A LearnWorld Text

Race to Oblivion

A Participant's View of the Arms Race

Herbert F. York




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




There are a staggering variety of missiles in the world today. They come in a great range of sizes; they exhibit many forms and shapes, and they are designed to accomplish a wide variety of specific objectives. Some missiles are nothing more than pilotless aircraft. Their weight is supported on wings, and they are propelled by jet engines which burn common liquid fuels such as kerosene and gasoline. Missiles of this type are usually known as "cruise missiles" or "air-breathing missiles." Other missiles are shaped more like elongated bullets and are rapidly accelerated to high velocities by rocket motors. Then, like a bullet, they coast along a trajectory determined by their initial momentum and the force of gravity acting on them. Missiles of this type are called "ballistic missiles."

Some missiles are designed to be used for strategic warfare; i.e., their purpose is to eliminate the basic war-making potential of any enemy by killing his people and destroying his factories. Other missiles are designed to be used tactically: their purpose is to destroy enemy soldiers and weapons in direct combat circumstances. Still other missiles are designed to be used for defense: their purpose is to intercept and destroy attacking aircraft and missiles.

The rockets which launch manned and unmanned satellites into orbit around the earth or onto a trajectory to the moon


are, of course, technologically identical with those which are used to launch ballistic missiles on their paths toward an enemy target. The only really important difference between the two is in the cargo. Thus it is not surprising that the technological and political histories of the space program and the missile program are tightly interwoven.

The only strategic missiles that have ever been fired in anger in modern times were the German V-1 and V-2. In the waning days of World War II these missiles were used to bombard London and various seaports in Allied hands, both in England and on the Continent.

The V-1 was a winged air-breathing cruise missile propelled by a pulse jet engine. It weighed just under five thousand pounds; its fuselage was twenty-seven feet long and thirty-three inches in diameter. It carried a one- ton chemical-explosive payload. Its rudimentary control system was preset for the range and direction of its target before takeoff, and as a consequence its accuracy was very poor. It was also guise unreliable and highly vulnerable to defensive measures. Considerably less than half of the eight thousand V-1s (or buzz bombs, as the English called them) launched toward London actually arrived there; the majority were either intercepted by British air defenses or went completely astray.

The V-2 was a ballistic missile propelled by a rocket motor using alcohol as its fuel and liquid oxygen as its oxidizer. (Rockets normally carry with them both of the chemical components necessary for the combustion: a fuel and an oxidizer. Jet-type missile engines, of course, get the oxygen they need for combustion directly from the air, as do airplane and automobile engines.) The V-2 was forty-six feet long and sixty-five inches in diameter, and it weighed 27,000 pounds. Just as the fuel was exhausted (or "at burnout," as it is usually called) the missile reached a speed of very nearly one mile per second. As a direct consequence of this speed its maximum range was approximately two hundred miles. At the time, it was the


fastest vehicle that had ever been built, and it was not much slower than even today's satellite launchers and intercontinental ballistic missiles, which reach speeds in the neighborhood of five miles per second. The V-2 was steered to its target by what is called an "inertial guidance system." That means that it carried within itself a mechanism which could (1) sense how fast and in what direction it was moving and (2), based on that information, give orders to the propulsion and flight-control systems to make it arrive at a predetermined destination. The accuracy of this guidance system was such that most of the missiles landed within three to five miles of the target selected, or, to put it differently, within a distance from the target which was about two percent of the range to the target. The V-2's payload of chemical explosive was, like the V-l's, just about one ton in weight. However, entirely unlike the V-1, none of the V-2s were intercepted by British air defenses; 1,500 of them landed in England and killed more than 2,500 persons in all.

Both the V-1 and the V-2 were developed during World War II at Peenemünde, Germany, in a laboratory under the direction of General Walter Dornberger and Wernher von Braun. After the war most of the key personnel, including both Dornberger and Von Braun, plus much equipment and much documentation, were brought to the United States as part of Operation Paper Clip. Some of the persons and equipment ended up in Russian hands, and a few individuals ended up in France and Britain. Much,, but by no means all, of the development which followed in the U.S.A. and the U.S.S.R. was based on the German experience and involved the participation of these former members of the Peenemünde team.

During the war none of the Allied powers carried on any program aimed at the development of large strategic missiles in the V-1 or V-2 class, but they all did develop and manufacture smaller missiles and rockets for other purposes. One of the best-known of these is, of course, the hand-held bazooka.


One of the most important, insofar as its long-range consequences were concerned, was the JATO (jet-assisted takeoff) rocket used to assist slow, heavy aircraft (such as seaplanes) in taking off.

In the United States there were several different projects for developing and producing these JATO rockets. One of these, supported by the Navy, was under the direction of Robert H. Goddard, the "father" of rocketry in this country. Others were conducted by Reaction Motors, Inc. (originally formed by members of the American Rocket Society), by GALCIT (Guggenheim Aeronautical Laboratory, California Institute of Technology) and the Aerojet General Corporation. Some of these wartime JATO units developed thrusts in the neighborhood of three thousand pounds.

Two entirely different types of rocket motors were used for these JATO applicators; one kind used liquid propellants and the other kind used solid propellants.

In the case of liquid-propellant engines, such as propelled the V-2, two liquids, one a fuel and the other an oxidizer, are pumped into a combustion chamber. There they mix and burn, producing high temperatures and high pressures which force their combustion products to exhaust through a nozzle in the tail of the rocket with great speed. In the process of exhausting to the rear they cause the rocket to recoil in the opposite direction, thus driving it forward. The most practical of the various fuels used during the war was aniline, and the usual oxidizer was red fuming nitric acid. Both of these liquids are nasty to handle, but they have the advantage of being storable at room temperature (unlike liquid oxygen) and of igniting on contact, so that no special ignition system is needed to get things started (unlike the case when kerosene is used as the fuel).

The solid-propellant motors are more like those used in Fourth of July rockets: the fuel and the oxidizer are already mixed in a dry form, and then are lighted at one end. A narrow burning zone rapidly moves through the mixture, creating high


temperature, high-pressure gaseous combustion products as it does so. These, as in the ease of the liquid rocket, exhaust rearward through a nozzle, causing the rocket as a whole to recoil and thus move forward in the process. A mixture commonly used in World War II consisted of asphalt (which acted both as the fuel and as the binder holding the mixture together) plus potassium perchlorate, a powerful oxidizer in the form of a crystalline powder.

The reason the Allies did not attempt to develop missiles for strategic bombardment was a very simple one. With our technological industrial base and with the state of the art as it existed at the time, aircraft were far superior to missiles for such purposes. Aircraft were far more flexible they could carry much larger payloads, they had much longer ranges, and they could deliver their bombs with much greater accuracy. The V-1 and V-2 missiles were really just psychological terror weapons with very limited direct effectiveness, whereas bombardment aircraft could, in principle at least, go after specific targets such as particular factories, bridges, railroad yards, etc.. Also, by the time the Germans began to use the V-weapons, the war was already going very badly for them, and these weapons may have had certain special advantages in just such a situation. For example, the V-2 could be launched from small, easily concealed pads rather than from large, easy-to-see, easy-to-disrupt airfields, and a single missile could get through; whereas the Germans at that time could no longer put up enough aircraft to gain the necessary advantage of saturation.

With the invention and demonstration of nuclear weapons at the close of the war, the whole technological picture changed. The destruction radius of even the earliest nuclear weapons against people, houses, and factories was one or two miles instead of only a hundred feet or so as in the ease of even the very largest chemical bombs. This large increase in destruction radius, combined with foreseeable improvements in the accuracy of guided missiles, led weapons-systems designers to


give serious consideration to the possibility of producing a new kind of weapon system: the nuclear-armed, intercontinental guided missile. It has become clear such weapons could perform at least some of the functions of bombers.

In the United States, strategic warfare was largely the responsibility of the Army Air Corps (soon to become the U. S. Air Force). Hesitantly at first, it began to explore this new possibility. Both air-breathing cruise missiles and ballistic missiles received serious attention as candidate systems.

Three air-breathing strategic missiles eventually came out of these studies. The first of these was the Snark, a subsonic, winged cruise missile with a range of five thousand to seven thousand miles. Northrup had the prime contract for the development of this missile, and Pratt and Whitney made the jet engines. It was boosted up to operational speed by two large JATO rockets. These were the solid-propellant type and generated a thrust of 33,000 pounds each. The vehicle had a gross weight of fifty thousand pounds and it could carry a payload of five thousand pounds, easily enough for the A-bombs of that day. After a long-drawn-out program, the Snark finally became operational in 1958; it was phased out only a few years later. We realized, several years before Snark became operational, that it would become obsolete by the time it was finally deployed, and repeated recommendations for dropping the project were made. However, in this case as in so many others, the momentum of the project and the politics which surrounded it made it impossible to do so.

The second of the strategic cruise missiles was Matador, for practical purposes a scaled-down Snark. It was built by the Glenn L. Martin Company; it had a gross weight of twelve thousand pounds and a range of 650 miles. It too carried a nuclear warhead. In order to be able to strike at our potential enemies, it had to be based in forward areas such as Japan and Germany. It became operational in 1955 and in the early sixties was replaced by the Mace missile, which had a longer range, greater speed, and superior accuracy.


The third of these large air-breathing winged cruise missiles was the Navaho. This missile was developed by North American Aviation under an Air Force contract. It was started much later than the first two, and consequently its design was based on a much more advanced technology. It was ninety-five feet long, it had small delta wings, and it weighed 300,000 pounds at takeoff. It was boosted into the air and accelerated up to three times the speed of sound (or about two thousand miles per hour) by three large liquid-propellant rocket engines, each one generating 135,000 pounds of thrust. After it reached that speed, ram jets took over and continued to propel it to its target a quarter or more of the way around the world. It was guided by a self-contained all-inertial guidance system. Navaho was flown for the first time in November, 1956, by which time the huge intensive programs to develop intercontinental ballistic missiles were already well under way. It was therefore again recognized, as in the case of Snark, that this weapon system would become obsolete before it was deployed. This time the fact that Navaho was less far along the road toward deployment made it possible to heed the repeated recommendations to cancel the program, and the missile was scrubbed well before it reached operational status.

Even though the Navaho was never deployed, it did play a very important role in our missile program. Its development provided the technological base for much of what was to follow. The large liquid- propellant rocket engines were the first of a long series of similar engines designed and built by Rocketdyne. Later engines in this series powered the Thor, Jupiter, and Atlas missiles and the huge rocket that started Apollo 11 on its journey to the moon more than a decade later. Also, the Navaho inertial-guidance system provided the technological basis of later, more advanced systems which were used in other missiles, aircraft, and even submarines.

In addition to these cruise-type missiles, the Air Force explored the possibility of using large rocket-powered ballistic missiles as intercontinental strategic weapons. In 1946 Con-


solidated-Vultee (later Convair) was given a contract to look into the matter further, and it began the design and development of the MX-774, an early version of what later became known as the Atlas. Soon after this work was started, and as a result of further internal reviews, the Air Force concluded that, given the existing state of the art, such a weapons system, even with a nuclear warhead, would take too long to develop and would not be competitive with other alternative approaches. The Air Force thereupon decided to discontinue any support for large long-range rockets and to concentrate its resources and energies in the more straightforward and conservative aircraft and cruise-missile programs. Consolidated-Vultee did manage to keep the program going along for a time on its own funds, but for all practical purposes the big-rocket approach to strategic weaponry was dead for the time being in the United States.

During the postwar years the Navy also conducted programs in the development of missiles for a variety of applications. These included the Rigel and then the Regulus I and the Regulus II cruise missiles which could be launched from surfaced submarines and used against either strategic or tactical targets hundreds of miles away on shore. The Navy, like the Air Force, also developed a number of air defense missiles, most of which were rocket-powered. The Army too had a vigorous program of rocket and missile development immediately-after World War II. Some of these rockets were developed for the purpose of upper-air research. Other surface-to-surface rockets, including the Corporal, were intended for use in tactical warfare. And still others, including the Nike series, were intended to be used for air defense.

In 1949, the Army decided to establish its major missile development activities at the Redstone Arsenal in Huntsville, Alabama. Among those transferred were 130 members of the original Von Braun team, as well as many military personnel and other government and contractor employees. The Korean


War broke out soon after this new missile-development group was established. As one result of this new stimulus to weapons development, the Redstone group was given the assignment to develop a longer-range ballistic missile for use in tactical warfare. After a brief period of some uncertainty, it was eventually established that this rocket should have a range of two hundred miles and carry a nuclear warhead. It was to be a mobile weapon capable of being launched under battlefield conditions by regular combat troops. The missile finally designed to meet these objectives was named the Redstone after the Arsenal. It was propelled by a slightly modified version of the 135,000-pound liquid-propellant rocket engine which was being developed by the Rocketdyne Division of North American Aviation for the Air Force's Navaho missile. The Redstone was the largest American rocket under development in the early fifties and, as we shall see, was to form the base for much of the program which followed. As a weapon, however, it never served any important function, and it was often referred to by other Army personnel as "the world's most expensive roadblock."

Starting in the fall of 1952, a series of closely spaced events wrought revolutionary changes in the content and style of our missile-development programs and, through them, all other military development programs as well. These were the invention and demonstration of the hydrogen bomb, the election of Eisenhower and the concomitant extensive personnel changes throughout the executive branch (the first complete change in twenty years), and the growing accumulation of intelligence reports which first indicated and then confirmed that the Soviet Union had already launched a major program for the development of large long-range rockets.

Six crash programs to develop long-range nuclear-tipped strategic missiles were born out of the confluence of these three events. These programs dominated the technological scene in this country throughout the fifties. The names of these missiles


all became well-known: Atlas, Titan, Minuteman, Thor, Jupiter, and Polaris. The first three are ICBMs (intercontinental ballistic missiles), the next two are IRBMs (intermediate range ballistic missiles), and the last is an SLBM (sea- or submarine-launched ballistic missile). All six programs were started within three years of each other, and for a time all six were being carried out concurrently. All six missiles had the same principal purpose: massive retaliation in the event of an attack. In retrospect it is clear that three such programs would have been entirely sufficient, and it would not have been necessary to have even these three all conducted on a crash basis. Why, then, did we overreact? The reason, once again, was the inflation of the anticipated threat and the multiplication of programs designed to counter it in short, excessive caution and prudence inspired by the thought which is the motto of the arms race: "Let us err on the side of military safety." In addition, the personal ambition of clever men played an essential role in at least one case.

Among the new officials who came into office as a result of Eisenhower's election were Donald A. Quarles, the new

€ Assistant Secretary of Defense for Research and Development (and soon after Secretary of the Air Force), and Trevor Gardner, whose title was Special Assistant to the Secretary of the Air Force. Quarles had been president of the Sandia Corporation, where the final engineering of nuclear weapons is done; he was intelligent, conservative, cautious, and unflappable. Gardner was intelligent, vigorous, somewhat volatile, and impatient to make changes quickly.

Each of these two men in his own style was responsible for first initiating and then guiding the thoroughgoing reviews of the defense research and development program that soon followed the change in administration. The reviews themselves were carried out at first by a number of committees working separately in each of the military services and reporting to dif-


ferent levels of program management. These committees had a strongly interlocking membership, a feature of the advisory apparatus which then, as now, allowed information to travel both up and down within agencies and laterally between agencies and thus to pass over and around the various barriers of secrecy, propriety, and bureaucracy which would otherwise cripple technical progress. (This style of organization and operation, of course, also allows a few strong-minded individuals to dominate a broad part of the scene.)

After about a year, in the interest of making more rapid progress and more profound changes, the originally large number of committees were regrouped into a much smaller number and John von Neumann became the chairman of the most important of them. The "Von Neumann Committee" (under different formal names) advised the Secretary of the Air Force on the projects that were under the direct control of that service, and it advised the Secretary of Defense on all large military rocket programs, including those of the Army and the Navy. Von Neumann was extremely intelligent, and curious about everything. He looked like a cherub and sometimes acted like one; my three- and five-year-old daughters delighted in climbing on him when he came to call at the house. He was very powerful and productive in pure science and mathematics and at the same time had a remarkably strong streak of practicality. He was one of the earliest pioneers in the design and construction of large electronic computers, he developed a strong interest in the technology of nuclear and other weapons, and he made a number of elegant inventions in each of these fields. This combination of scientific ability and practicality gave him a credibility with military officers, engineers, industrialists, and scientists that no one else could match. He was the clearly dominant advisory figure in nuclear missilery at the time, and everyone took his statements about what could and should be done very seriously. Other key members of his committee were George Kistiakowsky and Jerome Wiesner (both


of whom later became Special Assistants to Presidents of the United States), Simon Ramo and Dean Wooldridge (both of whom soon left the committee to organize the Ramo-Wooldridge Corporation), and Charles A. Lindbergh. Darol Froman, from the Los Alamos Scientific Laboratory, and I, from the Livermore Laboratory, were also members and in an informal sense represented the nuclear-weapons laboratories. The committee also maintained very close working relationships with the key experts in other essential subfields of technology. Especially important among these was Stark Draper, the director of the Instrument Laboratory at M.I.T. Draper knows more by far about the science and technology of inertial-guidance systems than anyone else in the Western world. He is a great optimist, and, although he is usually right, he sometimes seems to know even more than is so.

Working very closely with Gardner, Quarles, and Air Force Generals James McCormack and Bernard A. Schriever, the Von Neumann Committee in a series of intensive meetings in 1953 and early 1954 concluded that:

1. The state of the art in the relevant branches of technology had reached the point where a practical rocket-powered ballistic missile capable of carrying a nuclear warhead intercontinental distances and delivering it with sufficient accuracy could be built.

2. The Soviets had a head start of some years in this field, and therefore the United States should initiate a program with the "highest national priority" to build such a missile (the Atlas).

3. A new style of program management and technical direction was needed to coordinate properly the disparate fields of technology involved and to conduct concurrently the development, production, and deployment phases of the program. - In response to this third recommendation the Air Force established the Western Development Division (WDD) of the Air Research and Development Command (ARDC), selected


then Brigadier General Schriever to command it, and gave him an unusual combination of authorities over the research, development, and procurement phases of the program. Schriever used his special authority brilliantly, and much of what we now take for granted in the methodology of systems development and systems management was pioneered under his direction. He was able to integrate a wide variety of persons and facilities effectively and with finesse. In contrast to that of the more typical project officer, his personal influence made a real and positive difference to the program. In addition, and at least as important, the Air Force gave a contract to the then spanking-new Ramo-Wooldridge Corporation to provide what the weapons trade now refers to as general systems engineering and technical direction, or, more commonly, GSETD. At first this contract covered only the Atlas ICBM, but it later was extended to include all of the large Air Force ballistic missiles. The Ramo-Wooldridge organization supplied the scientific and technical expertise needed to complement and supplement the expertise in program administration and military matters which Schriever had directly on his staff. Ramo and Wooldridge assembled with remarkable speed a very strong group of scientists and engineers and did another excellent job.

Ramo and Wooldridge themselves had been members of the earlier versions of the Von Neumann Committee, and it was no coincidence that their new corporation was given this GSETD assignment as its very first job. There was absolutely nothing improper or wrong about this arrangement, but it does indicate that the "arm's-length" relationship between government and business which is generally thought of as being part of the free-enterprise system was even then missing from what years later came to be known as the "military-industrial complex." The Ramo- Wooldridge Corporation at first was strictly limited to planning and technical management and was prohibited from working on any hardware development or pro-


duction. However, the corporation later merged with Thompson Products (which always had been in hardware) and now operates as a major, well- known general defense contractor under the name TRW, Inc. One of its subsidiaries, TRW Systems, still performs general systems engineering and technical direction under contract to the Air Force on the Minuteman ICBM, one of the later missile programs which I will discuss shortly. (In 1960, a new, nonprofit corporation, the Aerospace Corporation, was established to provide GSETD for most Air Force missile and space programs. Many of the key personnel, including the senior vice-president technical, Allen Donovan, transferred from TRW to the new corporation in order to assure the smoothest possible transition.)

History can't be run by twice in order to check out alternatives, but I believe that without either WDD and General Schriever or the Ramo- Wooldridge group, the Atlas project would have taken more than a year longer to complete and would have cost much more as a consequence.

After the decision to go ahead with the development of large long-range rockets, or ICBMs, and after this new organizational structure came into being, the Von Neumann Committee, then working very closely with both WDD and the Ramo-Wooldridge Corporation, focused its attention on the technical details of the system. By an iterative process, the committee established both the performance goals (or "specs") and a first rough outline of a design which would meet them.

The performance goals set for the Atlas, the first ICBM to be started, were the following: a one-megaton-warhead explosive yield, 5,500 nautical miles range, and five miles or better accuracy, or CEP ("circular error probable," the distance from the target within which, on the average, half of the warheads (all). The accuracy specified for this missile thus called for the warheads to land within a distance from the target which was only one tenth of one percent of the range to the target. In this sense, the accuracy required was twenty times that actually


achieved by the V-2 ten years earlier. I emphasize this matter of accuracy because steady improvements in it have had a more profound effect on the arms race than any other technological factor, save only the invention of the H-bomb.

It is most instructive to examine the origin of these numbers. Draper, always a technical optimist, said he could foresee much better CEPs than five miles, and we simply took that figure as a conservative estimate. The range figure of 5,500 nautical miles is not, as one might suppose, the distance between two given points, one in the U.S.A. and one in the U.S.S.R. (Such distances vary from five miles to eight thousand miles or so.) Rather, it is precisely one fourth of the earth's circumference. It was, however, a sensible figure to focus on, being easily typical of heartland-to-heartland distances. Further, at a range of 5,500 miles the range of a rocket depends very strongly on its initial speed, so only small fractional changes in this latter quantity enable the range to be varied from, say, 3,500 miles to eight thousand miles.

The origin of the remaining performance goal, the requirement of a one-megaton yield, is much more interesting to examine. It is true that the damage radius of a one-megaton bomb is more or less compatible with the accuracy prescribed for the missile. However, the accuracy goal itself was arbitrary; the definition of damage is quite imprecise, and the radius at which a given blast pressure is generated varies only very slowly with the yield (as the cube root, in fact). Thus, the damage radius of a ten-megaton bomb and a one-hundred-kiloton bomb could equally well be said to be compatible with the prescribed accuracy. So, why 1.0 megaton? The answer is because and only because one million is a particularly round number in our culture. We picked a one- megaton yield for the Atlas warhead for the same reason that everyone speaks of rich men as being millionaires and never as being ten-millionaires or one-hundred-thousandaires. It really was that mystical, and I was one of the mystics. Thus, the actual physical


size of the first Atlas warhead and the number of people it would kill were determined by the fact that human beings have two hands with five fingers each and therefore count by tens.

What if we had had six fingers on each hand and therefore counted by twelves instead of tens? As any school child who takes modern math knows, the number one-million in base twelve is fully three times as big as the number one-million in base ten. Thus, if evolution had given us six fingers on each hand, our first ICBM warhead would have had to be three times as big, the rockets to deliver them would have threatened the lives of up to three times as many human beings, and it would have taken one or two years longer to carry out their development program. Similarly, if we had had only four fingers, like some comic-strip characters, the first warheads and missiles would have been only one-fourth as large, we could have built them somewhat sooner, and the present overkill problem would not be nearly as serious as it is. The only funny thing about this story is that it is true. It really was that arbitrary, and, what's more, that same arbitrariness has stayed with us. The initiation of other, later missile programs, including the Minuteman and the Polaris, was in fact delayed until nuclear-weapons technology advanced to the point where a one- megaton warhead could be forecast for each of them too. To go further, but still to remain within the realm of the truly possible, there might never have been a large liquid-fueled Atlas at all if we had had four fingers and hence a base-eight number system. We might have gone directly to the smaller and simpler solid-propellant Minuteman, which eventually did replace the liquid missiles.

Since these performance goals were set, rationalizations for them have been worked out, and many people, including especially those in the military services, think that the yield and all the other numbers characterizing the system can be derived from complex mathematical formulae connecting explosive yield, damage radius, target vulnerability, and other numerically defined quantities.


This particular ease of arbitrariness I have been discussing is by no means unique. Defense planning is full of arbitrary figures and figurings that have been thoroughly rationalized only after the fact. The number of units of many types of equipment is almost as arbitrary; so are the total numbers of men in the various services; and hence so is the total defense budget itself. I would say that the defense budget is arbitrary by at least a factor of two. The fierce arguments that can break out over a cut of, say, five percent have their origins in the very great difficulties of making changes in large traditionbound systems and not in the fact that the numbers as they originally stood were correct in any absolute sense. Thus, the real reason that this year's defense budget is so and so many billion dollars is simply that last year's defense budget was so and so many billion, give or take about five percent. The same thing, of course, applies to last year's budget and the budget of the year before that. Thus the defense budget is not what it is for any absolute reason or because in any absolute sense the total cost of everything that is supposedly truly needed comes out to be precisely that amount, but rather it is the sum total of all the political influences that have been applied to it over a history of many years, and that have caused it to grow in the way that it has grown. (The foregoing, of course, excludes special events like the Vietnam War, but even at its height that conflict has absorbed considerably less than half of the defense dollars.)

A similar situation also prevails with respect to over-all manpower needs. I became convinced when I was in the Pentagon later in the "missile gap" days that the apparent approximately ten percent shortage of technical personnel had nothing to do with the fact that there was a real need for an absolute number of persons which happened by a remarkable coincidence to be just ten percent more than we then had. Rather, the number of technical people then working could generate and sell ideas that needed ten percent more people to accomplish them. If there had been twice as many scientists and engineers in our defense programs, there would still have


been about a ten percent shortage, and if there had been half as many, the shortage we would have seen would even so have been only about ten percent.

Getting back to the Atlas story, after setting the performance goals, the Von Neumann Committee, Ramo-Wooldridge, and the Western Development Division were then able to outline the basic design of the system. The committee first estimated that the "required" one-megaton explosion could be produced by a nuclear weapon which, together with the protective reentry nose cone, would weigh 3,500 pounds. (A year previously, an Air Force Science Advisory Board panel, also chaired by John von Neumann, and including Teller and me as members, had predicted that a warhead of that weight would soon become feasible.) With the fuels then available, and for a given design style, immutable laws of physics connected the payload weight with the total rocket gross weight, and this was determined to be in the neighborhood of 250,000 pounds. With this as a base, it could then be decided that two large booster engines plus one smaller sustainer engine could accomplish the task of propelling the missile. The booster engine selected was the large liquid-powered Navaho engine uprated to 150,000 pounds of thrust each, and the sustainer engine was to be of similar design but reduced in size and generating only sixty thousand pounds thrust. It was further determined that the guidance and control equipment should be of the inertial type but assisted by radar observations of the rocket made from the ground during the takeoff. Such observations helped to further refine the ability of the missile to know where it was, how fast it was going, and where it was headed. The airframe, most of which consisted of the tanks containing the propellants, was to be made of very thin steel so that it would be light enough to enable the rocket to reach the high velocity necessary for an intercontinental range. In fact, the airframe had to be so lightly constructed that it could not stand up on its own. It had to be continuously internally


pressurized like a large balloon in order to hold its own weight plus the warhead and the other equipment placed up forward. In over-all dimensions, the Atlas was over eighty feet long and ten feet in diameter and weighed about a quarter of a million pounds.

Under the general technical direction of Ramo-Wooldridge, Convair (later called General Dynamics/Astronautics) was the contractor selected to build the airframe and to manage the over-all assembly of the missile, Rocketdyne Division of North American Aviation was given the propulsion contract, General Electric/Syracuse was given the contract for the inertial-guidance and control subsystem, General Electric/Philadelphia was given the contract for the reentry nose cone, and Los Alamos was assigned the job of designing the nuclear warhead. I must note here with some personal chagrin that, while it had been the Livermore Laboratory that was brash enough to promise that a one-megaton warhead could be made in a small enough physical package, it was the Los Alamos Scientific Laboratory that was mature enough at the time to actually provide one.

The task of developing materials and shapes suitable for the reentry nose cone, so that it could withstand the extreme conditions encountered at reentry, was often held up as the most difficult of all the problems associated with ICBMs. The public press, including the news magazines, was full of vivid stories comparing the reentering missile warhead to a meteor. I felt then that this was a grossly overestimated problem. As it finally turned out, it was necessary to approach this problem with some care, but a number of quite different but adequate solutions were discovered. As a direct result of the exaggerated views of the difficulty of this problem, much more money was spent on this aspect of the program than was really necessary.

The Atlas program was very successful in most respects. It exceeded most of its performance goals, and it met the schedule laid down at the very beginning of the program. First


flight with all three engines going was in August, 1958, and resulted in a range of about 2,500 miles. By 1960 an Atlas launched out over the Atlantic reached a range of nine thousand miles, and in that same year the first units became operational. However, over-all reliability from countdown to launch to flight was never good, especially for the operational units. Perhaps only as few as twenty percent of them would have reached their targets in a real war situation during the first year or so of so-called operational readiness.

In 1955, immediately after the Atlas program was under way and had been given the "highest national priority," two additional missile programs were initiated by the Air Force. The responsibility for the development of the Thor and the Titan, as they were called, was placed under the same management combination, and both programs were given the very same "highest national priority."

The range goal of the Thor was set at only 1,500 miles, but its other performance goals were left to be the same as Atlas. To meet these goals it needed only one engine instead of three, its tankage could be constructed with heavier walls so that it could stand without internal pressurization, and its top speed would be enough slower so that the reentry problem would also be much simpler. It was to be a "technical fallout" of the Atlas program and therefore was to use Atlas components to the maximum extent possible. The Thor program was quite successful from a strictly technical point of view. In addition to being briefly deployed as a strategic weapon, Thors have been used as the main booster stage for launching a great many satellites into orbit.

Along with the fact that Thor was a simple, direct and therefore hopefully inexpensive "technical fallout" from the Atlas program, there were two other rationalizations given for building it. The first of these, which made considerable sense in the context of the time, was that we thought it could be deployed one or two years sooner than the Atlas, and therefore we could


perhaps eliminate some or all of the time advantage the Russians had because of their earlier start. Of course, since these missiles did not have a truly intercontinental range, it would be necessary to deploy them overseas close to the Soviet Union. However, with our worldwide system of alliances, that did not appear to be a serious problem, and in 1960 they were deployed in the United Kingdom. A second rationalization put forward by a number of influential persons at the time was simply that since the Russians were believed to be working on an intermediate-range ballistic missile (IRBM), we must have one, too. It turned out be correct that the Russians were working on an IRBM of their own, and they now do have a great many of them deployed within the Soviet Union. However, the Soviet situation and the American situation are by no means symmetrical insofar as the need for missiles in this range is concerned. The Soviet Union has a great many potential enemies within 1,500 miles of its borders, whereas at the time the United States had no enemies that near its borders. Furthermore, the United States had a great deal of war materiel in the form of tactical aircraft and many other kinds of equipment deployed within 1,500 miles of the Soviet Union, whereas the Soviet Union had no equipment of any kind deployed that near the United States. There was, therefore, no symmetry whatsoever as far as the possible need for such a weapon was concerned, and this argument really had no more intellectual content than the familiar household notion about keeping up with the Joneses no matter what they do.

Also in early 1955, the Air Force initiated the development of a third large liquid-fueled rocket, the Titan. Its performance goals were the same as those of the Atlas, and its over-all dimensions were very nearly the same, but its design differed in a fundamental way. In contrast to Atlas, Titan employed a straightforward two- stage design. In such a design, the first, or booster, stage accelerates the missile up to about half of the final velocity needed to reach the target. Then, with the


first-stage propellants exhausted, the engines and the tankage of the first stage drop off. At that point the second-stage engine starts up and the payload is accelerated on up to full velocity. This is the style of design used for all long-range rockets except the Atlas and allows a number of simplifications over the Atlas design. The Atlas employed a modified single-stage design in which the two large booster engines were dropped off part way along but the main propellant tanks were carried along the whole way. As a result, the airframe, which consisted very largely of the propellant tankage, had to be so lightly constructed that it could not stand up on its own. It had to be pressurized internally like a large steel balloon to support its own weight plus the weight of the payload. In the case of the Titan, and of all true multistage rockets, it is possible to use heavier construction for the tankage and hence it is not necessary to rely on internal pressurization to supply the needed rigidity and strength.

We had realized when we started the Atlas back in 1954 that the Titan design was the ideal one, but we had avoided it because it required that the second-stage engine be started at altitude. At that time, there was practically no experience with starting liquid-rocket engines under such conditions i.e., in a vacuum and perhaps in a state of free fall. The Atlas design, awkward as it otherwise was, allowed all engines to be started on the ground before launch, so at the time it seemed prudent to use it for the first approach. We anticipated even then that we would later start another program parallel to Atlas. The Titan project brought in a complete new set of contractors for all subsystems and therefore provided a backup not only for the Atlas as a whole, but, indeed, for any part of it. As it turned out, the altitude-start problem gave us no difficulty at all, and the development of Titan proceeded reasonably smoothly. The first Titan was placed in service in 1962, less than two years after the first Atlas. Eventually a later, much improved and somewhat larger version of Titan replaced both


the Atlases and the first Titans in our operational inventory. In retrospect, it is clear that our approach to ICBM design was unnecessarily conservative in the way it treated unknowns such as reentry and engine starts at altitude. If we had started with the Titan design in the first place in 1954, we could have done a somewhat better job on about the same time schedule. Such an approach would have cost much less, since no second large liquid-propellant rocket system would have been needed.

In 1957, the Air Force also started yet another ICBM development program, this time based on the use of solid propellants rather than liquid propellants as had been the case with the Atlas, Thor, and Titan. This new missile, called the Minuteman, eventually became the most widely deployed of all and, through the last half of the sixties, provided the main component of our strategic forces.

It had long been held that for most military purposes solid rockets were potentially superior to liquid rockets. They promised greater reliability, readiness, and ease of handling, but their relatively poorer performance in terms of net payload weight for a given total missile weight had prevented their use in long-range ICBM applications. However, by 1957, improvements in this performance factor, coupled with further reductions in the weight of one-megaton thermonuclear weapons, made this use appear to be feasible, and the Minuteman program was accordingly initiated. Even with these advances, three stages were required in order to achieve intercontinental range. The project was considered at first to be marginal, but it eventually worked out very well indeed. A Minuteman flew out of an operational underground silo in late 1961, and about year end 1962 the first Minutemen were deployed. The original version of Minuteman was approximately fifty-six feet long, had a diameter of six feet at its base, and weighed 65,000 pounds. The weight was actually determined by practical considerations involving the transportation of the fully assembled rocket over the interstate highway system. The explosive yield


of its warhead was roughly the magic one megaton. The first model of this missile had a less than adequate range, but eventually the range was increased to about 6,500 miles.

The improvement in rocket performance that made intercontinental- range solid rockets feasible did not come from any so-called breakthrough in fuel performance; rather, it came from advances in the much more mundane technology of rocket-case design. In the early fifties and before, the cases containing the mixtures of solid rocket fuels had weights equal to twenty percent and more of the weight of the fuels themselves. Under such circumstances, intercontinental rockets cannot be built without an impractically large number of stages. In the late fifties techniques developed in the fiber-glass industry were adapted for making cases that would withstand the high temperatures and pressures of the burning fuels yet had weights equal to only ten percent of the fuels they contained. As is all too often the situation in the American aerospace complex, a great many people and organizations were working on the exotic problem of developing high-performance fuels, and relatively few were working on the much more important but less sexy problem of case design.

During the period in which the Air Force had been developing its missiles, the Army and the Navy had not been inactive. Determined not to stand aside and let the junior service monopolize the most crucial defense efforts of the day, they were quick to advance their own projects.

The Army's entry in the big-rocket derby was the Jupiter missile. The story of the development and the demise of the Jupiter is a textbook example of how personal determination and zeal, combined with interservice rivalry, can fuel the arms race and result in the production and the deployment of needless weapons and in the needless expenditure of billions of dollars. For all practical purposes, the Jupiter was identical to the Thor. It had the same range, the same-size warhead, the


same accuracy and even the same engines. With only minor variations (its diameter was specially selected so that it could pass through Swiss railway tunnels!), it had the same physical dimensions and it was built on the same schedule. It was developed by an Army organization under the command of General John Bruce Medaris and under the technical direction of Wernher von Braun. It was officially started a few months later than the Thor, although the people who promoted it had evidently been thinking about it for quite some time.

It seems clear, from various historical accounts of the origins of the space program, that perhaps the most important driving force behind the Jupiter program was the same as one of the major reasons for the existence of the V-2: Von Braun's abiding interest in space flight. In Germany at the time of the V-2 development, it was not possible to sell large space programs to Hitler's government, but it was possible to sell terror weapons. In the United States in the early fifties, a modest space program had been authorized as part of the U.S. contribution to the International Geophysical Year. The Von Braun group submitted a proposal for launching a satellite as part of the program, but it lost out to a Navy group in what seems to have been a fair competition. It was not possible to sell yet another space program to our government, but it was possible to sell weapons of mass destruction to the military authorities. And, since long-range ballistic missiles are virtually identical with space boosters except for the payload, Von Braun was "practical" and sold these rockets which really had one purpose on the basis of another.

Of course, many rationalizations for the Jupiter program were offered. These included a purported need for a land mobile tactical weapon having a range of fifteen hundred miles and an explosive yield of a megaton, although to my knowledge no neutral study then revealed the need for anything remotely similar. Another rationalization was that this system could serve as a backup to the Thor, although, since the Thor


itself was justified principally as a temporary stand-in for the Atlas, it is difficult to see now how this could have been taken very seriously, even in the context of the missile race as we then understood it. A third rationalization was that the Jupiter could serve as an intermediate-range missile to be launched by ships or submarines stationed in the open seas near the Eurasian land mass. The Navy did have, then as now, a role in strategic warfare, and, as a result, a joint Army-Navy group whose purpose was to adapt the Jupiter missile for shipboard and possible submarine use was established in November, 1955. The Navy never really liked this arrangement, since it involved having another service build a missile for its use. (The same thing was seen again years later in the TFX controversy, when Secretary McNamara tried to arrange for the design of a single aircraft suitable for both Air Force and Navy use.) Furthermore, the Navy took a very dim view of the idea of placing large liquid-fueled missiles on board ships and, even worse, on board submarines. These missiles, as then designed, had to be fueled with large amounts of liquid oxygen at the last minute, and the problems associated with this seemed to be extremely difficult.

Therefore, right from the start the Navy was on the lookout for some way out of this arrangement. The steady march of technical progress soon provided it with one. The Livermore Laboratory predicted that the package that would produce a one-megaton explosive yield could be further reduced in size. At the same time, important improvements in the performance characteristic of solid-propellant rockets were being made at Aerojet and other industrial laboratories. A special group under the leadership of Admiral William F. Raborn reviewed these advances and concluded that it had become possible to build what we now call the Polaris system.

The original Polaris was a solid-fueled rocket capable of throwing a new one-megaton warhead a distance of somewhat more than a thousand miles. It was 28.5 feet long and 54


inches in diameter and weighed 28,000 pounds. It was sized so that sixteen Polaris rockets could be fitted aboard a nuclear-powered submarine. The potentially nasty problem of handling liquid fuels, especially liquid oxygen, was completely eliminated. The joint Army- Navy committee was dissolved, and the Special Project Office, which performed brilliantly under the leadership of Admiral Raborn and then Captain Levering Smith, was set up and took over the program, now entirely within the jurisdiction of the Navy. The Polaris program itself proceeded very smoothly, and the first Polaris missiles were deployed at sea aboard the George Washington in November of 1960 In so doing, and despite its late start, the Polaris program became the first of our strategic missile programs to make a substantial contribution to our strategic power.

Soon after the initiation of the Polaris program, word got around that Secretary of Defense Charles E. Wilson did not consider the remaining rationalization sufficient justification for the existence of two nearly identical liquid-fueled IRBM programs, and that therefore one of them, probably the Jupiter, which by then had become a backup to a stopgap, would have to go. Medaris and Von Braun got wind of this potential calamity in time, and, with Secretary of the Army Wilber M. Brucker as a more than willing ally, they were able to ward it off, though not completely. As a result of their efforts, the Jupiter program was allowed to continue, but the operational responsibility for all missiles in the intermediate and intercontinental ranges was given to the Air Force. While not a very desirable solution to the problem from the Army point of view, this did have the virtue of keeping the program alive a little longer. The Jupiter was eventually deployed as an IRBM weapon in Turkey and Italy. There, nestled in the left armpits of the Russians, so to speak, it probably helped to inspire them to deploy some of their intermediate-range ballistic missiles in a similar fashion in a corresponding location, namely, the island of Cuba.


There was some useful technological fallout from the Jupiter program. This included the successful development and testing of the cheaper ablative type of reentry nose cone, and the development of a quite different rocket system, known as Jupiter C, which was later used to launch the first United States satellites. In addition, it kept the Von Braun team employed during what would otherwise have been some lean years. Even so, it seems to me that justifying the program because of these secondary benefits is like justifying five years' study of Latin so as to more clearly understand the difference between direct and indirect objects of verbs in English. There clearly were better ways of achieving the same objectives.

Altogether, during the fifties, we had initiated six crash programs to develop long-range nuclear strategic missiles. In retrospect, it is clear that three would have been sufficient.

Which of the three would have sufficed? Why did we overreact? What were the consequences of the overreaction?

A sufficient program would have consisted of the Titan plus the Minuteman plus the Polaris. Excessive technical conservatism led us to build both the Atlas and the Titan. The two offered the same general performance characteristics in terms of payload, range, and accuracy. We knew that the Titan design was intrinsically superior, but, because we had doubts about solving certain problems inherent in it, we started with the less risky Atlas first and did not take up the Titan until about a year later. As it turned out, both worked, but only one was needed. To be sure, hindsight is better than foresight, and probably we did the best we could under the circumstances at the time. But, even so, there is a valuable lesson here to be remembered the next time this sort of problem arises.

I regard both Thor and Jupiter to have been unnecessary. As I related above, the main justification for Thor was that it could serve as a stopgap for some period of time before the


more complicated and difficult Atlas ICBM could be deployed. As things turned out, it never did so in any strategically significant sense. The same arguments apply, with greater force, to the Jupiter. Both of these rockets did later serve as booster stages for space- vehicle launchers for a time, but other alternatives were available which would have cost much less in the long run.

As a result, we spent about twice as much money and we employed about twice as many people on these development programs as we should have. Furthermore, from the point of view of military security such excesses were harmful because they caused us to stretch our resources thinner than was really necessary. The early operational versions of these missiles all had a very much lower reliability than was predicted. They were therefore rather ineffective as strategic weapons for the first several years after their deployment began. It is quite likely that if we had been able to concentrate our top talent on fewer programs, this situation would not have been quite as bad. The same thing can be said about other problems and "bugs" that took time to fix. All in all, I believe the three sufficient programs could have been accomplished significantly sooner than they were if they had not had to compete with the three excess programs.

From the point of view of arms control and the arms race, these excesses in dollars and people also had serious consequences. The extra organizations and the extra people resulted in a larger constituency favoring weapons development. This larger constituency in turn strengthened those forces in the Congress "which hear the farthest drum before the cry of a hungry child," and consequently the whole arms race spiraled faster than before. Many of the leaders within this overexpanded missile industry correctly foresaw that they would be in trouble when all of these concurrent crash development programs finally resulted in some deployed hardware. They rightly anticipated that any follow-on develop-


ments would have a very hard time competing with the even larger funds needed for such deployments, and they provided some of the most strident voices among those proclaiming the "missile gap" of the 1958-60 period, but that is the subject of a later chapter.

In addition to all these missile programs, there were under way in the United States in the immediate pre-Sputnik era three entirely separate programs whose objectives were orbiting artificial satellites.

One of these was the Vanguard program, under the general supervision of the National Academy of Sciences. Its objective was the orbiting of a small artificial satellite as part of the United States contribution to the IGY, or International Geophysical Year. This rather special year was eighteen months long, extending from mid-1957 through 1958. The Navy, which was assigned the specific responsibility for developing the launch vehicle, was told to make it no bigger than absolutely necessary and to avoid interfering with any military-rocket programs. These two requirements together almost foredoomed it to failure, though it did eventually successfully launch three very small payloads weighing three, twenty-one, and one hundred pounds.

The second U. S. satellite program then under way was under the direction of the Air Force. It was actually a collection of a number of projects with different specific military objectives. In the immediate post- Sputnik days we usually do. scribed these as "communications relay," "navigation aids," "weather," "reconnaissance," "early warning," etc. From the first the payloads involved were scheduled to weigh over one thousand pounds, and eventually they in fact weighed many thousands of pounds. They were scheduled for first flight is 1959, when the IGY would be over.

These Air Force satellites were to be launched into space with large rockets consisting of either Thor or Atlas as the first


stage plus a special upper stage called Agena. The Agena was built by Lockheed on an Air Force contract let in 1956. It went through a series of modifications during its service as the workhorse of the military space program. On February 28, 1959, a Thor-Agena combination injected a 1,450-pound satellite into orbit.

The third pre-Sputnik satellite program was bootlegged by the Army. The Von Braun group had earlier submitted a proposal for a rocket for launching the IGY satellite to the committee duly charged with launcher selection. In what I understand to have been a fair competition, the winner was the Navy's Vanguard proposal. However, the Medaris-Von Braun group was not one to be stopped by a mere decision of higher authority, and they went ahead and designed a new satellite launcher which they named the Jupiter C. Writing in 1960 about the controversy over this matter, Medaris recalled that, after a successful cancer operation in 1956, he developed "the ultimate conviction that the Lord still had work for me to do or my life would not have been spared."

This Jupiter C was not really a Jupiter; rather, it was a Redstone plus upper stages consisting of clusters of small solid rockets. Its ostensible purpose was testing nose-cone materials for Jupiter, but the actual velocity attained (and not accidentally) was more nearly that of an Atlas, the development of which was the sole responsibility of the Air Force. Even on its very first launch, it carried an additional dummy stage, "filled with sand instead of power," which if properly filled and fired could have been used to send it into orbit well in advance of Sputnik and the IGY. According to Medaris in his book Countdown for Decision, it was given the name Jupiter C so that it could benefit from the high priority the real Jupiter had. On January 31, 1958, Jupiter C was used to launch America's first satellite, the Explorer I, weighing just thirty-one pounds, into orbit.

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

British SSBNs

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

LearnWorld Explained FAQ Bosnia War Theory Templates

Our Home Page Our Home

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