagasaki bomb as well as for the Alamogordo test explosion. Philip H. Abelson directed work on a thermal-diffusion process for separating uranium isotopes. Although it was started quite late, that work too made an important contribution to the total effort. Nearly all of these men, plus a few other scientific leaders, were involved in (though not in control of) the discussions of what to do with the Bomb after it was ultimately produced.

The Manhattan Project showed what American industry

and American science could do when fueled by money, patriotism and fear. It has been held up ever since as a model, very often in situations where it is totally irrelevant. But there are other lessons, too.

At the time, all of us thought we were in a serious and deadly race with the Germans for the atomic bomb. Einstein's August 2, 1939, letter to Roosevelt implied this, and a follow-up letter dated March 7, 1940, said, ". . . interest in uranium has intensified in Germany . . . research there is carried out in great secrecy and ... has been extended...." The postwar survey of the situation described in Samuel Goudsmit's book Alsos showed that we were not in feet in any such race. For a variety of reasons, including a mistaken belief on the part of the Nazis that the war would be very short, lack of interest on the part of many German scientists, and resource allocation problems and internal jealousies, the Germans never mounted a serious effort parallel to the Manhattan Project. It was, however, entirely plausible for us to think that they had. After all, fission had been discovered in Germany, and, despite the fact that many physical scientists had fled from the Nazi regime, it seemed that a sufficient number of competent scientists and engineers remained to initiate and carry out such a project in Europe's most technologically advanced country. Furthermore, the bulk of our intelligence effort was concentrated on classical targets; our technical intelligence was not yet sufficiently extensive and reliable to convince us, by its lack of any real reports about a major program, that such a program did not exist,

The Manhattan Project, then, was based on the first of a long series of mistaken beliefs in our being "raced." Because this was the first such case, our mistake in this regard was, in my view, entirely justified, but our subsequent failure to learn anything from repeatedly making this same mistaken judgment about the existence of a "race" is less so. Time after time since then our leaders have become convinced that the

enemy, or the competition, or whatever one chooses to call it, was up to some technological mischief designed to do us in or at least to cause us very serious trouble. And time after time these convictions proved to be either completely untrue or at least grossly exaggerated. True, the detailed reasons why the Germans did not conduct a Manhattan Project of their own were unusual circumstances applying only to that specific situation, but unusual circumstances are like unusual weather: any specific variety of unusual weather may be rare, but having some kind of unusual weather is usual.

After the war, the great scientific teams broke up. Some men went back to where they had been recruited from. Many others went to new positions. Nearly all thought they were leaving the war and war research behind them. Meanwhile the production plants continued to produce U 235 and plutonium and to fabricate them into bombs, and the Los Alamos Scientific Laboratory, even though most of its wartime leadership had left, continued to concern itself with the science and technology of nuclear weapons under its new director, Norris E. Bradbury. Up until August, 1949, when the U.S.S.R. tested its first A-bomb, the U. S. enjoyed a complete monopoly on nuclear weapons. In the interim between the end of the war and that date, the U. S. conducted two series of atomic tests. One, called Operation Crossroads, was conducted at Bikini Atoll mainly to determine the effects of nuclear weapons on naval vessels, as well as the general effects of underwater nuclear explosives. The second, called Operation Sandstone, was conducted at Eniwetok Atoll, primarily to test evolutionary improvements in weapons design. The technical situation in those days, as described by Edward Teller in The Legacy of Hiroshima, was frustrating for the scientists involved. Many of them wanted to get on with new and different designs (as a minimum, different in terms of weight, explosive yield and general dimensions), but the military services were having difficulty imagining what they would do with anything different

from the Thin Man or the Fat Boy (as the first two bombs were called) that had been designed specifically to fit the B-29.

The stockpiles of fissionable materials and atomic bombs continued to grow. We may estimate that the United States had some hundreds of bombs by about 1950 and that each of these could result in a release of explosive energy equivalent to a few tens of kilotons, so that the total explosive yield of the entire stockpile would have been some few millions of tons of TNT. Ralph Lapp estimates that the total at the time was ten megatons ten million tons. By comparison, this amount of explosive is considerably larger than the total used in all of World War II. In particular it compares strikingly with the estimated 2,700,000 tons of explosives that were dropped from the air on Germany. Thus, by 1950 the nuclear-arms race had reached a point such that we could duplicate the destruction of World War II by using nuclear weapons, except that while that conflict had lasted for more than five years, the devastation could now be reproduced in a single day.

Some authors have suggested that the effects of atomic bombs and the viciousness of atomic war have been exaggerated One of the things they have in mind is that if we compare the strictly physical effects of two bombs, one of them a thousand times larger than the other (say a ten- kiloton atomic bomb and a ten-ton chemical blockbuster), then the area subject to some given blast pressure is only one hundred times as great for the larger bomb. Hence, they say, on a per-kiloton basis, atomic bombs are ten times less effective than the largest chemical bombs. The physics of these notions is correct as stated, but this does not completely determine what happens when these weapons are used against human beings and the structures men build. Partly because of complex reinforcing effects, the devastation within the area of effectiveness of an atomic bomb is much more nearly complete than in the case of an attack with chemical bombs. The net result is that A-bombs and chemical bombs are roughly equiv-

alent to each other on a per-kiloton basis in their ability to destroy life. The great air raids on Hamburg in 1943, which were carried out entirely with chemical explosives, destroyed about fifty percent of the housing but killed only about three percent of the people. Where atomic weapons were used, the percentage of the area of the cities destroyed and the percentage of the populations killed were more nearly equal. Therefore, we may say that by 1950 we had a stockpile capable of somewhat more than reproducing World War II in a single day.

We might have gone on for years slowly accumulating A-bombs and slowly modifying them in an evolutionary way to fit new types of military equipment, if it had not been for two closely spaced events which caught us completely by surprise and which raised new anxieties and generated new possibilities. These were the first Russian A-bomb test, in August, 1949, and the Korean War, which broke out suddenly in June; 1950. The Russians exploded their atomic bomb much earlier than most American experts had predicted. Up to that time, the Russians had not particularly impressed us by either their scientific or their technological capability, and many experts expected that they would require somewhere between a minimum of five and as much as twenty years before they could duplicate the atomic bomb. Of course, since then Russian accomplishments in missiles and space and in science quite generally have completely changed our estimates of Soviet scientific and technological capability, but we must recall that at the time, in the late 1940s, the general view of the Russians was quite different from what it is today.

We still do not know how much the relatively short gap between the first American and the first Russian explosion was due to successes in Soviet espionage, and how much it was due simply to native Russian capability. At the time it seemed to many of us that espionage must have been far and away the main reason they were able to accomplish the job so

quickly, especially after the devastation that had been wreaked on them by World War II. Having since seen some excellent Russian technological progress in other fields, we are no longer quite so sure that this was the case. We should note, furthermore, that it is always easier to do something a second time, even if the only thing known from the first time is that it can be done.

In order to understand the reaction of many Americans to this new Soviet accomplishment and to the breakout of the Korean War, it is necessary to recall that Stalin was still very much alive and that Stalinism was entering one of its more paranoid phases. The Sino-Soviet bloc was newly forged and seemed to most of us in the West (as well as to the leadership of the Soviet Union itself) to be a monolith of tremendous potential power. The surprise and the fear and anxiety generated in this country by these events brought about a number of changes in American life and politics. Of most interest here was the power struggle that developed within the American scientific community over the issue of whether or not to initiate a priority program to develop the H-bomb, or, as it is more properly called, the thermonuclear weapon.

Instead of deriving its energy from nuclear fission, like an A-bomb, an H-bomb derives its energy from a process called thermonuclear fusion. Thermonuclear reactions had been hypothesized by Hans Bethe in the thirties as being the energy source that powers the sun and the stars. For our purposes here we may say that they involve the transformation of various isotopes of hydrogen, into isotopes of helium. Sometimes another light element, lithium, plays a role which may be crudely likened in some ways to that of a chemical catalyst. These thermonuclear (or fusion or hydrogen or H-) weapons differ in three very important ways from the earlier fission or A-weapons:

First, the energy released per pound of reactants consumed in thermonuclear explosions is between three and ten times as

great as the energy released per pound of reactants consumed in fission explosions.

Second, the materials needed to construct thermonuclear weapons are much cheaper than the materials needed to construct fission weapons. (There is only one exception, tritium, and it is not important just here.)

And third, thermonuclear weapons, unlike fission weapons, may be of practically limitless size and power. The first hydrogen device ever exploded, the Mike Event on November 1, 1952, was a thousand times more powerful than the first atomic bomb, which had been exploded only a little more than seven years earlier. It is now practical to make hydrogen bombs having explosive yields ranging anywhere from ten times the power of a nominal atomic bomb up to ten thousand or more times that amount. The cost of such bombs is, to a first approximation, practically independent of their explosive yield.

The American scientific community, or rather that part of it which was informed about the nuclear-weapons programs of the late forties, was split deeply over the issue of whether or not to go ahead with the development of the H-bomb on a high-priority basis. The numerically smaller side in the split, led by Edward Teller, felt that such a program was the appropriate U. S. reply to the Russian A-bomb and the political realities most recently illustrated by the Korean War. This Teller party admitted that there was as yet no specific promising idea about how to build an H- bomb, but they contended that a project equivalent to the earlier Manhattan Project would inevitably soon generate one. This group also argued that since the Russians had shown they could produce an A-bomb much faster than anticipated we must expect that they could do the same thing with the still hypothetical H-bomb, and that if they did so first the U. S. would suddenly find itself in serious jeopardy.

The other side, led by J. Robert Oppenheimer, opposed the

initiation of a high-priority program on technical, political and moral grounds. They emphasized that none of the proposed general methods for building an H-bomb was at all promising, and they suggested that even if one could be built it would probably involve so much tritium that it would be quite impractical economically, even if technically feasible. (Tritium is the most reactive of the hydrogen isotopes and the only one that is exceedingly expensive. As it finally turned out, large amounts of tritium are not needed.) Further, and partly because there appeared to be no limit to how big an H-bomb could be if one could be built at all, this group felt that the successful development of an H-bomb would lead to a new and very much graver level of danger not only to civilization but to the human race as a whole. Therefore, this group argued, the development of an H-bomb should be avoided if at all possible. A necessary (though clearly not sufficient) step in avoiding such a development was, of course, our refraining from the initiation of a program such as Teller wanted. Oppenheimer was Chairman of the General Advisory Committee of the Atomic Energy Commission at that time, and at a crucial meeting of that body his views were supported by all the members present.

Most politicians, like most scientists, were uninformed about and uninterested in this secret and arcane battle. Of those who were informed and interested, a few, and most of these halfheartedly, supported the Oppenheimer point of view. A larger group very strongly endorsed the views of the Teller party. This group was especially influential, since it included the highest-ranking Air Force officers and civilian officials, plus Lewis L. Strauss, then a member of the Atomic Energy Commission, and some of the key members of the very powerful Joint Congressional Committee on Atomic Energy, notably the Chairman, Senator Brien McMahon, now deceased, and then Representative Henry M. Jackson.

The resulting struggle was presided over at first by President

Truman and then, in its last phase, by President Eisenhower. In January, 1950, President Truman announced his decision that an effort be made to determine the technical feasibility of the thermonuclear weapon. He ordered the AEC and the Defense Department to fix jointly the rate and scale of the program, and he ordered State and Defense to review American foreign policy and military policy in the light of the existing and prospective Soviet nuclear capability. From that point on, the matter was, for all practical purposes, in the hands of the generals and the technologists. They presented a series of faits accomplis that really determined what happened next. After most of the major technical and political issues were resolved, but before all the loose ends were neatly tied together, Dwight D. Eisenhower became President of the United States. He appointed Lewis Strauss at first as his Special Assistant for Atomic Energy Matters, and then later, with the advice and consent of the Senate, as Chairman of the U. S. Atomic Energy Commission. In that role, Strauss proceeded to clean up the loose ends with a vengeance.

The drama has been ably and fully described from all points of view by the protagonists themselves as well as by others, some in semifictional form, and I will refrain from giving my own description of it. What matters here is the results of the struggle: ( I ) the invention of the hydrogen bomb, (2) the establishment of a second nuclear-weapons-development laboratory at Livermore, California, and (3) the defamation of J. Robert Oppenheimer, surely one of the most miserable events in American history and entirely unnecessary to boot, since by that time the first two results had long since been accomplished and the Teller-Strauss view had completely carried the day.

As I mentioned earlier, on the basis of my comprehension of the then existing general political situation, I supported Teller in his efforts to achieve the first of these results, the invention of the hydrogen bomb, and I ended up being the

first director of the Livermore Laboratory, which was the second of these results. However, along with a number of others who supported the accomplishment of these first two results, I opposed the prosecution of Oppenheimer. But I protested neither very loudly in private nor at all in public; I refrained from doing so for what I think were two good reasons. My first reason was my special relationship with Ernest O. Lawrence, who, as has been widely reported, quite generally supported Edward Teller in his views on all these matters. Ernest was my immediate boss and much more besides. I was extremely fond of him and felt I owed everything I was professionally to him. I still consider myself in a sense to have been his last student. I absolutely could not do anything which would add to the burden these problems were already causing him, and which in fact were making him seriously ill.

My second reason was my relationship with a number of other good friends and colleagues, including especially Harold Brown, John Foster, Gerald Johnson, Mark Mills, Arthur Biehl, Ernest Martinelli, and Kenneth Street, as well as Edward Teller himself. We were all committed to each other in the task of establishing the Livermore Laboratory. Because of the Oppenheimer affair and other problems which predated it, the times were especially difficult and our enterprise was just then in a particularly delicate condition. An open break between me and Teller or Strauss would have very seriously imperiled the whole endeavor, and so I avoided one. I still think that I did right, considering what the circumstances were and how I comprehended them, although sometimes on rainy nights I have some doubts.

The technical development of the H-bomb itself, which flowed steadily forward despite this maelstrom of political events, has been described in Teller's book The Legacy of Hiroshima and elsewhere. In May, 1951, the first thermonuclear reaction to take place on earth was produced at Eniwetok as a part of Operation Greenhouse. The device which pro-

duced it was not in any sense an H-bomb, but the subsequent analysis and study of the physical environment and sequence of events which enabled the reactions to occur led Edward Teller and others at Los Alamos (especially including Stanislaus Ulam) to invent the basic design which then led directly through only one intermediate step to the practical H- bombs of today. This intermediate step was the Mike Event of Operation Ivy. The explosive mechanism was properly called a "device" and was in no sense a practical bomb, but its successful detonation led immediately to the design of devices that were. The first true H-bomb was tested at Bikini in Operation Castle in the spring of 1954. From then until the test moratorium which began in the fall of 1958, subsequent H-bomb and A- bomb tests were largely devoted to developing nuclear explosives for particular applications. Especially important among these were the warheads of the various long-range missiles that were being developed on crash schedules at the same time. Also as a result of technical progress during this period, it became possible to build nuclear weapons of much smaller dimensions and weight. This permitted widespread deployment and dispersal on high-performance tactical aircraft all over the world, which, in turn, increased the danger of unauthorized use. The authorities responded to this danger with still more complicated gadgetry.

Meanwhile, in early August, 1953, midway between Operations Ivy and Castle, and only nine months after the Mike Event, the Soviets exploded their first crude experimental hydrogen device. They too developed this two years later into a variety of practical bombs and warheads. The close spacing of all these events makes it clear that the Russians started a serious H-bomb research and development program at just about the same time we did, perhaps a little before, perhaps a little after. It is still unclear whether or not restraint on our part would have resulted in restraint on their part, though there certainly is no evidence it would have. Nor is it clear

whether we really would have lost anything by refraining from going ahead until after they had made a first test. Had we waited, our program would have had much larger and wider support. Such support would most probably have enabled us to catch up easily with the Russians before any politically significant imbalance in the development and deployment of nuclear weapons (both A- and H-bombs) could have developed. In other cases which I will discuss later and which involve both missiles and space vehicles, the overreaction which has occurred whenever we first became aware of their having started something before we did has more than compensated for their earlier start. We have always had plenty of time to block any hypothetical attempt on their Dart to take advantage of their having started first.

In addition to achieving a thousandfold increase in explosive yield by changing from fission energy to thermonuclear energy, we also increased our destructive capacity by accelerating our unit production rate. We expanded existing plants and constructed new diffusion plants for separating uranium isotopes at Paducah, Kentucky, and Portsmouth, New Hampshire, and new reactors for producing plutonium at Savannah River, South Carolina. By about 1960, we had H-bombs running into the thousands and probably around ten thousand nuclear weapons in all. (By 1964 Senator John O. Pastore was able to say, "We number our weapons in the tens of thousands.") In 1960, John F. Kennedy estimated that the world's nuclear stockpile contained the equivalent of thirty million kilotons, and the Sixth Pugwash Conference used sixty million kilotons as a working assumption. Most of this was in the American stockpile. While most individual weapons were for special purposes such as air defense, tactical applications, and naval uses, the overwhelming bulk of the total explosive power resided in those weapons designed for strategic purposes that is, for the bombardment of enemy cities and strategic military installations. We are therefore safe in assuming that the United

States possessed at the beginning of the sixties a strategic-weapon stockpile containing twenty to forty million kilotons of explosives, or the energy equivalent of some ten thousand World War IIs, most of which could be released in a matter of hours. We had reached a level of supersaturation that some writers characterized by the word "overkill," an understatement in my opinion. Moreover, we possessed two different but reinforcing types of overkill. First, by 1960 we had many more bombs than they had urban targets, and, second, with a very few exceptions such as Greater Moscow and Greater New York, the area of destruction and intense lethality that a single bomb could produce was very much larger than the area of the targets. Since all or practically all strategic weapons were -by then thermonuclear, it is safe to assume that those Soviet or Chinese cities which were equivalent in size and importance to Hiroshima and Nagasaki were, by that time, targets for weapons from one hundred to one thousand times as big as the bombs used in history's only two real demonstrations of what actually happens when large numbers of human beings and their works are hit by nuclear weapons.

The lethality of bombs in the megaton class, due to blast and thermal effects, increases neither directly wish' the explosive yield nor (as the physics of the explosions might indicate) as the two- thirds power of the yield. Rather, the lethality increases only very slowly in most real applications. But the reason for this is to be found not in the realm of physical phenomena, but in the fact that the size of the bomb has outrun the size of the target and most of the lethal effects are wasted on the sparsely populated areas surrounding the urban centers. Some people may believe, and some officials may argue, that the foregoing is untrue or at the very least irrelevant because our bombs are aimed at "hard military targets" rather than against populations. Unfortunately, such is not really the case, for two reasons: First, certain common military targets such as missile silos will be largely empty when it comes time

to launch our weapons if we continue to stick with our present policy of deterrence (i.e., the policy of being ready to launch an unacceptably large second strike in retaliation against anyone who strikes first). Second, many so-called military targets are government and command centers such as Moscow and Peking and Warsaw and other capitals, major transport and communication centers such as Kharkov and Kiev, major seaports such as Leningrad and Vladivostok, and major scientific centers such as Novosibirsk. In short, a list of this second type of so-called military targets is for all practical purposes identical with the list of major population centers, and so any such distinction, even though it is often made, is meaningless. To be sure, there are some military targets which are not parts of major cities and which will not be empty when it is time for retaliation. These include, for example, air defense installations in the Far North, but the relatively small number of weapons which would be diverted to them does not substantially change the picture.

So far we have not considered fallout in this discussion of the holocaust that nuclear war can produce. The details of how heavy the fallout is, where it goes and how soon it comes down depend on such factors as the composition of the bomb, the composition of the soil, the altitude of the burst, and the wind conditions at all levels up to one hundred thousand feet or so. For ground bursts, such as would be used against hard targets or against generally soft targets which unhappily happen to contain a hard target within them, we may conservatively and for convenience assume that for every two kilotons of explosive energy about one square mile will be bathed in lethal levels of radioactive fallout within hours after the explosion. Thus, by about 1960 the two major nuclear powers together possessed stockpiles which could bathe tens of millions of square miles in lethal levels of radioactivity, an area larger than the total land area of the two principals themselves. Of course, the fallout would not be distributed uniformly; some

areas would get much more than a lethal dose and others would get less. Also, much of the lethal area would be outside the boundaries of the two powers. For example, an attack on the U. S. industrial heartland running from Chicago along the Great Lakes to the East Coast megalopolis would result in lethal fallout all over southeast Canada, which is where most Canadians live. Beyond this matter of lethal fallout lies the still more ominous matter of genetic and other long-term effects of fallout on the human race as a whole. This is still not fully understood, but appears to be extremely serious. It is this awesome fallout situation which has led to the various doomsday predictions such as that which was fictionalized in Nevil Shute's On the Beach.

If that's where we were in 1960 in nuclear weapons technology and capability, one might ask, what have we been doing since then and why? The answer to this question is that both the U. S. and the U.S.S.R. have continued throughout the sixties to develop, test and stockpile new, often baroque, sometimes rococo, varieties of A-bombs and H-bombs. The testing of these bombs and other nuclear devices has been conducted underground except for a brief period extending from late 1961 into 1963. The reason for this is that the Limited Nuclear Test Ban Treaty which came into force in the summer of 1963 prohibits testing anywhere else. China and France are not signatories and have not complied with this prohibition. In the case of the Soviet Union, these tests and developments have included bombs of still greater size than those developed in the fifties. It is generally thought that a Soviet test in 1961 involved a device yielding about sixty megatons. In the case of the United States, weapons development and testing since 1960 have been largely devoted to adapting A-bombs and H-bombs to special purposes and special environments and to making other qualitative improvements, rather than to making them bigger and more powerful. Thus, our test and development program has been directed toward developing bombs which

emphasize such special weapons effects as high X-ray yields for use in a hypothetical antimissile defense; to developing techniques for maintaining better command and control over the widely dispersed stockpiles, including especially those in foreign countries; to developing warheads which could better cope with more hostile environments, including very high accelerations and the effects of hypothetical enemy explosions designed to intercept our warheads; and to reducing the size of thermonuclear weapons so that they could be adapted to other special purposes, one of the most important of which is the provision of multiple warheads for a single missile. This last development program goes under the general name of MIRV, an acronym standing for "multiple independently targeted reentry vehicles." Numbers of warheads per missile of more than ten have been mentioned.

One of the political prices paid for getting wide acceptance of the Limited Nuclear Test Ban Treaty in the Congress consisted in a promise by the AEC to conduct an underground-test program vigorous enough to "satisfy all our military requirements." But, clearly, nuclear-weapons tests and experiments conducted underground are more complicated and less productive than they would be if they were held in the atmosphere. It has therefore been judged necessary to conduct such testing at an even faster rate than in the fifties, when aboveground testing was possible. During the sixties the operations at the Nevada test site were conducted on a year-round basis rather than in batches as they were in the forties and fifties.

Also during the sixties our strategic forces changed over from sole dependence on aircraft (in 1960 mostly B-47s and B-52s) to a combination of land-based missiles (now Minutemen and Titans) plus sea- based missiles (now Polarises) plus aircraft (now mostly just B-52s). The missiles deliver a smaller payload than aircraft do, and a larger part of the payload has to be devoted to coping with severe environmental conditions

such as those met on reentering the atmosphere or in a near-miss by a hypothetical intercepting explosion. For these reasons, plus the fact that more of our officials had come to appreciate how huge our overkill capability was (even though they may not have used that word), the total yield of our stockpile did not continue to skyrocket upward during the sixties as it had during the fifties. Rather, it probably reached a peak in the early sixties and then either leveled off or began to slowly decrease. However, before anyone breathes a sigh of relief, let me note it is only in the area of total yield and hence total fallout potential that there has been any slowdown in the mad rush toward doomsday. The total number of warheads, and therefore the total number of targets which can be hit, has continued to rise and is about to rise especially rapidly in the near future as a result of the deployment of MIRV. And in the case of the Soviet Union, all of these numbers, including even the total megatons in the stockpile, have been increasing especially rapidly in the last couple of years.

Some, such as Eugene Wigner in his speech to the spring 1969 meeting of the American Physical Society, have seen this decrease in the total megatonnage of our stockpile accompanied by an increase in the total number of individual warheads in our possession as a serious attrition in our over-all capability. This is sometimes referred to as a "virtual attrition" which has been brought about by the Soviet ABM system now deployed around Moscow. The idea here is that whether or not this Soviet ABM system works, the U. S. invention and deployment of MIRV and other penetration devices was prompted by the deployment of the Soviet ABM, and this MIRV deployment is one of the direct causes of the decrease in total megatonnage. It is a fact that our MIRV and other penetration devices (more on these later) were our response to Soviet developments in the ABM field, but even so the concept of "virtual attrition" is basically false, especially in the case of a strategic posture whose purpose is to deter

war--that is, to be able to make an unacceptable second, retaliatory strike at the cities and other vitals of the nation which strikes first. The notion of "virtual attrition" is based on the false notion that there is such a thing as a "small nuclear weapon." Small nuclear weapons are like northern Mississippi: there is no such thing, except in a detached theoretical sense. Some parts of Mississippi are, to be sure, farther north than others, and some nuclear weapons are, to be sure, smaller than others; but on a human scale, and compared with the structures that human beings ordinarily build, all A-bombs and H-bombs are very large indeed. In his discussion of this matter, Wigner stated that the present single ten-megaton warhead now carried by the Titan was to be replaced by a MIRV warhead consisting of ten individual nuclear warheads of fifty kilotons explosive power each, and he bemoaned what he calculated to be a loss in the lethality of the system. I know of no plan to do any such thing; apparently Wigner had become confused by plans for equipping the Navy's Poseidon missile with a MIRV such as he described. In any case, there simply is no fundamental reason for such a large change (twentyfold) in the yield to occur when single warheads are replaced by MIRVs. Some small decrease is inevitable, but the total area covered by some given blast level is not necessarily reduced; on the contrary, it normally increases somewhat.

Even so, let us suppose Wigner's notion about the plans for a MIRV for Titan were true. Each of those ten 50-kiloton warheads is almost four times as big as the warhead which gutted Hiroshima and killed nearly one hundred thousand people in the process. Except in the narrowest technocratic interpretation, it makes absolutely no sense to call such bombs small. Unless, perhaps, either Moscow or Leningrad is the target, the imaginary Titan MIRV (i.e., the real Poseidon) is deadlier and more vicious than the real single-warhead Titan.

Between 1950 and 1960, the explosive yields of nuclear weapons changed a thousandfold, the means for delivering

them (subsonic aircraft) changed rather little, and the number of persons in the Sino-Soviet bloc who were "at risk" to the U. S. strategic forces changed from a few million to a few hundred million, or by about a hundredfold. Between 1960 and 1970, the explosive yields of nuclear weapons changed very little, the speed of the delivery systems increased thirtyfold (from subsonic aircraft to supersonic Mach-25 intercontinental rockets), but the number of persons in China and the U.S.S.R. who were at risk remained more or less a few hundred million. The reason for the lack of a significant increase in the number of persons at risk during the second decade, as contrasted with the huge increase during the first decade, is obvious: it is the supersaturated, or "overkill," capacity we reached in about 1960. This static situation has been maintained now for almost ten years despite all the breakthroughs and other technological advances that took place during the sixties, and we may therefore reasonably expect that new breakthroughs and deployments of new types of weapons in the near future will similarly produce little or no change in the number of persons at risk on either side. However, as we shall see, the situation has worsened in other ways. And, unless some large change happens, we can expect it to continue to do so.

THE BOMBER BONANZA

At the same time that America was pursuing its spectacular accumulation of nuclear weapons, it was devoting considerable attention to perfecting the means of delivering them. At first this effort naturally focused on the bomber.

The strategic bombardment of Germany in World War II by the United States Army Air Force was carried out largely with B-17s and B-24s. These subsonic propeller-driven aircraft flew at speeds of a couple of hundred miles per hour. They could carry payloads weighing a few tons for distances ranging up to a thousand miles or so. They dropped the bulk of the 2.7 megatons of chemical high explosives which were showered on Germany during the latter part of the war. The bombardment of Japan, including the fire-bombing of Tokyo and other major cities and the delivery of the two atomic bombs, was carried out with B-29s, another type of subsonic propeller-driven aircraft. These had been especially designed for the Pacific campaign, where a much longer range was needed than in the European campaign. Aircraft continued after World War II and up until 1960 to be the sole means of delivery for our strategic weapons. And even in 1970 the great bulk of the total megatonnage in our nuclear stockpile is still programmed for delivery by aircraft. After World War II the B-36 bomber, an extremely

long-range propeller-driven aircraft, was introduced into the force to give us a home-based intercontinental strategic bombardment capability. In keeping with the general development of aviation, both civil and military, all of these propeller aircraft were eventually replaced by the B-47 and the B-52 jet bombers.

The first American jet bomber, the B-47, was introduced into the Air Force inventory in the late forties. It did not have either the desired range or the desired payload-carrying capability, and so a still larger airplane, the B-52, was designed specifically to meet these needs. The B-52 was simply a much larger version of the B-47. Both have speeds in the high subsonic range (about six hundred miles per hour). The B-52 began to be phased into the force in 1954. Today the largest part of our strategic stockpile in terms of total explosive power is still programmed to be delivered by one of the later models of this same B-52.

The B-58 was designed in the early fifties to meet the desire for a plane capable of supersonic speeds over enemy territory. It became operational in the late fifties and obsolete in the late sixties. This aircraft was programmed to fly most of the distance from its home base to a target at subsonic speeds, and then to dash in over the target and out again at about twice the speed of sound. The inventory of these airplanes never became very large, and consequently they never played a major role in our strategic-delivery plans.

In addition to gradual progress in the development of airframes and aircraft propulsion systems, the years since World War II have seen further developments in devices to confuse air defenses (electronic countermeasures and chaff), in weapons designed to roll back the defenses, and in the development of "standoff" weapons. These last include various types of small pilotless aircraft and guided rockets. These can be launched directly from the B-52s, and they can then fly ahead on their own for hundreds of additional miles, making it un-

necessary for the aircraft themselves to penetrate all the way to the immediate target area. The longest-range missile of this type ever to be seriously considered was the famous Skybolt missile, which after several years of on-again, off-again and development effort, was finally canceled at the beginning of the Kennedy administration.

Since the first plane rolled off the production line in 1954, the B-52 has gone through a series of major model changes. Some of these changes resulted from modification in operational concepts and requirements. But most of them resulted from the steady advance of the technological state of the art in engine design, aerodynamic design, and "avionics" (a general term coined to cover all of the various electronic devices carried by modern aircraft for communications, navigation, weapons control, defense penetration, etc.). As a result, the B-52G's and the B-52H's (the last one came off the line in 1962) were very different from the original B-52A. In fact, the difference between the B-52H and the B-52A was almost as great as the difference between the B-52A and the B-47.

Except for the B-58, no entirely new strategic bombardment aircraft has been introduced since the B-52. The reason for this has not been a lack of ideas, but rather that attention since the mid-fifties has been primarily focused on an entirely new kind of intercontinental delivery system, the long-range ballistic missile to be discussed later. Nevertheless, the Air Force has continued to propose several new radically different types of aircraft. Because of what their stories reveal about the dynamics of the arms race, two are especially worthy of mention: the huge supersonic B-70 and the nuclear airplane, or the ANP, as it was usually called.

The B-70 program was initiated in the late 1950s. Only

two prototype aircraft were ever built. These were flown in the mid-sixties, and the program is now dormant. The plane was to be capable of flight at Mach 3 (i.e., at three times the speed of sound, or about two thousand miles per hour), to fly at

attitudes in the neighborhood of 100,000 feet and to carry payloads of many tons to targets anywhere in the world. Its gross weight at takeoff would have been over 500,000 pounds. The airframe was to have been constructed of titanium and stainless steel, since the aluminum alloys used for all other aircraft would not stand up under the extreme temperatures generated on the leading surfaces at such high speeds.

The history of the B-70 is closely linked to the political and technological environment of the era.

The program began in the mid-fifties as a study of what was then called the WS-110. By early 1957, development work (mostly done at the old NACA laboratories) on the various kinds of components and design ideas necessary for long-range supersonic flight had reached a point where it became clear to everyone concerned that a Mach-3 airplane of intercontinental range could really be built. As a result, in mid-1957 the Air Force authorized both Boeing and North American Aviation to engage in a competitive design study.

Meanwhile, the huge programs to develop the intercontinental ballistic missiles, or ICBMs, had been started and had been given the highest national priority. As we will see, they soon came to dominate the technological scene in the U. S., and they absorbed the bulk of the resources, including both men and money, which the Air Force could devote to research and development. Thus, even i! the studies showed the project practicable, it was not likely that the U. S. would be able to commit the necessary resources to it. But on October 4, 1957, shortly after the study started, Sputnik, the first artificial earth satellite, was launched into space by the U.S.S.R. The political atmosphere both in Washington and throughout the country was transformed by the sudden shock of discovering that the United States was not automatically first in technological feats of that sort. Frightened by the Soviets' apparent technical superiority, Americans were disposed to listen sympathetically to anybody with an advanced-technology program to sell.

Thus, when North American Aviation was selected as the prime contractor for the project at the end of 1957, the firm was ordered by the Air Force to proceed on a high-priority basis with the development of what had become known by then as the B-70. The official priority rating given was just below that of the ballistic-missile development program. The B-70 appealed particularly to the flying generals, who did not look forward to becoming "the silent silo-sitters of the sixties." They took a different view from those who advocated the primacy of the ICBMs. General Curtis E. LeMay, the man with the cigar, was the commander of the Strategic Air Command (SAC) at the time. As I recall his personal view of the priorities, he placed the B-52H first (it was then called the B-52 Squared) and the B- 70 second (it was then called the WS-110). The nuclear airplane (ANP) was somewhere in the middle of his short list, and the long-range missiles were at the bottom. He and other leading Air Force generals managed to make it clear to the contractor that they personally considered the B-70 to be at least as important as the ICBMs, whatever the official priorities might be, and they ordered first flight by the end of 1961.

Before the first full year under contract was over, there were more than forty first- and second-tier subcontractors, and approximately two thousand vendors and suppliers were by then involved in the total program. Seventy of the then ninety-six United States Senators had a major part of the program in their states, and something like a majority of the Congressional districts had at least one supplier of consequence.

Many different arguments in favor of the B-70 program were presented by its proponents to the Congress and in the aviation and missile press. It was said that, as compared with conventional bombardment aircraft, its speed gave it certain very important advantages: it could respond much faster, it could arrive in the target area much sooner, and it could penetrate air defenses more surely. Compared with missiles, its

main advantages were said to reside in its greater "flexibility." That is, like any other aircraft, it could be launched into the air very soon after receiving a warning and it could be easily recalled if the warning turned out to be a false alarm. It could carry out a search-and-destroy mission on arrival in the target area, and it could deliver larger weapons with greater accuracy than was generally foreseen for missiles. In addition, the claim was made that it could also be useful for such purposes as "showing the flag" and serving as a high-velocity platform for launching artificial earth satellites. In this latter application, it could in theory substitute for some other non-recoverable rocket first stages, such as the Atlas or Titan booster. Because of it' enormous speed and great flight altitude and these other real and hypothetical advantages, it was sometimes called the "manned missile."

At the time the contract for building the B-70 was awarded to North American in 1957, the multistage intercontinental ballistic missiles, or ICBMs, were still in a very early test phase, and the single-stage intermediate-range ballistic missiles, or IRBMs (Jupiter and Thor), had been successfully launched for the first time only months before. The first Atlas B, with all engines operating, would not be launched until some months after the B-70 program was initiated, and even then only to less than half of its intercontinental range. The guidance and control systems for these long-range missiles were also still in a very early stage of flight test. Although some of the leaders in this field, particularly Stark Draper of M.I.T.'s Instrument Laboratory, were predicting extreme accuracies down the road, it was reasonable to believe then that aircraft delivery accuracy would continue to be very significantly better than missile delivery accuracy. Hence, the B-70 received its priority go-ahead at the time when advanced technology programs generally received a friendly welcome from the Congress and the people, and when its technical competition was still in too early a development stage to offer a sufficiently convincing alternative.

By the end of l 958, though, an Atlas-D missile had reached a range of four thousand miles, and the first field test of an operational missile was conducted in September, 1959. By October, 1959, some $300 million had already been spent on the B-70, and many hundreds of millions more were scheduled to be spent during the remainder of that fiscal year (i.e., through June of 1960). The Air Force was asking for an additional $460 million for the next fiscal year. In the budget planning for fiscal year 1961, which took place as usual six to nine months before its beginning--that is, during the last three months of calendar year 1959--we pared this figure back in the Office of the Secretary of Defense to about $360 million. We then discussed this and other development programs with representatives of the Bureau of the Budget and with Dr. George B. Kistiakowsky (then the President's Special Assistant for Science and Technology), and it soon became evident that there was rising opposition in the White House staff to spending anything like these amounts on this program. In a later meeting with President Eisenhower himself at his vacation headquarters in Georgia, it was finally decided that we should cut the program all the way back to an annual level of $75 million for the fiscal year beginning July, 1960, and that we should reduce expenses immediately for the remainder of the then current fiscal year so as to reach that goal in 8 rational manner. Such a program level would allow exploratory development work on components and certain advanced subsystems, but it would eliminate the construction of any prototypes, and, of course, indefinitely postpone any deployment plans. North American Aviation received an order from the Air Force to this effect in early December of 1959.

An intense campaign to save the B-70 was immediately launched in the Congress, in the aviation and missile press, and on a wider front in the general media. All kinds of visions of potential national dangers (and, just incidentally, lost jobs) were conjured up. Carl Vinson of Georgia, the chairman of the House Armed Services Committee, said, " . . . by cutting

back the B-70 we have increased the danger to our survival . . ." Senator Barry Goldwater, who was a brigadier general in the Air Force Reserve, personally appealed the case to President Eisenhower. Senator Clair Engle, Democrat of California, and also an officer in the Air Force Reserve, characterized the cutbacks as a "blunder which may have the gravest consequences to our national security," and he claimed that the B- 70 was needed to make up for what he said was the "five-year lead of Soviet missile developments." Even so, the lower budgetary level was maintained throughout the first half of 1960.

Then, during the 1960 campaign for the Presidency, the B-70 was given a brief new lease on life. Even before the new fiscal year started, on July 1, 1960, about $60 million had been tacked onto the originally planned $75 million. This extra money was supposed to be used for development work on some of the most critical weapons subsystems; and in combination with other readjustments in the project, it was to make possible the construction of a single prototype aircraft. However, a program leading to only one prototype never made sense, and going through such a step was nothing more than an exercise in salami tactics. Thus, in August, another $20 million was added for a second plane. Then, just days before the Nixon- Kennedy election contest in November, 1960, the Department of Defense announced that it was bringing the total B-70 budget for the then current fiscal year up to $265 million. As a result of these increased funds, the number of airplanes to be built was increased to four for sure, with eight more possible, and the four were to be prototypes of a "usable weapon system." In California, the announcement of this new lease on life was accompanied by a detailed statement by North American Aviation about the recent sad history of declining employment in southern California and how these funds would change all that.

Although Nixon did carry California in 1960, Kennedy won

nationally, and the B-70's new lease on life ran out almost immediately. Shortly after the inauguration, I, along with Secretary McNamara, Jerome Wiesner, then Kennedy's science adviser, David Bell, the new Director of the Budget, and a few others, attended a meeting with the new President where the outlines of Kennedy's first defense program were developed. By the time of that meeting long-range missiles of several different types had been successfully flown, sufficient reliability and accuracy had been demonstrated, several models of missiles were already deployed, and the first Minuteman (a second-generation ICBM) had just been test-flown successfully. And, no doubt most important, the new Administration kind confirmed the claims of the prior one that the missile gap for all practical purposes did not exist and had stopped claiming that it did. As a result, the B-70 program was cut back once again, this time to one which would produce two prototype aircraft but no more. When President Kennedy announced this decision, Senator Goldwater immediately denounced it and said that it would "go down in history as one of our worst tactical blunders." Other members of the Congress, including especially some from California, made similar statements And the Los Angeles County Board of Supervisors passed a motion urging the Administration "to reimplement the program for mass construction of the B-70 supersonic bomber." The Board duly noted that "much of the work of constructing these bombers would be done by Los Angeles area concerns."

Over the next three years, a battle of growing intensity raged between the executive branch and the Congress over the B-70. McNamara requested an appropriation of about $200 million annually for the program and stated that his objective was to build two prototype aircraft. Characteristically, the Air Force generals made it known that that was not their objective, and the Congress, inspired especially by Chairman Carl Vinson of the Armed Services Committee, appropriated or tried

to appropriate about $300 million more per year than the Secretary of Defense requested. It also tried to convert the program to one committed to produce about two hundred so- called reconnaissance-strike bombers at a cost usually estimated by friends of the B-70 at $10 billion. It was not for naught that the generals and the admirals referred to the chairman as "Uncle Carl."

For a time the argument was raised to the level of a constitutional issue. At one point Vinson said in a speech, "What is Congress's function in defense? Is it a partner? Does it have a voice? Or is it just a Mr. Moneybags, to give or withhold funds? That's not what the Constitution says; the Constitution grants the Congress the exclusive power to raise and support and make rules for our military forces."

I think those were, in the abstract, very good questions. Unfortunately, though, the real issue was then not so much about the merits of and need for the B-70 as over the question of military versus civilian control of defense planning. It was not until years later, in the 1969 ABM debates, that the Congress got around to arguing the real merits of an important weapons system, and, in the process, declined to accept the word of either the civilian secretariat or the generals as gospel. (Like the B-70 dispute, the TFX controversy which raged during the middle sixties was largely over the issue of civilian versus military control.)

But McNamara, with the support of the President, stood his ground and refused to spend the extra funds on the B-70, even when they were fully authorized and appropriated as an integral part of the final defense budget.

The program did finally work out middling well in a technical sense. Two prototypes were built, and the first of them flew in September, 1964. In another flight a year later, a speed of Mach 3--two thousand miles per hour--was reached and maintained for two minutes during a 107-minute, 1,900-mile flight out of Edwards Air Force Base in California. One of the