Edward Teller
Edward Teller; drawing by David Levine


The Geneva summit has come and gone, leaving Mr. Gorbachev adamant that the Strategic Defense Initiative program is a critical impediment to any significant nuclear arms control agreement—for the simple reason that it would inevitably drive the arms race into space. President Reagan on the other hand, remains bewitched by what he continues to call his dream, a dream of a shield of defense systems in space which would liberate mankind from “the prison of mutual terror.” So there it is—as the USSR sees it, a choice between survival and mutual suicide; for Mr. Reagan, a beautiful dream. Where does reason lie? Will there be anything new at the summit later this year?

Had anyone other than the American president ever invited scientists to try to render “nuclear weapons impotent and obsolete,” the suggestion would probably have attracted no more attention than had they been asked to square the circle or solve the problem of perpetual motion. But it happened to be the President, and he spelled out his vision of a future over which the nuclear bomb no longer casts a shadow in such homely terms that it all sounded real. How could the message fail to appeal? There was also a promise of vast resources for R&D—a vision therefore not only of peace but, at least in the meantime, of work, prosperity, and excitement for some. For those who might object that the idea was strategically naive, the President even acknowledged that it would “take years, probably decades of effort” for the dream to become a reality, and that in the meantime defensive systems, “if paired with offensive systems,” could be “viewed as fostering an aggressive policy.”1 However fantastic it was, the challenge therefore had to be taken seriously, even by the President’s defense secretary who, it had been widely rumored, had been skeptical about the idea until the moment it was suddenly proclaimed to the world.

The upshot is that within the space of two years, SDI has become one of the best-known acronyms in the world. It has stimulated a global debate. Instead of reducing tensions between East and West and “introducing greater stability into the strategic calculus of both sides,” it has exacerbated the tensions. It has also generated strains in the Western alliance. Even more important, it has divided that part of the American scientific community to which the challenge was particularly addressed, with respect both to its technological implications and to its strategic desirability—a part of the debate in which politicians, military people, and ordinary citizens have also engaged. And of course the debate has produced a mountain of comment, including books such as the three under review.

In some respects the debate is a rerun of the controversies that culminated in the 1972 ABM Treaty, when both sides implicitly acknowledged that it was then beyond their power to design meaningful defenses against intercontinental ballistic missiles (ICBMs). Both had set about the job in the same way, just as they deployed the same variety of anti-aircraft defenses. There were “acquisition radars” which scanned the horizon for incoming warheads; “tracking radars” linked by computer to nuclear antimissile missiles whose explosion outside the atmosphere would emit X-rays to which the attacking warheads would in theory be vulnerable at great distances; and then there were terminal radars and terminal anti-missile missiles to deal with such warheads as would not have been destroyed. By the late Sixties enough hardware and computer links had been developed to justify deployment, or so it seemed.

But doubts had already started to set in.2 Could an ABM system work? It would have to deal not only with nuclear warheads but with a variety of decoys and other “penetration aids” which the missiles would release in order to confuse the radars. Warheads might be exploded outside the atmosphere to create an electromagnetic blackout that would make the task of the radars almost impossible. The large radars themselves were clearly vulnerable to direct attack. The scale of an attack could itself be so great as to swamp any defensive system. Each ballistic missile could carry not one but several warheads which, as was clearly recognized as early as the mid-Sixties, could be made independently maneuverable—what we now call MIR Ved.3 And then there was a political problem—people did not want defensive nuclear missiles planted in their back yards. Finally, neither the US nor the USSR could afford to deploy more than a handful of defensive complexes. If these could be made to function effectively, which was the first question that needed an answer, there was then a second problem, who or what was to be defended.

Despite all the doubts, in 1967 the United States started deploying a “light” ABM system, code-named Sentinel, to defend against a possible missile attack from China. The USSR had started a few years before to deploy one for the presumed defense of Moscow. For, as Mr. Khrushchev saw it, if his ballisticians knew how to “hit a fly in the sky,” so too they could hit incoming warheads. It was therefore only rational to try to defend his capital city. President Johnson was not so sure. In 1967 he asked the only question that mattered: Would an ABM system work? The answer from those best qualified to judge was “no.”4 No ABM system could reduce significantly the vulnerability of the United States; no president could initiate or agree to the initiation of a nuclear exchange without realizing that once it had begun, he could never be sure where it would end—that the risk, were he ever to agree to the actual use of nuclear weapons, was the total devastation of his country. In 1967 President Johnson and Robert McNamara, his defense secretary, tried hard at Glassboro, New Jersey, to persuade Kosygin, Khrushchev’s successor, to accept these propositions. Gradually he and the Politburo saw the light. Dubious ABM systems only destabilized a state of mutual nuclear deterrence.


The result was the ABM Treaty of 1972, a treaty that limited ABM deployment to two sites only—later changed to one—in each country. The treaty did not bar development work that improved the radars, computers, and defensive missiles deployed within the two sites, but specifically prohibited the development of any type of space-based ABM system. Stability was then the order of the political day.

And that was the moment—not March of 1983 when President Reagan spoke—when SDI really began. For, not surprisingly, the American and Soviet scientists and engineers who had been working on lasers and particle beams as possible Ballistic Missile Defense (BMD) weapons did not cease their experimental inquiries when the 1972 treaty was signed, any more than did the scientists and engineers who were trying to improve the power of the permitted radars and computers, and the design, thrust, and speed of their defending missiles. The military chiefs on both sides, who had anyhow been dubious about the wisdom of the ABM Treaty, were only too ready to encourage them to continue, however little they understood the intricacies of the systems concerned. Most of the scientists and engineers needed little urging. After all, it was their jobs that were on the line.

An important figure who was in no need of any encouragement was Edward Teller, the well-known refugee theoretical nuclear physicist who had worked on the atom bomb under Hans Bethe during the war years. Teller is regarded by some as a distinguished, by others as a notorious, physicist. During the McCarthy years he had played a critical part in the downfall of Robert Oppenheimer, the wartime scientific director of Los Alamos, whether because of jealousy and frustration or because he had conceived of himself as some kind of superpatriot—plus royal que le Roi—it was difficult to say. Whatever his motives, Teller lost the respect of most of his scientific peers, from whom he rapidly became isolated.5 On the other hand he was eagerly supported by members of the defense establishment, particularly in the Air Force, who were only too ready to agree that a more powerful nuclear device than the atom bomb, the “second generation” hydrogen bomb, would be a valuable addition to America’s nuclear arsenal.

They also supported him in his campaign to found a second nuclear warhead laboratory at Livermore as an offshoot of the University of California. Teller had persuaded them that the Los Alamos research center was too liberal. He vehemently opposed the Partial Test Ban Treaty of 1963, basically because it interfered with the testing of new warhead designs, but protesting too that the Soviet Union would be bound to cheat—and that anyhow there was no reason to suppose that the radioactive fallout from nuclear tests in the atmosphere did any harm, it might even do good.6 He became the chosen scientific mouthpiece of the “hard-line right,” a term that Europeans have come to identify with those Americans who are intrinsically against arms control, who uncritically assume that more destructive nuclear power than what already exists means more military and political strength, and who, whatever the risks, wish to oppose the Russians and communism at all times and wherever possible.

Teller was also loud in his protestations against the ABM Treaty and against SALT I and II. The Livermore laboratory, his creation, was going to give birth to a third-generation nuclear device that would transform the entire strategic scene. According to William Broad, the author of Star Warriors, the picture of this third generation of nuclear devices that Teller painted for the President was largely instrumental in instilling in Mr. Reagan’s mind a vision of a future in which nuclear weapons could be made impotent and obsolete.


Teller thus lurks behind almost every page of Mr. Broad’s book, which focuses on a small but select group of the employees of Livermore, who now number, so we are told, some eight thousand, and who cost the federal government more than $800 million a year. Although Livermore does many other things, its primary function is the design of warheads, a field in which it competes fiercely, and presumably very successfully, with the older Los Alamos laboratory. A glossy brochure that was issued to celebrate the station’s silver anniversary claimed that Livermore was responsible for nine of the ten strategic warheads now in the American nuclear stockpile. As Mr. Broad was told by a member of the special group with whom he spent a week in the Livermore compound, warhead and weapon designers are free to follow their heads—the number of possible designs is “limited only by one’s creativity.” The young men Mr. Broad was getting to know were the ones who were responsible for Teller’s third-generation nuclear breakthrough.

Their leader, and Teller’s main disciple, is Dr. Lowell Wood, now aged forty-two. For a week Mr. Broad stayed with him, consorting during all hours of the day and night with his host’s team, which was designated O Group at Livermore, and which numbered no more than a dozen or so young scientists of average age less than thirty. Associated with them were as many part-time workers, some of whom were no more than graduate students. Many of the team had begun as research fellows of the Hertz Foundation, on whose board both Teller and Wood sat, and for which Wood served as the recruiting sergeant. With employment prospects bleak, and competition for jobs fierce, he was able to select from all the universities of the US young scientists and engineers in whom he discerned “outstanding capability that has been developed and exercised in some direction”—usually in mathematics or physics. Apparently men with general interests but no specialized technical accomplishment were not wanted.

Successful candidates were invited to work at Livermore for a summer in an intern program, and were kept on if they made the grade. All but a few of the group were, like Wood himself, bachelors. Few had set out to be bomb makers; but it was either that or, as one of the group told Mr. Broad, working in a beet factory. There was the further attraction that Livermore had the most marvelous equipment with which to work, as well as access to the underground nuclear testing grounds of Nevada, which were shared—in effect as an outstation—with Los Alamos and the Sandia nuclear development establishment at Albuquerque.

Lowell Wood’s young men both collaborated and competed with one another, and celebrated their triumphs at parties at which they ate masses of ice cream and drank gallons of Coca-Cola. Mr. Broad tells us that there were no women around and that O Group was not entirely popular in the main Livermore establishment, one member of which told Mr. Broad that the team was made up of “bright young hotshots” with “no outside interests…who are socially maladjusted.”

If the week that Mr. Broad spent with the hotshots was typical, they also seemed to converse only with one another, and when not discussing their work, exchanged naive views about politics. One would imagine from the conversations Mr. Broad describes that the only problem in the world for O Group is the competition for power between the USSR and the US. Their part of the problem was to construct a shield to keep out Soviet warheads. One of the group told Mr. Broad that as soon as that was done, the US would leave the USSR technologically “in the dust,” and that success would “prove to the world that democracy works.” Another told him that if the Russians “owned the planet” they would not allow the evolution of technology to continue. So far as this young man knew, “the only reason they are going with technology is that they can’t afford not to.” He clearly was unaware that in the 1930s the USSR had shocked the West with a revelation of a totally utilitarian view of science and of its absolute commitment to technology. As propounded by B. Hessen, the Russian ethos holds that science cannot advance in a society which restricts technological advance, that “science develops out of production, and those social forms which become fetters upon productive forces likewise become fetters upon science.”7 I imagine that this proposition would have appealed to Lowell Wood and his team. They are doers, not philosophers or political scientists. Their business, like that of their opposite numbers in the USSR, is to put scientific knowledge to work.


Long before any of them was born, long before the era of ICBMs, physicists had been building machines—for example, cyclotrons and proton synchrotons—in which the subatomic particles that make up the atom are accelerated into extremely powerful beams of energy. These “particle beams,” if directed into space, might, it was later thought, intercept and destroy nuclear warheads. Then, in 1960, came the laser. Ordinary light, as emitted by the heated filament of a light bulb, consists of an incoherent emission of a very wide band of electromagnetic waves—from the longer ones at the red end of the visual spectrum to the shorter ones at the blue. The laser focuses all the energy of a very narrow band of the electromagnetic spectrum within a coherent beam or jet. The discovery8 was seized upon by scientists the world over for a myriad of different purposes, from an instrument that can be used for operations on the retina of the eye, to an aiming device for marksmen.

It was not surprising that “defense scientists” also saw in the laser, as in the particle beam, a device which, if furnished with sufficient energy, could operate at great distance—the sort of thing an older generation would have called a death ray. Retired generals started to talk about particle beams as though they were particles which could be poured from one hand to another. The newspapers were not slow to hint at a new generation of wonderful weapons.

The main achievement of O Group, and in particular of Peter Hagelstein, whom Mr. Broad introduces to his readers as the brightest star of Lowell Wood’s team and as a young and troubled engineer who is also interested in classical music and French literature, was the presumed invention of the “nuclear-pumped” X-ray laser. Other workers, including an older Livermore scientist, had also bent their talents to this problem, but in vain. X-rays belong to the extreme shortwave end of the electromagnetic spectrum (about one thousandth the wavelength of visible light). If a coherent beam of X-rays could be provided with sufficient energy, it would travel outside the atmosphere at the speed of light for thousands of kilometers, imparting its energy to the “first fraction of a millimeter of the aluminum skin of a missile [in its path]. This paperthin layer would explode, sending a shockwave [‘thump’] through the missile,” so destroying it.9 This is the concept that was Teller’s basic justification for believing that a space-based ABM system was a possibility. A sufficiently powerful X-ray or other laser or particle beam traveling at the speed of light, that is to say at 186,300 miles a second, could, if properly aimed, destroy a warhead whose maximum speed was less than ten miles a second.

Were an X-ray laser to serve as an ABM weapon, it would, however, be necessary to use as a source or “pump” of energy a nuclear device, i.e., bomb, of significant force (maybe 100 kilotons in yield or more). On the other hand, in theory the X-ray laser is not the only laser that could do the trick. Los Alamos, among other laboratories, is working on an “excimer” or chemical laser whose wave-length, although much longer than those of X-rays, would be equally effective (but by heating, not “thumping,” the target), without the disadvantage that X-rays could be made to lase only at the enormous temperatures associated with the explosion of a nuclear weapon.

Since X-rays are absorbed by even a thin layer of the atmosphere, another disadvantage of the nuclear-pumped X-ray laser is that it is a device which in practice could only be effectively fired when shot up into space, or shot from a space satellite, a so-called space battle station—which indeed would be necessary for most subatomic particle-beam weapons. An X-ray space battle station would, of course, be a one-shot device, since the whole thing would be destroyed an infinitesimal fraction of a second after the nuclear explosion that generates X-rays, which would be directed along, and amplified by, a series of lasing metal rods built around the whole device. Given certain conditions, the rods could in theory be independently aimed in that millionth of a second at a number of enemy launchers as they rose from their silos.

Only land-based weaponry was involved in the ABM systems with which the 1972 agreement was concerned. There was no possibility then of hitting ballistic missiles during their launch phase; since decoys and other countermeasures ruled out effective interception in space, warheads would have become vulnerable only when they reentered the atmosphere on the way to their automatically designated targets.

The 1983 system, if SDI can be called that, differs completely because it is a space-based concept. The theory is that beam weapons or rocket fire could be directed from artificial satellites against enemy missiles during the few minutes of their launch phase, before the ejection of their multiple warheads, and thousands of miles from the targets which they would be programmed to destroy. The same arguments would apply to the electromagnetic rail-gun, another device now being worked on, which uses intense magnetic fields to create the force to shoot out small projectiles (“smart rocks”) at very high velocity.

Were it ever possible to bring laser, particle-beam, or electromagnetic rail-gun weapons into action during this initial phase of the flight of a missile, the defensive system would also have to include enough artificial surveillance satellites to ensure that as they circled the globe, there would at all times be at least one that was looking down on the Soviet missile fields. Otherwise the curvature of the earth would make it impossible for one or the other side to see its opponent’s missiles before their warheads were well into space. The weapons on the “battle stations” circling the earth would have to be ready to be aimed and to strike on automatic command.

But here lies the first major problem. Teller, who we have been led to believe started the whole thing, is convinced that battle stations permanently in space are too vulnerable to enemy attack to be contemplated. Even if, as Lowell Wood suggested to Mr. Broad, they were placed in geosynchronous orbit more than 20,000 miles above the earth, they could in theory be “fooled”—for example, by decoy launches on the ground or by decoys in space furnished with transmitters to send out false signals to confuse the BMD sensor systems.10 Or they could also be destroyed by space mines, small satellites that would follow the battle stations and would always be ready to explode.

Space-based attacking systems also suffer from an additional handicap—the power sources by which they would be activated would be both very heavy and very bulky. Teller’s view is that the X-ray laser, his favorite weapon, should be carried in submarines, and launched into space—“popped-up” is the happy-go-lucky term that is used—by ballistic missiles which would react automatically when commanded to do so by the surveillance satellites that registered the Soviet SS18 and other missiles rising from their silos or launch pads.

Once shot into space, the X-ray laser devices would automatically be focused onto the presumably unprotected boosters, which, as they rose above the atmosphere, would be “thumped” by an X-ray laser beam set off by the explosion of a hydrogen bomb. Excimer or chemical lasers on the ground might in theory reach their targets by way of a system of folded mirrors that would be orbiting the earth, ready to open up on computer command to reveal themselves as perfect large reflecting surfaces. These would change their orientation in split second after split second as they aimed the beams impinging on them either directly to their targets, or redirected them to other mirrors that would do the focusing.

Then there would be a computer network that would tie all the surveillance satellites, targeting devices, beam and ray weapons into a single system competent to deal not with one or a few enemy missiles but, if the space shield were to be truly impregnable, with hundreds, even thousands.

What all this means is that if it ever came to action, heaven would become hell within a few minutes, and, given a failure of the system, that hell would also break out on earth in less than an hour. What is more, even though the whole system would have to start reacting automatically at a moment’s notice, somehow or other—no one has said how—there would have to be time for a human link in the chain of interacting processes. As a sop to the doubters, the proponents of SDI agree that the fate of mankind is not something that should be simply committed to a computer.

Teller, Lowell Wood and his whizkids, as well as their opposite numbers in Los Alamos and such supporters as they have in the Pentagon and the Department of Energy, believe that all this can be done, or at least that it is worth spending tens of billions of dollars to see whether it can be done. Little time passed, however, before it became clear that some members of Congress had doubts, and that the views of the space warriors were not shared by a number of scientists who know about these things, both within and outside government laboratories. Lowell Wood asserts that all the opposition emanates from a very few scientists. At a small international meeting, not mentioned by Mr. Broad in his fascinating book, Wood told his audience that the number of scientific skeptics could be counted on the fingers of one hand. Unfortunately he said on the fingers of a maimed hand, which rather shocked his audience and reduced the force of his argument.


In fact, the situation is the reverse of what Lowell Wood believes. According to Dr. John Bardeen, twice a Nobel Prize winner in physics, there are few scientists either within or outside the administration who believe that President Reagan’s dream could be realized in the foreseeable future. Dr. Bardeen was a member of the White House Science Council at the time of Mr. Reagan’s SDI speech, about which both the council and Dr. George keyworth, its chairman, were ignorant until five days before it was delivered.11 Teller too does not share Lowell Wood’s views about the number of scientific doubters. He told Mr. Broad that “a great many American scientists, perhaps the majority,” are against SDI. The fact is that only a very few independent scientists have come forward to offer their support to the Livermore and Los Alamos enthusiasts. Of these, the quickest off the mark was Dr. Robert Jastrow, a well-known popularizer of science, and a professor of earth sciences at Dartmouth College. His unswerving loyalty to SDI shines out in How to Make Nuclear Weapons Obsolete.

Dr. Jastrow’s short book begins with a number of fairly unassailable propositions. Defense, he tells us, is always a good thing; a policy of mutual nuclear deterrence is inhumane since it implies a willingness to destroy populations; if one side acquired an effective defense against ballistic missiles, it could attack the other with impunity; if both had a defense, nuclear arms would become useless; even an imperfect US defense that left some of its retaliatory nuclear weapons untouched would foreclose the possibility of a first strike by the USSR. Why the USSR should in any circumstances want to risk such a strike, knowing that the considerable submarine missile fleet of the United States would be immune to attack, Dr. Jastrow does not make clear. As former president Nixon has recently reminded us, the Soviet leaders are neither madmen nor fools.12

Dr. Jastrow then gives an account of the buildup of Soviet land-based missiles in the years since the signing of the SALT treaties, implying that doing so was contrary to what the treaties allowed. For Dr. Jastrow, the USSR has only one end in view, namely the destruction of the land-based components of the US nuclear arsenal in a first strike. Here Dr. Jastrow’s echo of the conventional Pentagon view again clashes with the position of Nixon, who in his recent article in Foreign Affairs observes that the Russians have gained whatever “superiority” they have “in strategic land-based missiles not because of what they did in violation of arms control agreements but because of what we [the US] did not do within the limits allowed by those agreements.”

Dr. Jastrow writes about the airborne and submarine elements of the US nuclear arsenal, including the Trident missile, in terms that rather belittle their value. He talks mysteriously of work going on which will make it possible to detect deeply submerged submarines. This is a possibility that has been continuously discussed and explored over the years, but so far with no results that would undermine the view that nuclear submarines are, and will continue to be, effectively invulnerable. The picture Jastrow paints seems to imply that America is wide open to attack by the more powerful armory of the USSR. The only real hope, therefore, is “a defense that shielded the American people.” And despite what the critics say, that, he asserts, is already available. The new secret weapon is the 670-million-mph laser beam. With this introduction Dr. Jastrow takes us back to SDI.

It turns out that he was so inspired by the President’s speech of March 1983 that he immediately and publicly gave it his scientific imprimatur. He then became fortified in his faith by a talk given by Dr. Keyworth, until recently Mr. Reagan’s science adviser who, Mr. Broad tells us, was recommended to the President by Teller. Dr. Keyworth is a former member of the staff of Los Alamos, outside which he was little known before, and is a friend of Teller. It would have been surprising if he had not been an ardent crusader for space defense.13

Much of the material for Dr. Jastrow’s book was provided by Gregory Canavan of Los Alamos, and by Lowell Wood of Livermore, by General James A. Abrahamson, the head of the Pentagon’s SDI office, and by a few other officials whom he names. The book contains no original analysis, which perhaps is not surprising since it would seem that Dr. Jastrow has not himself been involved in research either on nuclear weapons or on lasers, radars, or computers. He is a missionary for SDI.14 What the reader therefore gets is a highly optimistic account of the same hypothetical space defensive system of which countless descriptions have already been published.

Can “inventive genius,” Dr. Jastrow asks, find a device that can shield the American people? Of course it can. The invention is already there. “It is called the laser.” And the way Dr. Jastrow writes makes it all but child’s play to fit together the whole defensive complex. The US could deploy a Mark I system by the early 1990s15 and all for a cost of $60 billion, for which could be bought one hundred satellites, each carrying 150 interceptor rockets, four early-warning satellites in geosynchronous orbits, lower altitude satellites for surveillance, acquisition, tracking, and terminal defense, all the necessary but as yet nonexistent computer networks and other accessories. Everything can be “easily” achieved. Terms such as “easy” and “not too difficult” characterize Dr. Jastrow’s rosy picture.

His optimism is matched only by his breathtaking simplifications. War in space—that is to say, intercepting nuclear warheads with laser or particle beams or with pellets shot from electromagnetically driven rail-guns—is for him like an infantry battle. If the battle-management satellite loses touch with its weapons satellites, they can function autonomously—“like a machine-gunner cut off from his unit.” It would, however, be better, so he writes, were they under the control of the master satellite which, like the general in charge of a land battle, can oversee the whole operation, moving his forces as required. The control function would be exercised by a master satellite—not, it should be noted, by the president of the United States in consultation with the heads of NATO governments—during the three to five minutes of the boost phase of the enemy missiles, whose targets this time would not be hostile soldiers, but defenseless cities with millions of inhabitants in peril of instant death. It reads like a film script. I suspect that were Dr. Jastrow’s book to be made required reading for the leaders of America’s NATO allies, what reluctant political support some of them have been prevailed upon to give to President Reagan’s dream would vanish overnight.

Dr. Jastrow fully realizes that a large number of highly reputable American scientists regard the entire idea as technical and strategic nonsense. Yet almost the only point of criticism on which he concentrates relates, first, to an erroneous early estimate, in a report by scientists opposed to SDI, of the number of surveillance satellites that would have to orbit the earth in order to keep the Soviet missile fields constantly in view, and, second, to an estimate of the considerable weight of a satellite that would be demanded by a particle-beam weapon. Dr. Jastrow did not himself spot the errors. He says he learned about them when they were rumored by “professionals in the field.” In fact the authors of the report in question,16 which included such distinguished scientists as Hans Bethe, Richard Garwin, Victor Weisskopf, Kurt Gottfried, and Henry Kendall, themselves drew public attention to the two errors five weeks after their report was issued, and before anyone else had done so.17

They also made quite sure that their subsequent publications were free of computational errors, at the same time emphasizing that estimates of the numbers of surveillance and laser satellites that a defensive system might call for depended on a varying number of assumptions. Dr. Garwin has subsequently published what seems to be the most complete and unchallenged set of estimates, given several different assumptions.18 At any rate it is judged as such by Edward T. Gerry, 19 the chairman of the relevant panel of the Pentagon’s Fletcher study team,20 which the administration set up in 1983 to advise whether the pursuit of a space-based defensive system was technically justifiable.

In reality the two computational errors did not affect any substantive judgment about the feasibility of a space-based defense, as emerged clearly from a vigorous and lengthy exchange of letters published in Commentary in March 1985. Dr. Jastrow, who took part in the exchange, nonetheless again hammered away at the errors in an article published later in the summer,21 in which he went so far as to imply that the views of his critics about the efficacy of Soviet countermeasures should not be “accepted”—by which I sense he means they should be disregarded. (It should be noted that his present book appeared a year after the corrections had been made by Dr. Garwin in his testimony to the Senate Armed Services Committee, and that Jastrow makes no mention of that testimony.) Someone not competent to follow the technical nuances of the debate could be forgiven were he to assume that Dr. Jastrow’s apparent obsession with the long-corrected computational errors reflects a determination to discredit his critics personally.

Dr. Ashton Carter, the author of the first report on SDI to be prepared for Congress’s Office of Technology Assessment (OTA), is also the target of Dr. Jastrow’s criticisms. He too has pointed out22 that Dr. Jastrow has never provided his own analysis of the problem. It would be unfortunate if the analysis included such meaningless statements as Jastrow’s observation, on page 95 of his book, that one molecule of oxygen always consists of two oxygen molecules bound together. In truth, the precision of Dr. Jastrow’s style, as manifested in his book, compares poorly with the appearance of scientific exactitude of the papers in which he attacks his critics, and in which he quotes extensively from documents provided him by proponents of SDI at Los Alamos and Livermore. While the voice, like that of Jacob, is obviously Dr. Jastrow’s, his papers often read as though the hands of more than one Esau had helped steer his pen.

Dr. Carter’s report of April 1984 considered the technical ideas that were discussed by the Fletcher study team as possible ways for attacking enemy ballistic missiles during their brief boost phase. In preparing it, he was helped by every official organization that was concerned, including Los Alamos and Livermore, as well as the CIA. But the conclusions that he drew were his alone, and the main one was that

the prospect that emerging “Star Wars” technologies, when further developed, will provide a perfect or near-perfect defensive system…is so remote that it should not serve as the basis of public expectation or national policy about ballistic missile defense.

Not surprisingly, he was immediately set upon by the proponents of SDI in Los Alamos, Livermore, and the Defense Department—not to mention Dr. Jastrow.

Dr. Carter’s study had been commissioned by OTA at the request of the House Armed Services Committee and the Senate Foreign Relations Committee. In view of the debate that his report stimulated, OTA then undertook an even more extensive study under the scrutiny of an advisory panel, which included among its twenty-one members Michael May, associate director-at-large of Livermore; Robert Clem, the director of systems sciences of the Sandia National Laboratories; senior representatives of several of the major defense contractor companies who are, or who would be, involved in SDI work; General David Jones, the former chairman of the Joint Chiefs of Staff; Robert McNamara, former defense secretary; Gerard Smith, the chief negotiator of the 1972 ABM and the SALT treaties; Major General John Toomay, who had served on the Fletcher study team; as well as Richard Garwin, Sidney Drell, and Ashton Carter, three who have criticized SDI on technical grounds. It would be difficult to conceive of a more distinguished or better balanced group. They advised a project staff which, in addition to writers of the studies they commissioned and an administrative staff, included nine researchers.

So far as I can judge, the new and lengthy OTA report, Ballistic Missile Defense Technologies, and the summary report accompanying it, touch on every aspect of SDI that has been publicly debated, and they set out both sides of every point at issue. The authors and the advisory panel acknowledge that the USSR is “vigorously developing advanced technologies potentially applicable to BMD.” But at the same time, and contrary to the somewhat equivocal views put forward by the proponents of SDI in order to encourage public support, the OTA report does not consider that the Soviet Union has any lead over the US “in any of the 20 basic technologies that have the greatest potential for significantly improving military capabilities in the next 10 to 20 years.” (These were the technologies which were recently reported on in the annual report to Congress of the under-secretary of defense for research and engineering in the Pentagon.23

The OTA report reviews the requirements that an effective BMD system would have to meet in the face of the obvious Soviet countermeasures. The reader is also warned that it is essential to consider more than just the feasibility of a host of separate technical ideas. What matters is operational feasibility—could the developed technical components be combined into an “integrated, reliable system that could operate effectively and maintain that effectiveness over time as new countermeasures appeared.” The report reaches the same general conclusion that Ashton Carter did in his earlier appraisal—“assured survival of the US population appears impossible to achieve if the Soviets are determined to deny it to us.”

Press reports suggest that the Pentagon’s reaction to OTA’s new assessment has been less hostile than it was to Ashton Carter’s, and that the defense authorities agree that during the years that it would take to move to a defensive strategy, new risks of nuclear conflict might well arise. On the other hand, the head SDI office in Washington believes that even a partial defense would increase the USSR’s uncertainties were it ever to contemplate a first strike against the US, and would therefore enhance deterrence.24

But while administration and congressional leaders, as well as many press commentators, accept the OTA report as a nonpartisan review, which is the way it certainly reads, some die-hards have condemned it. What I find surprising is that they have now been joined by Dr. Frederick Seitz, the chairman of the Pentagon’s Defense Science Board. He and Dr. Jastrow recently proclaimed at a meeting of the conservative Heritage Foundation that all the members of the OTA advisory panel except Dr. Seitz, as well as its staff, were strongly prejudiced ab initio against SDI. Dr. Seitz is also disturbed that the advisory panel did not vote on the report.25This, one might suppose, would have been a waste of time, since the vote would surely have gone against SDI in view of his assertion that the majority of those on the panel were in the anti-SDI camp.

General Daniel Graham of High Frontier withdrew from OTA’s advisory panel because he anticipated that he would not like the conclusions which were being reached by the study team. He, at least, appears to be committed to SDI whatever the scientific judgment about the program’s technical feasibility.26 It is an entirely different matter when a scholar of Dr. Seitz’s eminence27—he took General Graham’s place on the panel—disavows the report for such reasons as he has so far made public. These reasons add up to a blunt denial of what has been said by critics of SDI about the ability of enemy space mines to destroy battle stations, the ability of “spoof launches” to confuse space sensors, and so on. Surely the issue of the technological feasibility of the SDI concept has become far too important to the world at large for it to be argued about by accusations of prejudice, whatever the quarter from which they come, rather than by cogent analyses.

If one were to imagine that the President’s dream will one day be given substance, far-reaching political and strategic issues will have to be debated, and debated internationally, in a world in which the 1972 ABM Treaty would have become a dead letter, and which in the meantime would undoubtedly have been transformed by major political events. But that could be decades away. Scientific judgments must come first, and they are an entirely different matter. Regardless of whatever political views he may now entertain (he is on record as having declared that the US should be able to make a first strike against the USSR), Dr. Seitz should be expected to argue his case before those of his scientific peers who have reached judgments on the facts—some of them in the field of basic science—that are contrary to his.

Dr. Jastrow bluntly says that the views of “professionals,” who work full time in the “defense science community,” should be given greater weight than those of their scientific critics, however distinguished they may be, and whatever their previous experiences of defense science. Lowell Wood is, not surprisingly, in full agreement. He tells us that Hans Bethe, Richard Garwin, and others who have dared criticize SDI “have fared uniformly poorly in technical debate in the classified surroundings required by government regulations,” and that it is because of their failures in secret conclave that they carry the debate to the public “immune from the criticism of those who know better.”28

This contemptuous dismissal by Lowell Wood of his critics harmonizes well with his claim that all the technological criticisms of SDI emanate from a few physicists who could be numbered on the fingers of a maimed hand. In any circle where the rules of scientific discourse prevail, both remarks would be dismissed with an equal measure of contempt. Unfortunately laymen who write in favor of SDI and who presume to make judgments on scientific matters about which they have little or no understanding tend to cite any scientific claim—for example Lowell Wood’s—that reinforces the entrenched views in which they have a vested interest, be it political or financial. It is highly regrettable, therefore, that many of the most influential and ardent proponents of SDI are politicians and officials such as Richard Perle who have so far displayed surprisingly little critical understanding of the difficulties that the R&D program entails. It is surely absurd that matters which obviously first need to be strictly judged on their scientific and technological merits, and which are of such profound importance to the future of life on earth, should be pronounced upon by laymen lacking either a scientific background or any experience in the management of major R&D projects—or both. The technical feasibility of a space-based BMD system is not a matter that will be resolved either by a show of hands, or by a slanging match in which the pro-SDI side on occasion goes so far as to suggest that its critics are soft on communism. The laws of physics and judgments about what is technologically feasible are not yardsticks for the measurement of political attitudes, any more than Galileo’s discoveries were disposed of by the conventional dogma of the Church.


The resolution of the technical argument will depend on the clear formulation of a few basic questions and, following that, on those competent to express a view providing the wisest answers that can be put before the administration, Congress, and the people of the world. For example, a fundamental premise, given that a space-based ABM system could be devised, is that beam weapons can be aimed from space at a ballistic missile before it ejects its payload of warheads and penetration aids, that is to say, they can be aimed at a single target and not have to contend with tens and tens of separate targets. If, as Dr. Garwin and others have argued, and as the Russians claim, the separation of warheads from the missile can be made to occur within, say, the first hundred kilometers of the atmosphere, then X-ray lasers and particle beams fired from satellites would be relatively useless since they lose their effectiveness when they enter the upper layers of the atmosphere.

The primary question, therefore, is whether a ballistic rocket can be fueled and programmed to eject its warheads before reaching that height. The recent OTA report, as well as that of Ashton Carter, gave a positive answer to this question, which was what the Fletcher study team also implied the Russians could do, given time. If this is the consensus of those best able to judge, and if the USSR were to seek to achieve the necessary countermeasures over the next decade (if indeed it has not already done so),29 the complexion of the entire problem of a space-defense system changes completely.30 One critical part of the SDI concept would evaporate overnight.

Take another question—the enormous number of targets which a space-defense system should be able to engage almost simultaneously. A ship-defense system known as Aegis, which was designed to track two hundred incoming cruise missiles, and to engage sixteen of them at the same time, has not yet been shown to be able to manage two or three. 31Have the contractors and engineers who have been working for years on airborne and ship defensive systems given their views in public about the engagement pattern that is presumed to be possible in the SDI concept—the destruction every second of between ten and twenty ballistic missiles in a salvo of more than a thousand?

Towering above all such technical issues is the question whether it could ever be possible to design the computer links that would be needed for a BMD system to function as a whole. This matter, too, is discussed in detail in several reviews, with generally pessimistic conclusions. Dr. David Parnas, a consultant of the Office of Naval Research, and an experienced professor of computer science, spelled out in detail his reasons for resigning from the official SDI panel that is dealing with the computer problems of a space-based ABM system.32 They make formidable reading, adding up as they do to the general conclusion that the job of designing the necessary computer network is an impossible one. In the letter of resignation that covered his detailed submissions, Dr. Parnas wrote that he was aware that there were software experts who would disagree with him, and for whom

the project offers a source of funding, funding which will enrich some personally…. During the first sittings of our panel I could see the dollar figures dazzling everyone involved. Almost everyone that I know within the military industrial complex sees in the SDI a new “pot of gold” just waiting to be tapped.

Dr. Parnas is fully supported in his view by the computer specialists who have recently founded an organization called Computer Professionals for Social Responsibility. British computer experts have also expressed their skepticism about what has been proposed,33 and even more recently Herbert Lin of MIT has ended a review of the entire problem by stating that “no software-engineering technology can be anticipated that will support the goal of a comprehensive ballistic missile defense.”34 All this is in line with the conclusions of the recently published OTA assessment. The fact that Dr. Solomon Buchsbaum of the Bell Laboratories and Dr. Danny Cohen of the University of Southern California have publicly expressed more optimistic views, even if they do not claim that error-proof or tested software for the SDI concept could be devised, does not dispose of the criticisms. What is more, it is difficult to imagine the political uproar that would result were the public to become aware that in addition to having its destiny entrusted to a computer network, it was one not free from errors in software. I doubt if SDI could ever surmount this obstacle. It would be worse than having nuclear antimissiles in one’s back yard.

The OTA report undoubtedly reinforced the views about the strategic shortcomings of the SDI concept which have been so powerfully expressed by James Schlesinger, Dr. Harold Brown, and Robert McNamara, three former secretaries of defense; by General Brent Scowcroft, whom the President had earlier put in charge of the preparation of a major report on the strategic forces of the United States; by Gerard Smith; by at least five of the holders of the office of director of defense research and engineering since it was established in the late Fifties, all of whom know from bitter experience, as I do, how easy it is to waste hundreds and thousands of millions of dollars in the pursuit of a technological will-o’-the-wisp; and by a number of other prominent men who have held public office in the field of national security. There may have been some members of Congress who also found it odd when the Canadian government decided that it wanted no part of the SDI program, even though any hypothetical space-based defensive system for the United States would automatically provide a shield over Canadian territory. In view of all the doubts, it is no wonder that Congress has now reduced the SDI budget for the coming fiscal year.

In consequence we are told that next year’s SDI R&D program will focus mainly on land-based systems. On the other hand, it should not be expected that the setback to the program will put an end to the work being done in Livermore on nuclear-bomb-pumped X-ray lasers, or at Los Alamos on excimer lasers powered by electron beams. As I have said, both laboratories had embarked on their pet laser and particle-beam projects well before the President spoke in March 1983, and they did so without being disturbed by any thought that the 1972 ABM Treaty barred the development of space-based defense systems, or by the fear that long before any such system could even be devised, the testing of its components would almost certainly constitute an abrogation of the treaty.

There is also no reason to suppose that the men who are working on a super-computer and software for a space defense system are likely to bring their work to a halt because authoritative computer specialists have declared that it will never be possible to devise an acceptable network which could transform the separate components of a space-based BMD into a workable BMD system. The theatrical dream that was the background of the President’s challenge to the scientists of America should in retrospect be seen as a proclamation to the world that work on particle beams and high-power lasers was already in progress. In no sense did it set that work in motion. It would be equally sensible and prudent to suppose that research and development work on lasers and particle beams that is going on in the USSR was not halted by the announcement of the American SDI program.


One consequence of the criticisms of the SDI program has been the reduction of the SDI budget. Another is that many of the explanations that are now given by the administration for the need for the program to continue differ from the President’s original vision and from his view that a defense against ballistic missiles constitutes a higher category of morality than the maintenance of security through the threat of mutual annihilation. One major justification continues to be heard: that the Russians are engaged on work that corresponds to different elements of the SDI program, and that in many ways they are ahead of the United States. We have also been told that some Russian actions have already breached the terms of the 1972 ABM Treaty. Specific violations are spelled out in impressive brochures.35

The Russians counter by pointing to American actions which in their view are breaches of the treaty. They have even offered to suspend work on the much spoken of, and highly vulnerable, vast phased-array radar system which they are building at Krasnoyarsk if the United States abandons its program to modernize the radar complexes which it has at Fylingdales in the United Kingdom and Thule in Greenland. Their spokesmen argue that these modernization plans, and particularly the rebuilding of Fylingdales as what is rumored to be a 360-degree phased-array radar complex, is far more questionable than what the USSR is doing at Krasnoyarsk.

A further accusation by the administration is that the USSR has committed “a far greater investment of plant space, capital, and manpower” to advanced BMD technologies than the US has.36 This extravagant claim is not borne out by a CIA document about Soviet efforts which was presented to the Armed Services Committee of the Senate on June 26, 1985.37 Indeed, the document expresses doubt about the applicability of even a network of Krasnoyarsk systems—regarded as the most serious breach of the 1972 treaty—for widespread ABM deployment. Dr. Garwin, in a follow-up to testimony presented to a congressional study group on October 10, 1985, has also pointed out that the better part of the large Soviet program on strategic defense is devoted to the upgrading of its anti-aircraft defense system.38

But whatever the truth about Krasnoyarsk, it can hardly be a justification for the US deliberately interpreting the 1972 treaty so widely that the Russians are given cause to say that the US is proposing to contravene the treaty in a much more specific way, or ways, in order to gain the “advantage” of being able to launch a first strike against the USSR without fear of significant retaliation.

It was therefore unfortunate that immediately before the Geneva summit, Robert McFarlane, then the head of the National Security Council, declared that no aspect of the development of space-based BMD components is prohibited by the 1972 ABM Treaty, and that what was intended about testing and development merely implied a shift from the technology that was available at the beginning of the 1970s to what can be undertaken today. This statement could be taken as reflecting the hard fact that major vested interests are now involved in the SDI program—not only the men in the laboratories who started the whole thing and the authorities in the Defense Department who encouraged them, but also the industrialists who see in the SDI program a bonanza that they cannot afford to disregard. Unfortunately the statement also clearly implied an intended breach of the treaty. Indeed, Gerard Smith has pointed out that what McFarlane implied was not just a breach, but a new version of the treaty.39 That the statement was publicly played down before the President met Mr. Gorbachev was therefore only to be expected.

But it remains highly regrettable that the myriad and diverse arguments about SDI have now induced what might well be described as a state of schizophrenia among America’s European allies. All of them recognize that the coherence of NATO is a vital consideration, and one that makes it necessary for the United States, as the keystone of the alliance, to be supported in its policies whenever possible. But at the same time there is considerable skepticism in Europe about some of those policies, and particularly about America’s nuclear policies, including the SDI program, which is widely regarded as a threat to the 1972 ABM Treaty and as a spur to the nuclear arms race. The arguments about the deployment of cruise and Pershing II missiles on European territory caused considerable political trauma and their echoes have not yet died.40It would therefore be a major error of political judgment to treat lightly the fact that vast numbers of Europeans are fearful of any moves that might lead to a further buildup of nuclear armaments, or to assume that any deterioration in the relations between the US and the USSR as a result of SDI would not produce a new wave of antinuclear, and indeed of anti-American, protest in Europe.

The agonizing that is now going on about the US invitation to engage in SDI work is already a practical sign of the disquiet and suspicion which are entertained about the President’s initiative. Some NATO governments have declined because they dislike the entire idea on political and strategic grounds. The British government agreed to participate in the knowledge that if it refused to provide a formal blessing SDI scouts were already in the field seeking to entice European specialists with particular skills to work in the United States. Since the 1972 ABM Treaty bars the United States from sharing with others any technology that relates to strategic ballistic missile defense systems, cooperation will do little to help either the economies or the military defenses of European countries that formally bless collaboration on R&D, except insofar as such SDI R&D contracts as may be won in probably costly competition with American companies could provide employment for some European scientists and engineers in what may well turn out to be no more than a sharecropping exercise.

Europeans who concern themselves with these matters appreciate that even if the nuclear arsenals of both sides were cut by 50 percent—as has now been proposed by both the US and the USSR—more than enough destructive power would still remain, whatever way the cuts were made, to devastate not only the European mainland but also the United States and the western USSR. The concept of nuclear superiority has become meaningless. It belongs, in the language of Lord Carrington, NATO’s present secretary general, to the unreal world of “nuclear accountancy.”41 And Europeans no more believe that their countries could be defended by a space-based BMD than they imagine that the USSR would ever risk a first strike either in Europe or against the US. Many suspect that the picture of a layered space-defense system was fabricated in order to confuse the innocent into supposing that a space-based BMD would operate in a measured sequence, a proportion of the offending missiles or warheads being destroyed as they traversed the layers in turn. The greater the number of layers postulated, the more missiles would be destroyed, until in theory—and on paper—almost all were eliminated. But, as I have said, it is the first layer defense that is both decisive and regarded as unfeasible by independent scientists. There are also many European officials who, being concerned with real military security, wonder what SDI has to do with Europe. They know that while it is just conceivable that the Russians might one day attack across the Iron Curtain, their purpose would be to occupy territory—not radioactive territory that had been devastated by nuclear weapons.

The President and Mr. Gorbachev now seem to be locked into their respective positions. Time and time again the Russians have declared that if the US continues in its search for a space-based defense system, it will embark upon its own countermeasures, including the further buildup of its offensive forces. This is surely not propaganda. In the Weinberger letter to the President that was leaked just before the Geneva summit, the defense secretary warned that “even a probable territorial defense [by the USSR] would require us to increase the number of our offensive forces and their ability to penetrate Soviet defenses to assure that our operational plans could be executed.”42 That is precisely what the Russians also say they will do if the US continues to seek, through SDI, to develop a “territorial defense.” And, as Mr. Nixon warned in his recent article in Foreign Affairs, it would be easy for the Russians to triple in little time the number of warheads that are carried on their giant SS18s, a simple multiplication which in theory would by itself increase the threat that US missile silos face from an SS18 first strike from three thousand to nine thousand MIRVed warheads.

Richard Nixon and Henry Kissinger gave their support to SDI because they saw in it both a means whereby the Russians could be induced to return to Geneva, and a “bargaining chip” in arms control negotiations. But if one were to regard SDI as a bargaining chip, one would also have to accept that the US will gain only if it throws it away. If the SDI R&D program continues, the Russians will respond. Even were SDI to confound its critics and succeed in the sense that its separate components could be fitted together in a working system, the United States and the West as a whole would still lose, not only because the USSR would have devised measures for defeating a space-based BMD, but because there are ways other than land-based ICBMs, for example long range low-flying cruise missiles, whereby the US could be threatened with nuclear devastation.

President Reagan still speaks as though nothing has changed his original dream. But it has been changed. He himself changed it when he declared after the Geneva summit that what the United States was embarking upon was a non-nuclear space defensive system. That declaration, if acted upon, would be the death knell of the nuclear-pumped X-ray laser, the kernel of the scenario of a defensive astrodome first painted for him by Edward Teller.

Paul Robinson, the principal associate director for national security programs at Los Alamos, has been recently quoted as saying that the X-ray laser is in any event flawed because “it might inadvertently wreak havoc on other SDI components in space,” while his colleague, Steven Rockwood, the Los Alamos director of SDI research, asks whether an orbiting device containing a powerful nuclear bomb could ever be politically acceptable.

But, one now has to ask, did an effective X-ray laser ever exist, or could it be made to exist? Whatever the President’s motives in insisting in recent weeks that his SDI proposal implied a non-nuclear BMD, his protestations, no doubt inadvertently, coincided with a growing volume of informed comment, based on recently published statements by Livermore itself, to the effect that the claim that an effective nuclear-bomb X-ray laser has been devised was not only premature, but also based upon an unwarranted reading of measurements made in critical tests.43

What is more, some directors of SDI research at Livermore have publicly expressed concern because the success of the research for which they are responsible has been exaggerated by Pentagon officials. Dr. George Miller, head of defense programs at the Livermore laboratory, has been reported as saying that the public “is losing sight of how difficult the job is,” while his colleague Dr. Cornelius F. Coll III, who is director of “Star Wars” systems studies at Livermore, declared that “overstatements by Pentagon officials were imperiling the program…. This job is difficult enough without having to defend hyperbole and exaggeration.” It is even reported that a recent demonstration which was laid on to impress a selected audience about the effectiveness of the electromagnetic rail-gun was a spoof. The demonstration pretended to show that a mock-up of a Soviet SS18 missile could be destroyed by the rail-gun. In fact, General Abrahamson is reported by The New York Times as having later revealed “that the damage had not actually been done by an electromagnetic rail-gun but by a hardened projectile fired from an air-gun”—a weapon whose antiquity goes back to the early eighteenth century!44

Surely the President must now appreciate, possibly even from what Gorbachev told him, what the arguments against SDI are. Surely he realizes that the nuclear arms race is different in kind from the competition which takes place in the field of conventional arms; that the idea that the US, the USSR, and Europe could ever by subjected to a nuclear conflict is total madness; and that such a conflict could solve nothing. In the forty years since Hiroshima and Nagasaki, increasing numbers of nuclear warheads and delivery systems, not to mention presumed defensive measures, have not provided greater security to any party—not to the United States, not to the USSR, and not to Europe. What they have done is reduce security for all.

We often hear the homely term “leaky” in the course of the SDI debate, as though if a perfect BMD defense proves impossible, a “leaky” one would still be worth having. It is yet another of those words which helps to lull the senses, so that we fail to realize the hideous reality—that the fraction of warheads that would “leak” through would today be enough to cause what once used to be euphemistically called “unacceptable damage.” We continue to talk about numbers of warheads and megatons as though they were numbers of tanks and bomber aircraft. The brutal fact which our minds seem incapable of taking in is that were the explosion to occur over New York or Washington, London or Moscow, one megaton would be equivalent to a million instantaneous deaths (what matter if the figure were 100,000 or 200,000 more or less?).

The President may protest that his SDI dream implies a protection of people and not of silos. But however many times he does so, the fact is that were the “unthinkable” ever to occur, a future American president would probably never know how his enemy had behaved. He could well have disappeared in the nuclear Armageddon. If the SDI program ends up only in protecting America’s land-based missiles, no president could be sure that given a nuclear outbreak, the Russians would necessarily confine their fire to the American missile fields and not also aim at centers of population, any more than the Russians can be relied upon to believe that the United States would spare their cities. A “point defense” or SDI II, as some now call it, would, in short, take us back to square one—to the same argument that revealed the futility of missile defenses and which ended in the 1972 ABM Treaty.

Adhering to the strictest interpretation of that treaty has therefore become a vital consideration for all of us—not some so-called liberal interpretation of the way its terms were drafted, however legally argued, not some new version, as Gerard Smith has put it, but the treaty in the sense in which it was negotiated by the two sides. Were some demonstration test of a novel BMD component by either side to result in a unilateral breach, it would be but a short step to the abrogation of the few other treaties that have been so painfully negotiated in order to try to stem the spread of nuclear weapons.

A conflict in which nuclear weapons were used would not help solve any of the political disputes that now divide the two superpowers. It would certainly make it impossible for either to help solve the multitude of territorial and racial disputes and problems of social and economic development which now torment the nations of the world, and in the resolution of many of which both have a common interest. Both leaders should therefore remind themselves of the critical difference between the BMD of the Sixties and what is being discussed now. Twenty years ago, active defenses against missile attack were being devised by both sides in response to a formulated operational requirement which it was incorrectly assumed could be technically satisfied. Today SDI is a concept that is “technology led” by the belief that new technological wonders can be fitted together in order to create an effective operational defense system. No one, not even the President, believes that this could ever happen before the turn of the century, if indeed it ever proves possible. He also knows that in the interval there could be military conflict.

The two leaders should therefore keep on reminding each other that were the prevailing state of nuclear deterrence to break down, the result could be a catastrophe unparalleled in the history of warfare, and one which would make even the worst natural disaster of which history tells us seem like a gust of wind. Let us therefore hope that when President Reagan and Mr. Gorbachev next meet, even if they do not discuss technicalities, their visions of the dangers which they face in the years ahead will move them closer than they appear to have been in Geneva.

This Issue

January 30, 1986