Chapter 7
Deploy or Perish
SDI and Domestic Politics

Like all other large-scale efforts to develop military technology, SDI is inevitably deeply enmeshed in domestic politics. Simplistic formulas by which such entanglements are sometimes measured—such as the suspicion that all military projects result from promotion on the part of a "military-industrial complex"—hardly describe the realities well enough.^[1]    This is certainly true in the case of SDI, which, as we have seen, was not promoted by any of the usual suspects. Although the project had some preliminary support among certain researchers in the national laboratories and defense firms, as well as among a handful of members of Congress, it became a high-priority national program only because a popular president embraced it and used his powers of persuasion to win support from the electorate. In order to sustain a project of this magnitude, especially when it moves from the research phase into the far more expensive and significant development and deployment stages, the support of a strong coalition of political forces is essential. The early history of SDI has been a struggle between those eager to build such a coalition and those determined to thwart and counteract their effort. It thus presents an instructive case study of how major projects in need of political support do or do not take hold, and how pressures for early deployment distort the research agenda.

Democratic Principles And The Cold War

In democratic theory, a nation's commitment to shift to a radically new and exceedingly costly defensive strategy ought to rest on a widespread

consensus that the change is desirable and feasible. In the United States, decisions are often made and sustained without such a consensus. Acquisition programs develop a momentum that makes them hard to stop, even when they are not justifiable in military terms. In very complex R & D programs, it is often difficult to know with a high degree of certainty whether what is envisioned in the program will be beneficial. Much depends on whether enough of those with influence believe in the desirability of a new weapons system strongly enough to press for its continuation and then, finally, for acquisition and deployment. Some of the will to believe comes from self-interest, some from conviction. The domestic politics of defense is to a large extent a struggle among forces motivated by a combination of self-interest and public interest—a struggle fought out in the budgetary process, in the press, in public-opinion polls, and during elections. The struggle is also influenced by changing perceptions of the international environment. If the Soviet Union is perceived to be mounting some new threat, and a particular program seems to be a plausible response, it is likely to gain adherents. If the Soviet Union does not appear to be mounting such a new threat, the quest for a technological fix becomes less appealing. The mistrust of the Soviets, and thus the competition for advantage, has been hard to overcome, so promoters of the continuing development of new military technologies have usually won out over resisters.

Whether or not the deployment of strategic defenses makes sense, therefore, there is some danger that the expenditure of considerable sums of money in developing them could lead to a point of no return, at which partial deployments would be made in the hope that they will eventually lead to a comprehensive system, even though there is no reason to believe that it will ever become a practical prospect. SDI has been restrained so far out of concern for its technical uncertainties and the high projected costs of development and deployment. Many also view it as unnecessary and unwise in view of the changes in Soviet foreign and military policy introduced under Gorbachev. The domestic struggle over the direction of SDI will be influenced, as it already has been, by the interplay of these international considerations with domestic politics.

SDI And Earlier Military Projects: A Comparison

All major military projects have been "political" in the sense that those who decide to support them have policy objectives in mind. The United States' decision to develop the atomic bomb was a watershed in

the degree to which the political authorities became committed to a program involving research as well as development. Unlike other projects aimed at perfecting existing weapons, this one was designed to conduct a complex series of experiments in an effort to discover whether a wholly new weapon (which was thought capable of introducing a radically new dimension to warfare) could be developed. The decision was made in conditions of secrecy at the highest level of the executive branch and was influenced by the fear that a wartime enemy, Nazi Germany, might be the first to develop such a weapon. Scientists played an important role in the decision—far more important than they had played in the initiation of military-technology programs in the past—because it depended critically, first, on their judgments of technical feasibility and, second, on their commitment to make the concept work.

The success of the Manhattan Project set two fundamental precedents. One was that national security was now closely coupled to progress in science and technology, at least in the physical sciences. Although the team that had developed the atomic bomb was dispersed at the end of the war, the project was revived a few years later because of a growing appreciation in the United States that a continuing effort had to be made to support science and technology for purposes of defense. This decision was not a result of domestic pressures from defense industry or the military services but was largely a reaction to the intensification of the could war. The U.S. government, spurred on by public concern, was reacting to Soviet success in developing an atomic weapon, the Berlin blockade, creation of the Sino-Soviet bloc, and the outbreak of the Korean War. Even though the United States enjoyed a clear advantage in military technology over the Soviet Union, it seemed imperative that the partnership between the federal government and the military laboratories as well as with the research universities be maintained and strengthened in order to maintain that advantage, especially in view of the U.S.S.R.'s potential manpower and tactical superiority over Europe. This partnership represented a considerable break with a past in which the federal government was often constrained on constitutional and ideological grounds from either becoming an economic actor or from subsidizing private industry. Especially in the Progressive era, government had been conceived of as a watchdog over industry, a champion of free enterprise against combinations, trusts, and oligopolies. After World War II, however, concern for national security put the government in the unaccustomed role of underwriting the risk of private contractors and of reinforcing industrial oligopoly in order to support the most essential military contractors.

The United States has so far resisted pressures to go even beyond these arrangements. It has not created a cabinet-level department to coordinate science and technology, nor has it adopted a formula for national R & D expenditures based on the theory that overhead costs should be borne by the government. The United States' approach has been sectoral and pragmatic and has so far not engaged the federal government consistently and openly in the support of civil as well as military research and development. Despite the evidence that other countries have profited from government planning and direction, the United States has adhered to the classical liberal principle that private industry can do the best job of promoting innovation in the civil sector if left free of government interference. In the two principal areas of federal involvement, defense and space exploration, the U.S. pattern has been to emphasize work performed by private industry but subsidized by public R & D contracts and procurement. In supporting this R & D, the federal government relies on the "project system," except in the case of three nuclear-weapons laboratories—Los Alamos, Livermore, and Sandia—where research is supported on a nonprofit, "level-of-effort" basis—much as the Soviets do in their missile and other weapons-research facilities. Although some critics argue that the United States would be better off admitting that it had created a "contract state" or a "managed economy," and nationalizing those defense firms whose work is almost entirely dependent on federal contracts, such steps have been resisted on the ground that they violate traditional American norms and would be counterproductive.

Despite such disputes about methods, it was well recognized, especially in the late 1950s, that national security required a wide range of activities in R & D of all sorts. Various factors influenced decisions on research priorities: the perceived military need, technical feasibility, and the comparative cost vis-à-vis other existing or new systems. In an effort to achieve better control of the process, Secretary of Defense Robert S. McNamara introduced cost-benefit analysis for comparing weapons systems rather than allowing the services to pursue their own needs and to iron out priorities by interservice bargaining. Although the need to reconcile economic and strategic goals with the priorities set by each of the services continues to bedevil defense planning, McNamara's efforts at least compelled the planners to think about the interrelationships of weapons and their roles in general strategic doctrine.

Although it would be too much to say that strategic doctrine alone fueled the quest for new military technology, it is certainly true that the advent of the atomic bomb gradually led to the adoption of the belief in

strategic deterrence, and that this belief influenced the development of offensive forces.^[2]    The decision to employ intercontinental ballistic missiles as delivery vehicles for atomic weapons was only a more efficient way to implement a strategy already in place. The Soviet Union undoubtedly had something to do with this decision, but it probably would have been made anyhow. The decision to build the thermonuclear bomb was different. It was not obvious in 1949, when the matter was first considered, either that the bomb could be designed or that it would confer some military advantage not supplied by the atomic bomb. Accordingly, the scientists on the General Advisory Committee (GAC) to the Atomic Energy Commission recommended almost unanimously against a crash program to develop the new, more powerful bomb. President Truman overruled the GAC on the advice of a small group of scientists committed to the idea. Truman and his political advisors were uneasy that the Soviets had developed an atomic bomb of their own, fearing that they might develop a thermonuclear bomb ahead of the United States and use their resulting superiority to achieve advantages, much as Secretary of State James F. Byrnes had tried to do with "atomic diplomacy."

The H-bomb decision deviated from the wartime pattern in that the president decided to proceed over the opposition of key scientific advisors, who had warned that a crash program was not only premature but would also be costly in terms of other military needs, such as a continental air defense and tactical atomic weapons. Nevertheless, the president did not reach his decision arbitrarily or in defiance of informed opinion. In fact, the administrators of the Los Alamos Laboratory had recommended that research on the H-bomb be accelerated, along with other projects. Karl T. Compton, head of the country's highest-ranking defense-science advisory body, the Research and Development Board, which was attached to the Department of Defense, had recommended going ahead with the effort to develop the H-bomb. This decision, then, was based on recommendations of those with operative responsibility and on the opinion of senior scientific advisors, even though it had been opposed by the GAC. Congress's role in this decision was almost as limited as it had been in the case of the atomic bomb, except that one senator, Brien McMahon (D., Conn.)—who had a key role in such matters as chairman of the Joint Congressional Committee on Atomic Energy—worked with the air force in promoting the case for a crash program.^[3]   

Since then, the general pattern in weapons development has been that new ideas have percolated up, from the industrial contractors, the DOD laboratories, and the federally funded research centers. The ideas move

up through a hierarchy of decision makers, including those in charge of military R & D and the military planners, and are reviewed—if they reach the top levels of the Pentagon—by special outside advisory committees made up of experts from industry and academia, in addition to senior officials, elected as well as appointed. Ideas that involve new weapons that either parallel new Soviet developments or act to counter them are especially likely to run this entire gauntlet. In 1954, for example, a U.S. ICBM program both paralleled the Soviet ICBM project and seemed to solve the problem of how to penetrate Soviet air defenses. The ICBM's unique speed was also an important consideration. Such proposals reach the White House only after lengthy, thorough, and iterative reviews in the defense establishment. When proposals come to the attention of the president without undergoing this process, they have rarely if ever been acted upon.

The "Iron Triangle"

In general, the dynamic system now in place generates so much work and so many new ideas that it is sometimes accused of promoting projects for which there is no real military need. Hence, President Dwight D. Eisenhower's warning against the "military-industrial complex" and the scientific-technological elite, and the indictment of the Pentagon as a kind of state within the state, nurturing new weapons in order to perpetuate the cold war and the "warfare state." Recurrent scandals involving kickbacks, exorbitant consultant fees, and "revolving door" employment of civilian and military-procurement personnel by defense contractors have reinforced this suspicion. Nevertheless, the reality is more prosaic. The profit motive and corrupt practices oil the wheels, but they do not drive the machine. Cold war anxieties, coupled with the rapidity of technological change in modern times, provide the main impetus for weapons innovations, as is apparent in the fact that the Soviet Union, even without a private-sector economy, has developed a thriving military-industrial complex of its own.

More and more, this system has also come to involve Congress in detailed decisions on R & D and procurement. Although Congress has rarely if ever shaped new strategic or defense policy (such as the "New Look" or "Massive Retaliation," which have typically been introduced by the executive), both houses and their committees take a keen interest in the allocation of federal funds for defense R & D and procurement. Much of this is a pork-barrel interest. Members of Congress are eager to

secure new projects for their districts and to preserve those already created. Key legislators on key committees are in the best position to control the distribution of the largesse, resulting in an "iron triangle" that links the military services, the contractors, and the key members of Congress.^[4]    Their combined efforts create a formidable force in favor of the perpetuation of military spending and the initiation of new programs. When there is conflict, it is often because those in the triangle perceive their interests differently. When one aircraft has to be chosen from among several candidates, different constituencies become rivals, and the prime contractors seek to reduce political risk by spreading the work over as many congressional districts as possible. When the interests of one branch of the service have to be chosen over those of one of its rivals, the same kind of competition ensues. Congress sometimes plays a more constructive role when individual members expose scandals and use the legislature's investigative power to expose kickbacks, inflation of costs, and other forms of fraud and waste. The General Accounting Office, a major watchdog agency of Congress, has also achieved significant reforms in contracting procedures after discovering that contract reviews in negotiated R & D contracts were often inadequate and unduly influenced by relationships between the contractors and the military-procurement officers. But the opportunities and temptations have proved hard to control or police, and collusive practices persist.

To characterize the military R & D and procurement processes as a triangle, however, is to miss some of its dimensions. Much of the incentive for new programs comes from service rivalries, sometimes even within particular branches of the services. Thus, "bureaucratic politics" involving competition between two services, as in the conflict between the army and air force in the Thor-Jupiter controversy,^[5]    weighs heavily in the outcome of such struggles. The Defense Department's effort to gain a measure of control over such rivalries is also a factor in the picture, yet the DOD is itself a bureaucratic actor. It has sometimes promoted weapons systems and modernization out of its own "organizational imperative."

Disentangling self-interest from public interest is difficult. When the services and the DOD propose some pattern of military spending, it is not necessarily, or even usually, the case that they are acting purely out of organizational self-interest. These organizations tend to develop a concept of a public interest in national security that is presumably being served by the policies they advocate. They argue, often plausibly, that they cannot carry out their policies unless they can count on help from

others in the executive branch, especially the president, and unless they receive enough backing from public opinion to influence Congress to vote the funds in accordance with their priorities, rather than those dictated by political pressures.

Nevertheless, the general pattern of support for new military projects is one in which the laboratories tend to initiate ideas, the services become committed to particular projects and strive to have them adopted and brought to completion, the Defense Department reviews them and determines which of them make sense, and the secretary of defense, in collaboration with the contractors and Congress, decides whether or not to pursue them. By no means are all plausible candidates supported, and the evidence suggests that the contractors alone cannot carry the day unless they can succeed in forging a strong enough coalition with at least one of the armed services and interested politicians. A good example is the B-1-bomber program. When President Carter decided to shelve this program in favor of upgrading the older B-52 until the Stealth bomber would come into production, the proposal for a major new military program was temporarily defeated. This happened even though the prime contractor, Rockwell, had a great deal at stake and had assembled a formidable coalition of industrial subcontractors who stood to benefit from the adoption of the program. The formidable B-1-bomber lobby spent heavily on behalf of the cause and eventually succeeded, but it would not have done so without the election of Ronald Reagan, who was committed to a broad program of strategic modernization, including the B-1. Thus, the B-1 cannot be viewed as an example of a lobby's strength alone. In this case, the key to its success was the election of a president with a political mandate to support the program.^[6]   

The Political Auspices Of SDI

Political auspices were equally critical for the initial success of SDI. The SDI was promoted by a very small coalition of military officers, scientists, strategists, and politicians, but it enjoyed nothing approaching the support for the B-1 bomber. SDI did not carry the immediate promise of lucrative procurement contracts, and it did not arouse much enthusiasm among military researchers, who had come to rely more on continued modernization of offensive weapons. Because there seemed to be no effective defenses against nuclear attack, the important goal, in their view, was to assure that the nuclear-deterrent forces could be protected from attack, could penetrate prospective defenses, and would be

accurate enough to do whatever was required. Although there had long been a significant level of support for basic research into technologies that might have defensive applications, the 1972 ABM Treaty had ruled out their development. Indeed, as we note in chapter 1, early in the Reagan administration's first term, even the Defense Department's own internal review of the High Frontier proposal on strategic defenses had led to the conclusion, announced by Secretary Weinberger, that the project was premature at best, because it could easily be defeated and overwhelmed.

The president's decision to make SDI his own pet project made all the difference. As an air force general with planning responsibilities observed: "All other systems stem from some statement of operational need. SDI was generated by the president. We didn't have that operational requirement before the president's vision of a world with no nukes."^[7]    As we note in chapter 1, Reagan did not seek prior approval from the defense establishment because he had been warned that SDI would be resisted. He also did not consult the White House Science Council. When a broad review committee was finally established (the Fletcher Committee), it was not asked "whether," but "how." Instead, Reagan chose to listen only to enthusiasts and to trust his own judgment about what was desirable. That judgment was based on a faith in U.S. science and technology as well as on a personal frustration with the existing strategic situation and concern that a future U.S. president might not be able to rely on being able to deter an attack by threatening unacceptable retaliation. As we saw in chapter 1, this belief, coupled with skepticism about the prospects for negotiating a reduction of the nuclear threat and a dislike for the very notion of mutual U.S.-Soviet nuclear deterrence prompted the president's decision. It was successfully implemented not because the president could count on support in Congress, the DOD, and the contractors—or, in other words, from all points of the "iron triangle"—but rather because he went over the heads of the "establishment" and appealed to the electorate. His simplistic vision of a space shield against nuclear weapons aroused precisely the strong public approval that he needed and that his political instincts told him he could expect.

Once he advanced this vision and the people responded favorably, Reagan could count on enlisting institutional support among the contractors whose interests would be served, within the services where special divisions would be created with responsibility for the program, and among those who would develop the strategic rationale for the program.

Although he could anticipate opposition to the program, he could calculate that most of it would come from political liberals, whom he had defeated before. By couching the issue in ideological terms, as a test of his hardheaded conservatism versus his critics' supposed weak-kneed liberalism, the president could hope to fashion a foreign policy issue that would serve as a focal counterpart to his campaign for the domestic conservative agenda. SDI had a particular appeal for his ideological supporters because it represented an effort to achieve national security by relying on U.S. military-technological initiatives rather than on negotiations with the Soviets. At the same time, it undercut the moral case of freeze advocates and went them one better by proposing a nonnuclear defense that would make nuclear weapons "impotent and obsolete." In short, it was in many respects well calculated to (1) evoke the same fervor that other elements of the conservative crusade had elicited among Reagan's most ardent following and in the general public, and (2) buttress and institutionalize this appeal by building a coalition of support among contractors, the defense laboratories, and influential figures in the media. Opposition, the president knew, could easily be labeled as evidence of the liberals' eagerness to make lopsided concessions in order to achieve agreements with the Soviet Union. Ironically, liberal opposition would only make the cause of SDI that much more formidable politically.

Public Support For SDI

According to opinion polls, SDI has been favorably received by the American public. This general assessment may be deceptive, however. Polls measure simplified responses to complex and sometimes ambiguously worded questions, and some of the data suggest serious confusion and reservations among the respondents, especially in response to questions that link SDI to issues of arms control and expenditure. Public support never seemed to grow firmer than at the time of the 1984 presidential election; but even that support was ambiguous. During the campaign, the Democratic candidate, former Vice-President Walter Mondale, accused the president of wanting to "militarize the heavens" with a scheme that would not work and would prevent progress in arms control and cause grave destabilization. The public seemed to sympathize both with Reagan's proposal and with Mondale's doubts. Among viewers of a televised debate between the candidates, 57 percent of those polled found Reagan more persuasive on SDI; 23 percent found Mondale

more persuasive.^[8]    A week before the election, however, another poll showed that 48 percent of registered voters concurred with Mondale that SDI would "just speed up the arms race," while only 30 percent thought it would "make us more secure."^[9]    After the Reagan landslide, the confusion remained just as evident. Early in January 1985, 62 percent of respondents thought a defense in space could prove workable, compared with only 23 percent who thought it could not. In the same poll, however, 60 percent said that having such a system in space would worry them, whereas only 25 percent thought it would make them feel more secure.^[10]   

The failure of the meeting at Reykjavík to produce agreement on arms control or on SDI-testing seemed to strengthen public support. The public overwhelmingly supported the president's stand at Reykjavík. In one poll, 56 percent approved the president's refusal to trade away SDI for an arms-control agreement, compared with 16 percent who disapproved.^[11]    Another showed 62 percent supporting the president against 22 percent opposing him.^[12]    Still another showed a still larger disparity of 66 percent to 21 percent.^[13]    And when a poll stressed Reagan's leadership role by asking whether he did "the right thing" in "refusing to change his ideas for the development of 'Star Wars'"—72 percent agreed and only 18 percent disagreed. Asked in the same poll if they continued to favor SDI, a majority indicated approval by a margin of 60 percent to 26 percent.^[14]   

Several polls revealed considerable apprehension about the possible effects of SDI. In a November 1985 poll, 71 percent of the respondents agreed that negotiations for a reduction in nuclear missiles were preferable to developing SDI.^[15]    In October 1985, 74 percent of those polled said that it was more important for the United States and U.S.S.R. to agree to arms reduction than for the United States to develop space-based weapons to defend against nuclear attack.^[16]    Asked the same year about Gorbachev's proposal that both sides cut their missiles by 50 percent and negotiate a "total ban" on the development of space-based defenses, 47 percent approved, 32 percent did not.^[17]    Respondents in another poll that year agreed, 46 percent to 39 percent, that "some limits" on SDI should be accepted if no treaty could be negotiated otherwise.^[18]    By 53 to 33 percent, those questioned said they would choose negotiations over development of "Star Wars," if these were the only choices.^[19]    Almost a year after Reykjavík, in September 1986, by a slight margin of 47 to 43 percent, respondents favored cutting back on "Star Wars" in order to achieve an arms-control agreement.^[20]   

In addition, despite support for SDI, before and after Reykjavík, there was evidence of concern over the costs of the project. In January of 1985, 46 percent said SDI would not be worth the money it would cost, while 40 percent said it would prove worthwhile.^[21]    In October 1985, 48 percent said they would favor spending the "many billions" it would cost, but 46 percent were opposed.^[22]    In November 1985, by a margin of 59 to 27 percent, respondents agreed that the president's proposed budget of $26 billion over five years was too high.^[23]    In January 1987, however, after an election in which the president failed to persuade the electorate to send back members of Congress who would support SDI, 50 percent still thought SDI was "worth the amount of money it would cost," against 37 percent opposed.^[24]   

If any conclusions can be drawn from these responses, the most fundamental would seem to be that the president did succeed in persuading most respondents that his proposal was a good one, but that many remained concerned that SDI was blocking prospects for arms control and could turn out to be more expensive than it was worth.^[25]   

Creating A Special Office: SDIO

In order to give SDI better definition and better protection against cuts in the defense budget, and probably just as much to protect it against doubters in the defense establishment, the Reagan administration decided to establish the Strategic Defense Initiative Organization (SDIO) within the Department of Defense. The office initially succeeded in preventing the program from fragmenting and in providing a focus for the direction of research and public information. It has been less successful, however, in protecting the program from budget restraints imposed by Congress.

In appointing a military officer with a background that included direction of the NASA space shuttle program, Lieutenant General Abrahamson, the administration gained the assistance of an experienced manager who had shown himself able to work with a variety of agencies and congressional committees. The office drew upon the services of industrial and government laboratory scientists and engineers and immediately launched a major effort to invite proposals from hundreds of industrial laboratories. SDIO recognized from the outset that the agency needed to establish itself within the military services and in the laboratories in order to widen its base of support and to mobilize the degree of cooperation that would be necessary to implement the complex agenda.

Accordingly, funding authorities were set up in the various services through which SDIO channeled its contracts, creating a network of military officers whose careers would be affected by the success of the project. An Office of Innovative Science and Technology was established to encourage submissions from university researchers, thereby blunting the outspoken opposition that was emerging on the campuses. Over the first several years of the project, contracts were let to hundreds of prime and secondary contractors and university-based researchers.

While SDIO benefited from strong support of Weinberger and the president (who ordered that SDI be protected from defense-budget cuts required under the 1985 Gramm-Rudman Deficit Reduction Act), it had less success in establishing its credibility in Congress. A small group of strongly supportive congressmen tried valiantly to resist budget cuts and to trumpet the significance of the program's achievements. SDIO set its own technical milestones, and each time these were met, the claim was made that they vindicated the aims of the project and showed progress even more rapid than had been anticipated. Critics were not impressed, however, claiming that the milestones were publicity stunts. To some extent, perhaps, SDIO suffered from public disillusionment with high technology owing to the disastrous accident at the Chernobyl nuclear power plant, the debacle of the Challenger space shuttle, and the failure of the AEGIS system aboard the USS Vincennes in the Persian Gulf. Congressional opponents were able to cite the OTA reports that warned of technical and other barriers to the success of the project, as well as the opposition of leading scientists. Neutral members of Congress were influenced by negative assessments from within the defense establishment, openly advanced before the March 1983 speech but expressed confidentially thereafter.

In response to the urgings of its supporters and in the hope of counteracting the criticisms of its opponents, the SDIO decided to shift its strategy in 1986 to put more emphasis on near-term deployable systems and less on longer-term research. Although the office formally denied any such change, analysis of actual allocation of funds belies the SDIO's claim: constraints on expansion of funding had been met by shifting the allocations to nearer-term technologies. Both supporters and skeptics in Congress had joined in urging the SDIO to identify near-term prospects. The supporters were eager to show that the project was succeeding in order to prevent their opponents from voting low levels of funds and "researching it to death." The critics were convinced that SDI could not achieve the president's goal, but some thought that a deployment of

mainly terminal defenses for retaliatory weapons had merit. Hoping to weaken popular support, some of the critics were undoubtedly also interested in putting SDIO in the position of abandoning the president's vision of protecting population. SDI could then be attacked as an unrealistic and overly expensive way of protecting missiles. The issue became confused because some in Congress wanted to redefine the program so that it would lead to the deployment of silo defenses, while others believed that the goals of the program should continue to be defined in accordance with the president's vision of population defense.

Meanwhile, SDIO was also caught in the crossfire between the executive branch, which was forwarding a reinterpretation of the ABM Treaty, and an outraged Senate. SDIO originally assured Congress that its experiments would be configured so as to conform to the treaty, although Abrahamson and other DOD officials pointed out that at some point testing in space would have to take place if the validity of developing space-based systems were to be established. Then, when the administration wanted support for its reinterpretation, SDIO was asked to prepare a list of experiments that could not be done under the "narrow" interpretation. The SDIO therefore reported that the narrow interpretation would result in the waste of billions of dollars by prohibiting experiments that could save time and effort. Beset on two sides, Abrahamson had to walk a tightrope so as to satisfy supporters and influential critics alike, supporting the administration's policies with one hand while promising with the other to obey congressionally mandated restrictions.

Abrahamson probably became a more credible figure by allowing SDIO officials to disagree publicly. Thus, when others were urging early deployment of a partial shield, Louis Marquet, SDIO's technical director, said that by the time an early deployment of kinetic-energy satellites could be made, the Soviets could counter it by going to fast-burn boosters.^[26]    Marquet's comments contradicted arguments of SDI supporters, including a former deputy to Abrahamson, Col. Simon P. Worden, urging early deployment and contending that it would take the Soviets longer to build countermeasures and cost them more than they could afford. Marquet's candor, coupled with the care taken by Abrahamson not to enter the policy thicket himself, helped to preserve the credibility of the program in Congress, even though it did not prevent substantial and increasing cuts in appropriations requests.

In the longer run, it remains to be seen whether SDIO can survive as a separate entity. The more the defense budget is restricted, and the more skepticism is generated about SDI, the more pressure is likely to be

felt to reintegrate the elements of SDI into DARPA and the services so that the Department of Defense can make better judgments about tradeoffs between SDI and other R&D programs. In the short run, however, the establishment of SDIO was a politic move. It prevented the losses caused by such integration and enabled the program director to create a network of cooperative officials and agencies, more or less along the lines of the successful ICBM and Polaris projects of the 1950s and 1960s. In an effort to reinforce its status and perhaps to help assure longevity, SDIO has proposed the creation of a Strategic Defense Initiative Institute, in the belief that such a body would undertake useful, arm's-length reviews of contractors' work—much along the lines of the Aerospace Corporation, which reviews work done for the air force. Critics in Congress charge that the real aim is to create a public-relations agency for SDIO's activities. The proposal has so far been rejected.

The Debate Among The Scientists

In the debate over SDI, as in the earlier controversy over ABM, opposition from scientists has played an important role in shaping legislation and public opinion. Disagreements over defense policy within the nation's scientific community have become commonplace since the end of World War II, when scientists and engineers began to play a regular and important role in the nation's defense effort. Many scientists and engineers are employed in defense work—25 to 30 percent by one rough estimate^[27]    —and even those who are primarily engaged in nondefense employment contribute indirectly to the defense effort because their work has military applications and because some consult for defense agencies. In view of the importance of their contributions, many have come to feel a special responsibility for taking part in policy debates. Some have gone further, either deciding to opt out of military work or out of all science on the ground that their work could be misused. A few have taken matters into their own hands and practiced resistance, particularly during the controversy over the Vietnam War. Generally, however, most members of the scientific community acknowledge that their expertise gives them no special knowledge of matters of foreign and defense policy. A leadership elite, however, has been very active. Those who generally seek to limit the arms race or openly oppose many new weapons tend to join such organizations as the Federation of American Scientists and the Union of Concerned Scientists. Those who generally support modernization tend to serve on official committees or

to work for government and industrial laboratories; they sometimes support candidates for office who favor a strong defense program.

SDI's most vocal promoters have been a small group of physicists who work at the two major weapons laboratories—Los Alamos and Lawrence Livermore. There is also strong (though much dispersed) support in industry, "think tanks," and in the government. At Livermore, Edward Teller has been an inspiring leader of some of the key researchers in part as director of the Hertz Foundation, through which some of the most talented young physicists have been recruited to the laboratory.

The role is one to which Teller has grown accustomed. He has long been a believer in the exploitation of technology for military purposes without restraint. While the fission bomb was still under development, he agitated at Los Alamos for permission to work on the next stage, the thermonuclear bomb. This permission was granted by the laboratory director, J. Robert Oppenheimer, once the key theoretical task involved in the fission bomb was sufficiently accomplished. A few years later, when Oppenheimer chaired the GAC (which recommended against a crash program to develop the H-bomb), Teller criticized him passionately, accusing him not only of bad judgment (which, as he would later testify at Oppenheimer's security hearing, was ground for concluding that he was a "security risk") but also of interjecting political and moral views when asked his scientific opinions. He believes that the United States is in a life-and-death struggle with Soviet communism and that the best U.S. strategy would be the unrestrained pursuit of technological advantage.

Teller also believed in the virtues of defense, as we note in chapter 1. In the 1950s, he was an advocate of civil defense at a time when it was thought the Soviets were developing a bomber force that could strike at the U.S. mainland. As the Soviets developed a large ICBM force capable of devastating U.S. urban centers, most of those who had become interested in civil defense abandoned the cause, convinced that it would be a futile, tremendously expensive way to protect even a fraction of the population. Some also argued that building defenses would be interpreted as provocative because a commitment to do so could indicate willingness to contemplate a first strike. Teller disagreed and urged the adoption of a serious program to build shelters and to disperse industry and population.

Teller has remained consistent in his support for continuing improvements to defensive and offensive weapons. He has argued that some defense is better than none, and that all forms of defense, including both

passive and active measures, are desirable and even critical to the credibility of a deterrent posture. He opposed the limited test ban of 1963 because it would seriously interfere with U.S. development of an ABM (then always thought of as employing nuclear explosives). More generally, he has opposed the proposal for a comprehensive nuclear-test ban on the ground that it would inhibit research to develop newer atomic weapons of all kinds, including types that would produce less fallout and be configured for greater accuracy and thus less collateral damage. He also opposed the ABM Treaty on the ground that it would prevent the United States from developing new systems that might be more effective than those known in 1972 for intercepting nuclear attacks.

As we have noted, however, Teller did not endorse the Fletcher Committee's recommendation that a decision on deployment be deferred pending further research. He argued that elaborate space-based systems were vulnerable to countermeasures and were far too expensive as well as unlikely to provide a truly leakproof defense. For the same reason, he also declined to endorse High Frontier's space-based KEW system, arguing instead for terminal defenses, which were feasible and would provide affordable protection to some degree, and for a high-priority development program, mainly involving the X-ray laser.

The main political push for the president's program had come initially from the High Frontier organization, which was the brainchild of Lt. Gen. Daniel O. Graham, and which came to include a small group of relatively second-rank industrial military researchers. In 1984 a document marked "not for release" and apparently prepared under the auspices of the Heritage Foundation (though an official of that organization later attributed it to High Frontier) was obtained and widely circulated by SDI critics. It was seized upon with relish by these critics because it seemed to reveal a cynical and conspiratorial strategy for promoting SDI, which bore at least a superficial resemblance to actual events. The document, reportedly prepared by a consultant, set three goals for the campaign: to achieve "initial startup" of deployment during the second term of the Reagan administration, to develop sufficient momentum behind the project so that "it could not be turned off by a replacement or successor Democratic administration," and to foster the creation of "a broad political constituency favoring a new U.S. arms-control strategy featuring major BMD deployments at an early time." The strategy called for a unilateral U.S. effort to develop BMD to be couched as a "new approach to nuclear arms control" which "could be represented as a bilateral effort—one with Soviet reciprocation and participation." Among the virtues of this approach, the author argued, was that it could

"disarm BMD opponents either by stealing their language and cause (arms control), or … [put] them into a tough political corner through their explicit or de facto advocacy of classical anti-population war crimes." The author recommended that High Frontier enlist an "offshore" constituency among major allied governments and find some institutional "home" for the project outside its own ranks. This might be done by identifying a "forceful personality" who might champion the cause (Henry Kissinger was mentioned as a possible candidate), by inspiring the creation of a summer study group, or (in the author's opinion the best option) by organizing a group across the political spectrum in order to remove the perception that "BMD is primarily a 'right wing' cause." Earmarked for recruitment to the cause were conservative columnists, members of Congress, and defense analysts, "neutral" groups such as "pro-Israel political circles," financial and professional organizations, and—most ambitious of all—even such already committed groups as the FAS, SANE, freeze groups, and the Center for Defense Information.^[28]   

Graham did not succeed in enlisting nearly so broad a constituency, but he did find allies in Congress and in the White House. Defense Department officials turned a polite but deaf ear to High Frontier proposals until the president adopted the cause. From then on, the DOD reversed course until, in short order, its leaders were arguing that an intermediate deployment such as that advocated by High Frontier would make sense and that eventually beam weapons would be added to the system.

When critical reports began to appear, starting with the report for the Office of Technology Assessment prepared under commission by the physicist Ashton B. Carter^[29]    and a report prepared by the Union of Concerned Scientists,^[30]    under the direction of physicists Hans Bethe, Kurt Gottfried, and Richard Garwin, other defenders of SDI began to be heard from. One spokesman was Robert Jastrow, a professor at Dartmouth who had been director of the NASA Goddard Space Flight Center. Using data supplied by the pro-SDI scientists at Los Alamos, Jastrow disputed the figures that the UCS used to show that an SDI system would be impossibly expensive.^[31]    The UCS had calculated that in order to achieve adequate coverage against the existing Soviet ICBM force, the United States would have to orbit a fleet of more than two hundred satellites. Jastrow claimed the real number was closer to one hundred. The UCS conceded that its original calculation was in error, pointing out that the mistake had been corrected before its testimony was presented to Congress, but it also argued that Jastrow's numbers rested on

unrealistic assumptions about the brightness that could be achieved by chemical lasers and other relevant parameters. This phase of the debate was soon superseded, when attention shifted away from chemical lasers based on battle satellites toward a combination of kinetic weapons parked in "garages" in space and beam weapons fired from ground-based systems such as the free-electron or excimer laser using presumably cheap mirrors capable of being proliferated at low cost.

The anti-SDI forces were spearheaded by the UCS and the FAS. The FAS maintained a resident space specialist, John E. Pike, to comment on the program and to study its various facets, including the relationship of SDI to the ABM Treaty, the pattern of contracting, and the relationship of SDI to arms control. Both the UCS and FAS issued a drumbeat of reports critical of the project.

A subsidiary theme in the quarrel over the SDI has long been the question of its applicability to the nation's civil industry. Supporters, notably Keyworth, argue that SDI would be a tonic for U.S. high-technology industry. Knowledgeable skeptics like Harvey Brooks complain that SDI is an altogether inadequate substitute for a national science policy. Brooks notes that military projects tend to be directed toward narrow objectives and to yield products and processes that are of limited value in the civil economy, partly because of security restrictions but also because of the growing incompatibility of military (especially space) technology and civil needs.^[32]    John P. Holdren and F. Bailey Green argue that far from promoting useful spin-offs, SDI is likely to divert critically needed talent and resources from the civil sector and make the country even less competitive in civil markets.^[33]   

The anti-SDI forces also gained strength from the circulation of a variety of petitions and opinion surveys. A petition circulated by a group of physicists at Cornell University and the University of Illinois at Urbana-Champaign among physics departments in the nation's universities was signed by some 6,500 faculty members and graduate students, who pledged not to accept funds for research on SDI. The signers included majorities of the faculties at the nation's twenty leading departments.^[34]    In response to a questionnaire sent to members of the prestigious National Academy of Sciences (NAS) a large majority of respondents declared SDI's goals neither feasible nor desirable.^[35]    Another petition, initiated at AT&T Bell Laboratories and circulated among researchers employed in government and industrial laboratories, called for a curb on appropriations for SDI (but did not commit the signatories to boycott SDI-related projects). This petition, which had 1,600 signatures, took the form of an open letter to Congress urging that support for SDI

be limited "to a scale appropriate to exploratory research."^[36]    A poll taken by the UCS of 549 randomly selected members of the American Physical Society found that SDI was criticized as "a step in the wrong direction" by 54 percent of the respondents who did not oppose other military R&D; 29 percent were in favor.^[37]   

A counterpetition was circulated by a group of eighty scientists and engineers led by Frederick Seitz, a former NAS president. This petition chided those who opposed SDI for being unfaithful to the scientific method. Instead of condemning the project in advance, "hastily, unscientifically, or ideologically," the petitioners urged their fellow scientists to allow it to continue until the prospects for strategic defense could be fairly evaluated.^[38]   

Conflict over SDI also flared dramatically into the open in 1988 when Roy D. Woodruff, associate director of defense systems at Livermore, made public his complaint that Teller and others at the laboratory were passing on their own overoptimistic assessments of progress on the X-ray laser to senior officials, including Nitze and McFarlane, as though they reflected the views of the laboratory. In October 1985 Woodruff had resigned his position in protest, charging that he had been put in an "untenable position" because he had been denied an opportunity to show that Teller and Wood were "overselling" the program. Although Woodruff was subsequently reassigned to a new post, and a General Accounting Office inquiry found that several Livermore officials, including Woodruff himself, had made statements about the X-ray laser that had been no less optimistic than Teller's, it was also reported that Woodruff had specifically taken issue with Teller's claim, as early as December 1983, that research had been so successful that the laser was ready "for engineering." Woodruff eventually decided to file a personal grievance with the president of the University of California (which operates the laboratory under contract to the Department of Energy) that he had been "constructively demoted" because Teller and Wood "undercut my management responsibility for the X-ray laser program and conveyed both orally and in writing overly optimistic, technically incorrect statements regarding this research to the nation's highest policy makers."^[39]   

Scientific Judgment And Political Commitment

What significance should be attached to the expressions of dissent among scientists not at work on SDI is an intriguing and difficult question. In what proportions does the controversy among the scientists and

engineers reflect scientific judgment or political commitment? Many of the physicists who signed the Cornell-Illinois petition are probably not directly conversant with defense research and receive their research support from the National Science Foundation and other nonmilitary agencies. Strong opposition at the universities has not prevented SDIO from placing a large number of projects with university scientists. SDIO officials even claimed to have set up university-industry consortia, though this claim aroused protests from university administrators who noted that contracts accepted by individual researchers did not imply institutional endorsement. Some of the opposition among scientists clearly does not arise solely from scientific judgment. Many university physicists were apt to oppose the Reagan administration and its policies, including SDI, on political grounds. Even many of the critics agree, moreover, that some level of expenditure on strategic-defense research is warranted as a hedge against the possibility that the Soviets might stage a breakout from the ABM Treaty and might deploy certain space-based defensive technologies. Some also suggest that the United States might profitably consider the deployment of defenses aimed at protecting retaliatory capacity and command and control.

A similar division of scientific opinion was expressed in congressional hearings on the proposed deployment of ABMs in the 1960s. Then, too, as Harvey Brooks notes, policy views were couched in the form of technical judgments to suit political purposes:

Many of the technical witnesses, on both sides, were really motivated by strategic policy considerations, or their personal evaluations of the international situation, or the supposed intentions of the Soviets, but their political allies found it more politic for them to couch their arguments in narrow technical terms, partly because technical experts are often automatically regarded as having nothing useful to say on policy matters. Furthermore, technical testimony appears more "objective" and politically neutral, and it is thus thought to carry more weight with those politicians who have not yet made up their minds.^[40]   

Professional and personal interests are also at play. Some of those who do military R&D are no doubt committed to SDI because they see it as a source of continued funding, especially if the need for research on offensive weapons tapers off. In view of the many technical problems that remain to be addressed in order to perfect offensive systems, however, such self-interest seems an unlikely explanation for the attitudes of researchers—although industrial laboratories and firms may have their eyes on the vast potential represented by defensive procurement contracts.

Some scientists may believe, as David L. Parnas has charged, that even if nothing militarily useful can come of SDI, something beneficial can come for the specialties in which they work.^[41]    In that sense professional curiosity and self-interest may be a motivating factor for some of the scientists. In the case of some of the younger scientists, it is also possible that a desire to do something new and better than their elders is at play. Some of the scientists who work on military projects generally, including SDI, believe they have a duty to contribute to the national security of the United States and the security of the West generally, because, generally speaking, the United States has used its power for benign purposes whereas the Soviet Union is an aggressive, expansionist power whose leaders cannot be trusted. Still others view the world as a system of sovereign states with little law and no law enforcement governing the relations among them, and they argue that even if they did not do the research, someone else would. Motives are undoubtedly as varied and complex in this area as they are in others.

On the side of the critics, there seems to be a more explicit political motivation. Generally, those who are ready to renounce funding for SDI and are otherwise dubious about the motivations behind SDI argue that the political forces promoting it are eager to avoid arms control because they do not fear nuclear war as much as they fear Soviet superiority. In their view, the aim of SDI is not the president's benign goal of making nuclear weapons "impotent and obsolete" (because that is widely recognized to be unattainable) but a panoply of much less benign motivations: to challenge the Soviets to an expensive race in high-technology weapons development that they would lose, to establish a military presence in space for offensive as well as defensive purposes, and to sustain the military-industrial complex by promoting a new phase in the military competition between the superpowers. Whatever the goal, they argue, the result of the competition would be to make the world less stable and to promote the chances of nuclear war, because the Soviet response would include the development of a stronger offensive force and resistance to any proposals for arms control. In general, they perceived the Reagan administration to be guided by an obsessive anticommunism and mistrust of arms control, ironically at a time when, because of a major change in the Soviet leadership, the U.S.S.R. is more open than ever to the possibility of serious arms reduction and a renewal of détente with the West. If SDI had not been deliberately contrived for aggressive purposes, they suggest, the administration should have been willing to trade it away as a bargaining chip in negotiations.

The report prepared by a panel appointed by the American Physical Society (APS) suggests that despite these very real political differences—which produce powerful effects in promoting policy commitments—the structure of scientific evidence and thinking is such that consensus is possible on a good many, if not all, of the main technical points. Scientists are apt to disagree about what can be accomplished over what span of time and at what level of effort, but they do not disagree much about what is known and what is yet to be accomplished. The APS report makes it clear that some of the proposed technologies are at least theoretically conceivable, but it also points out that the most important of them are still orders of magnitude away from producing the energies or power necessary to be useful in defensive systems. Even if the technical parameters are met, the report cautions, it remains to be seen whether the devices will prove feasible as weapons or will be less expensive to deploy than offensive systems that could overwhelm or evade them.

In political terms, the APS report was widely considered hostile to SDI. Journalistic accounts emphasized the report's conclusion that it would be many years before the feasibility of space-based beam weapons could be demonstrated. That much was correct, and insofar as SDI has been perceived as a program that promises a quick answer to the question, the report was properly considered critical. If SDI is perceived as a research program, however, the report is not so much a critique as a status report. Nevertheless, it did not satisfy enthusiasts for SDI. In a presentation to the House Republican Research Committee in May 1987, two active SDI researchers, Lowell Wood and Gregory Canavan, criticized both the press coverage of the APS report and the report itself.^[42]    As we note in chapter 3, their charges provoked a rebuttal that left the report's findings undisturbed.

The SDI is only the latest chapter in a division among physicists that opened once the atomic bomb was developed and used against Japan. Until then, disagreements had been muted for the sake of winning the war against Nazi Germany. Once the war was over, a split developed that reflected differing appraisals of the need to gain control over atomic warfare and the possibility of rapprochement with the Soviets. Scientists took different views of defense policy, first with respect to containment, then with respect to the strategy of deterrence and arms control. Many have continued to work on weapons even though they may have disagreed with various aspects of government policy. They do so partly to continue to have some access to policymakers and thus to have influence over policy, partly because they are fascinated by the technology, but in

general because they are committed both to the design of military technology and to the effort to achieve their different policy goals. When Hans Bethe came to Livermore and acknowledged that the idea behind the X-ray laser was scientifically sound, Teller felt vindicated. In an open letter to Bethe in 1986, Teller criticized the scientists who had opposed SDI in the NAS poll, lamenting that the "World War II unanimity of the scientific community" in developing nuclear weapons had been lost. "I am writing this letter," Teller asserted, "in the hope that you and others may find some way to move from polemic debate and confrontation toward technical criticism, understanding, and cooperation."^[43]    Behind the show of indignation, however, lies a reluctance to acknowledge that more is at stake than the accuracy of a scientific theory or line of work. Teller and Bethe—and the scientists who align themselves with both—disagree about what is to be done with such weapons and how their use can be prevented. Both sides are inevitably enmeshed in a political struggle to organize constituencies for and against SDI.

Organizing A Constituency: SDIO And The Contractors

As Erik Pratt, John E. Pike, and Daniel Lindley have pointed out, although the formal rationale for SDI is that it is a research program to determine whether comprehensive strategic defenses are feasible and desirable—a program that is not intended to lead to a decision to develop or deploy until after the term of the Reagan administration—the expenditure of funds in this phase of the project will help build a constituency in favor of its continuation:

Before a formal deployment decision can be made, SDIO will have spent billions of dollars and involved hundreds of institutions and thousands of individuals. In the process, many of these people and organizations will have acquired vested interests in SDI and in the deployment of strategic defense systems. For many government organizations, research labs, and universities, SDI will provide a large share of the budget. For defense firms, SDI will be the source of significant profits. For some individuals, careers and promotions will be determined by the progress of the strategic defense program. As a result, these vested interests will become part of the "Star Wars constituency," seeking to influence policy and control decisions concerning strategic defense.^[44]   

This statement puts the argument sharply, perhaps too sharply. Not all of the institutions and individuals engaged in SDI work have an exclusive interest in its perpetuation. They are often also engaged in other

types of military research. For instance, the nuclear-weapons laboratories did not expand to accommodate work on SDI. As the program grows, it attracts people from elsewhere in the laboratories and, as an SDI project is completed, they may well shift to other work, so they do not necessarily have the same vested interest in SDI as other organizations would whose entire raison d'être is predicated on it. It is misleading to suggest that corporations engaged in SDI work depend on it for profits. Whereas military procurement has had a "critical role" in the growth of defense industries, military R & D spending has had "a much more modest role."^[45]    It is only insofar as R & D work may lead to procurement that profits enter the picture in a significant way, and in some cases the firms that are likely to secure the procurement contracts are not the same as those that carry out the R & D. Nevertheless, there is good reason to suppose that the expenditure of large sums of money on SDI will create a constituency interested in perpetuating it. By the time a decision must be made to move to the stage of development, at least $20 billion is likely to have been spent on the research phase. The development stage would involve far higher expenditures.

For most of the contractors, SDI's importance is accentuated by the fact that several of the large strategic programs, such as the B-1B, MX, and Trident II, are now being completed. New "black" or secret programs—like "stealthy" bombers, fighters, and cruise missiles—have been taking up any slack. Had it not been for SDI, however, funding of military R & D might have been reduced. As the new systems are deployed, SDI could become the major source for continued military R & D funding and for deployment contracts. Because deployment could involve much larger sums—estimates range from hundreds of billions to a trillion or more dollars—the stakes could be very high.^[46]    Interest in the defense contracting community in SDI has been keen. The contractors sent 1,200 representatives to an initial classified briefing. During the first three years of the project, 2,210 SDI-related contracts were awarded to 510 different entities.^[47]    These entities are located in the private, university, and government sectors and also include federally funded research and development centers (FFRDCs) as well as foreign contractors. Some 331 business firms and 15 nonprofit corporations have been engaged in SDI research, along with 14 of the 38 FFRDCs, entities that perform research on behalf of government agencies but are administered by nonprofit organizations. Two such FFRDCs are the Livermore and Los Alamos laboratories, both of which have played leading roles in SDI research. SDI work has also been performed at

eighty-one campuses and in sixty-seven U.S. government agencies. The largest share of the funds—almost three-quarters—has gone to the private contractors. Another 20 percent has gone to the FFRDCs. The remaining 5 percent is divided among the others, with universities accounting for 2.7 percent, U.S. government agencies for 2.3 percent, and foreign contractors for only .4 percent.^[48]   

Among the private contractors, awards tend to be concentrated among relatively few firms that inevitably develop an interest in promoting various SDI programs. The top one hundred SDI contractors account for 95 percent of the total, the remaining four hundred plus for less than 5 percent. This concentration is even higher than is generally the case in defense work, where the top one hundred contractors receive 88.3 percent of total prime contract awards. Ten firms alone account for 53.1 percent of total SDI awards, five for 42 percent.^[49]    More than 90 percent of space weapons' prime contracts in FY1983 went to four states: California (45 percent), Washington (22 percent), Texas (13 percent), and Alabama (10 percent); 77 percent went to states or districts represented by House and Senate members who sit on the armed services committees and appropriations defense subcommittees. Thus, the program promotes possibilities for conflict of interest for both legislators and contractors. Contractors with much to gain "have been assigned the task of both developing and assessing the technical feasibility and strategic advisability of proceeding beyond the research and development phase."^[50]   

As the program moves out of the concept-definition stage—in accordance with the "horse race" scheme adopted by Abrahamson—fewer contractors are selected for the succeeding stages, and the value of the contracts grows larger. The horse-race idea was originally employed for the study and development of alternative architectures. Contractors would be eliminated when a particular architecture (or a smaller set) was chosen for more detailed analysis. Four contractors were awarded Phase I contracts for the development of the National Test Bed. Two teams, one led by Martin Marietta, the other by Rockwell, were selected for the second phase, and Martin Marietta was chosen as prime contractor for the construction of the facility, possibly worth $1 billion. Because the prime contractors distribute some of these large amounts to subcontractors, the constituency remains fairly large, despite the concentration of the awards, as measured by prime contracts.

Although SDI support for university research remains relatively small, it contributes to a growing trend toward increasing dependence of university

researchers on military R & D support. As a fraction of all federal support for university R & D, the DOD share rose from 11.6 percent in FY1980 to 16.7 percent in FY1986. In constant dollars, DOD funding for university research during the Reagan years nearly reached the level attained in the peak years of the Vietnam War.^[51]    This dependence could dampen university protests against SDI, particularly in view of the comments by at least one DOD official, Donald A. Hicks, under secretary for research and engineering, who pointedly noted that although university researchers may be free to criticize DOD programs, the DOD is free not to provide them with support. During his Senate confirmation hearing on July 25, 1985, Hicks said: "I am not particularly interested in seeing department money going in someplace where an individual is outspoken in his rejection of department aims, even for basic research." In an interview with a reporter from Science, Hicks confirmed his belief that critics of defense policy should not expect research support from the DOD: "If they want to get out and use their role as professors to make statements, that's fine. It's a free country, [but] freedom works both ways. They're free to keep their mouths shut, [and] I'm also free not to give money." The DOD later claimed that Hicks was expressing his own views rather than department policy,^[52]    but the warning had been delivered.

The general point properly made by critics is that all support, even in relatively modest amounts, can serve to co-opt potential critics, or at least generate self-interested acquiescence, both domestically and abroad. But the point should not be exaggerated. An offsetting factor is that many of the largest firms are so large that SDI work constitutes only a small fraction of their total budgets. General Motors (GM), the second-largest SDI contractor, had net sales in 1985 of more than $102 billion. Less than one-tenth of 1 percent of the total came from SDI work done by Hughes Aircraft, a division then newly acquired by GM. Even for Hughes, however, which had annual sales of $6 billion, SDI became more significant in 1985 and could become still more so. For a number of other firms, SDI already represents 20 percent or more of their R & D contracts. By "buying in" early, they can hope to receive even larger amounts in the future and thus gain an inside-track position from which to compete for far more lucrative procurement contracts. For smaller firms, SDI contracting can be a very substantial part of their business—without which they would not survive. One such company, Sparta, whose president had served on the Fletcher Committee, had a nearly 1600 percent sales increase between 1981 and 1985, owing in large part to SDI contracts. Close to half the companies highly dependent on SDI

R&D funding are located in Huntsville, Alabama, near the army's Strategic Defense Command.^[53]   

Four FFRDCs perform 96.2 percent of all SDI work awarded to these institutions: Livermore, Los Alamos, Sandia, and Lincoln laboratories. At the first two, SDI-funded work grew to about 12 percent of the budget by 1985; at Lincoln, which does virtually all of the work on SATKA, the SDI budget grew to 22 percent of the total.^[54]    At these laboratories, government-established priorities set the pace for research efforts, and SDI is considered a major challenge, particularly for younger researchers. The competition among the laboratories, especially between Livermore and Los Alamos, is clearly evident in their competing designs for a free-electron laser. This rivalry is helpful because it provides critical reviews of ongoing programs and a stimulus to advance the project. Los Alamos scientists, for example, raised questions about the quality of Livermore's X-ray laser research and testing. The involvement of the various laboratories and industrial contractors makes it inevitable that those involved in the work will have a significant degree of influence over the evaluation of what has been achieved and the decision to move on to the next stages. Defense-review panels rarely include critics. At the same time, the greater the involvement, the less likely that laboratory representatives will favor terminating the projects altogether.

In addition, conflicts of interest arise because so many former military officers and DOD civilian employees enter employment with defense contractors once they leave their Pentagon positions. Although federal law requires that such employees notify the DOD if they join a defense contractor within two years of leaving the government's employ, a GAO study in 1986 concluded that only 30 percent of those who were required to file a report actually did so and that many others were exempt from reporting on technical grounds. The DOD, which has no institutional interest in closing the revolving door, has not been active in enforcing the requirement. In 1985, of the ten contractors with the most former DOD employees, nine were major SDI contractors. Several key figures in SDIO had previous jobs with defense industries and have taken subsequent positions with firms heavily engaged in SDI work. Although this flow between DOD and the laboratories and industrial firms is inevitable as both seek to obtain the best-qualified personnel, the result is that a strong relationship of shared assumptions is created, which is bound to affect judgments about research priorities. Thousands of jobs are already at stake, and a drastic cutback in SDI would have serious consequences.^[55]   

The existence of such a constituency is certainly a real enough phenomenon,

and its potential role is evident from previous cases, when defense contractors helped to stimulate initiation and perpetuation of projects that might have languished on purely military or economic grounds. Whether such constituency pressure will be enough to promote an increase of support for SDI is at least questionable, at this stage, especially in view of the opposition SDI has met in Congress. The existence of a constituency is not an automatic guarantee of its success, especially given the general interest of that constituency in defense funding, which can be spent on systems other than those under study in SDI. Even with respect to strategic defense, this constituency has actually restrained the more radical proposals for SDI. Staunch supporters of SDI have argued that the program is suffering in the hands of defense officials who do not share Reagan's belief in the possibility of a fundamental shift away from offense to defense, and a leakproof defense at that. Thus, the constituency could just as well frustrate the more radical dreams of SDI advocates as promote them. By emphasizing near-term options that can serve mainly to protect retaliatory capacities, they may actually be undermining support for the president's emphasis on a long-term effort to develop population defenses.

It is equally a mistake to suppose that all those who are in the employ of FFRDCs and industrial contractors are venal and will thus support any program that promises to provide them with research funding. They must recognize that funds for one program may come at the expense of others. They may have conflicting interests, which could lead them to prefer non-SDI programs to SDI, particularly if they work on technologies that lose out in the research phase. Those who work at FFRDCs are presumably freer to put their professional judgment ahead of their vested interests, if only because they cannot benefit from follow-on procurement contracts. When decisions have to be made to enter into development and procurement, their independent judgment could be crucial in blocking efforts by industrial contractors to promote the adoption of programs that are not likely to work.

Still, the actual experience of such costly failures as the division air defense (DIVAD) anti-aircraft gun suggests that the DOD does not have adequate mechanisms for ensuring unbiased appraisals and may be too reliant on the judgments of contractors who work closely with procurement officers. Both groups have a vested interest in not canceling expensive projects, even though they are not performing as expected. In order to assure that decisions with respect to development and deployment of strategic systems are not unduly influenced by the assessments of their

sponsors, better reviewing mechanisms may well be needed. The experience with DIVAD indicates that the influence of supporters can keep a program going even though it is not meeting prescribed goals. But it also shows that the truth cannot be suppressed indefinitely. Experience with strategic defenses is similar. Twice before, the United States has gone a long way down the road of deploying defenses. Each time, the effort was strongly supported by vested interests. Each time, however, the nation drew back when technical considerations as well as military and strategic factors made the efforts seem unattainable.

Congress And SDI: The Battle For The Future Of The Project

The Reagan administration's campaign for SDI ran into difficulty for several reasons. It was launched at a time when Congress was generally eager to assert a stronger constitutional role in foreign policy—an attitude originally generated by the Vietnam War but strengthened during the nuclear-freeze campaign when many members of Congress came to feel that they alone represented their constituents' concerns at a time when the administration was ignoring them. Arms control was so low on the agenda of the Reagan administration that many Democrats in Congress came to feel that if it were to be promoted actively, they would have to be the ones to do so. Members of Congress were influenced by the strong opposition to SDI from the scientific community and by the skepticism they knew prevailed within the Pentagon itself. In addition, their own Office of Technology Assessment produced reports that cast serious doubts on the project. Many in Congress seemed to conclude that although they might risk the ire of the voters by challenging the entire basis of SDI, they would be safe in taking a skeptical attitude by keeping appropriations below those requested and by insisting that the program conform to the ABM Treaty. Although the president was able to quiet some of the opposition in Congress by arguing that it ought not to tie his hands in negotiations with the Soviets before Reykjavík, the failure of that meeting to produce an agreement and the administration's insistence on promoting its reinterpretation of the treaty provoked a strong response from the Democrats, especially after they gained control of both houses of Congress in the 1986 election during which the president had made SDI an issue. When the voters ignored his request for support, SDI critics were all the more emboldened.

In the immediate aftermath of the president's SDI speech, both support

and opposition appeared in Congress, but neither was very strong. The division developed along party lines, particularly as the issue was injected into the 1984 election campaign. When Mondale charged that the proposal would militarize the heavens, the president responded that SDI was a research program—something even Mondale admitted was advisable—and argued that it was right to try to replace mutual assured destruction by a system that promised mutual assured survival. The president's personal popularity was a major factor in his reelection, but his policies were also popular. On the domestic front, the economy had revived after a shaky start, and the tax cut was widely approved. In foreign affairs the president's bold approach had restored a sense of direction and patriotism, even though there were significant failures not only in arms control but also in the abortive peace-keeping effort in Lebanon in 1983. SDI did nothing to detract from the president's electoral appeal. Accordingly, congressional opposition was muted.

As SDI became a more contentious issue, both in arms-control negotiations and domestically in the appropriations process, a group of Republicans emerged as strong champions. Senators Wallop, Quayle, and Pete Wilson, and Representatives Kemp (N.Y.) and Jim Courter (N.J.) came to be known as the "Gang of Five." In 1986 they became convinced that the anti-SDI forces had settled on a strategy of not attacking SDI directly and vigorously but, instead, of allowing it to proceed at only a low level of support. If no decision were made during the Reagan administration to exceed the testing limits of the ABM Treaty or to schedule deployment, a new administration might kill the program altogether. Otherwise, the program might simply wither away, as momentum was lost and public enthusiasm waned. The same fear was shared by others sympathetic to SDI. SDIO's former chief scientist, Gerold Yonas, wondered if the U.S. public had the patience to support a long-term project that had no immediate payoff. "It's the ultimate Catch-22," he observed: "Congress won't provide the funding unless you are serious. And yet you can't be serious without funding. The fundamental problem is with organizational psyche. We want instant gratification. Rather than discuss early deployment, the nation should debate capability, survivability, and feasibility—those are the real issues."^[56]    Experienced legislators like former president Gerald R. Ford warned that Congress would not continue to vote large amounts of money for SDI unless it could be assured that the research phase would be followed by deployment.^[57]   

To counter the presumed strategy of the Democrats, congressional supporters of SDI urged the administration to adopt the "broad interpretation"

of the ABM Treaty so as to permit advanced testing and, if necessary, to consider abrogating the treaty altogether. Late in 1986 Senator Wallop and Representative Kemp wrote a letter to the president in which they noted that to defer a deployment decision for five to seven years, as originally contemplated, "would place the United States in a no-win position and the Russians in a no-lose position." In an article in the conservative National Review , Kemp argued that SDI was the most important single initiative of the Reagan administration and that it would be the key issue of the 1988 presidential campaign. He made it clear that he at least was prepared to make it the key issue in his campaign for the Republican nomination. Kemp argued that the Democrats in effect should be "smoked out" on the issue, forced to reveal their true opposition to SDI rather than be allowed to appear supportive when in fact they were attempting to starve the program and stretch it out.^[58]   

Among the opponents of SDI, the leading figures were a number of Democrats, including Alan Cranston (California), J. Bennett Johnston (Louisiana), William Proxmire (Wisconsin) in the Senate, and George Brown (California) and Thomas Downy (New York) in the House. Proxmire and Johnston, along with senators Lawton Chiles (D., Fla.) and Dale Bumpers (D., Ark.), instructed members of their staff to prepare reports on the actual progress of the program. Three reports were prepared in consecutive years. The first, submitted March 17, 1986, reported on the basis of interviews with researchers that, contrary to some reports, there had been "no major breakthroughs" that promised to make deployment in the 1990s of comprehensive defenses any more feasible than it had been at the start of the project. The staff members found that the "schedule-driven" nature of the research had, in fact, aroused apprehensions among scientists at the national weapons laboratories by threatening to compromise long-term research to suit an arbitrary schedule and to promote a public-relations style of research. (SDI, said one senior Livermore scientist, was degenerating into "a series of sleazy stunts.") The results so far, they reported, had shown that the job of creating a boost-phase defense would be even harder than initially thought, and that other problems, such as the logistics of orbiting the needed hardware, were even more daunting because of the shuttle accident. In general, the report noted that although SDI continued to have broad bipartisan support as a research project, the pace at which it ought to be pursued and the direction it should take remained contentious issues.^[59]   

On March 19, 1987, the three staff members submitted a second, updated

report, in response to a request from Proxmire and Johnston, who specifically asked them to examine the effort to commit the country to a "near-term" deployment. They reported that if the president were to announce such a decision, there would be nothing ready to deploy, but they noted that SDIO was indeed reorienting the program to pursue near-term deployment, possibly in the 1994–95 time frame, even without a formal presidential request. Accordingly, SDIO had cut back its directed-energy program drastically and increased the kinetic-energy weapon budget. In a number of cases, more innovative technologies were being scaled down to permit greater emphasis on the near-term option. In particular, the office was emphasizing the rapid development of the space-based kinetic kill vehicle (SBKKVs) to pursue a near-term deployment in the mid 1990s. Such a system would have a more limited capability than the one previously discussed as essential for boost-phase interception. They had been informed that SDIO had a "black" program for "developing a reference architecture for a near-term deployment of strategic defenses"—a program hidden from most members of Congress. Based on what they claimed was a careful review, the staff members estimated that the near-term system contemplated by SDIO would have an effectiveness of no more than 16 percent against Soviet ballistic missile warheads. The system would comprise SBKKVs—since renamed space-based interceptors (SBI)—and ground-based ERIS missiles. Again they reported that SDI researchers were extremely unhappy over the reorientation of the program toward near-term deployment. While noting again that progress had been made in the research, they found no significant breakthroughs in the effort to achieve a comprehensive defense.^[60]   

After the second report was presented, the issue of adherence to the Nitze criteria became a bone of contention between the SDIO and its congressional critics, as the third Senate staff report, issued in June 1988, made abundantly clear. Formally, the Nitze criteria were endorsed by administration officials. "Within the SDI research program," a State Department report asserted, "we will judge defenses to be desirable only if they are survivable and cost-effective at the margin. … We intend to consider … the degree to which certain types of defensive systems by their very nature, encourage an adversary to try simply to overwhelm them with additional offensive forces."^[61]    In congressional testimony on October 30, 1985, Lieutenant General Abrahamson also endorsed the Nitze criteria: "We will not proceed to development and deployment unless the research indicates that the defenses meet strict

criteria: Within the SDI research program, we will judge defenses to be desirable only if they are survivable and cost-effective at the margin."^[62]   

To ensure that the administration would adhere to these commitments, Congress enacted them into law in 1985 in the Department of Defense Authorization Act of 1986 (P.L. 99-145), adopting an amendment introduced by Senator Proxmire. The provision spells out the meaning of the criteria and requires presidential certification of their observance as a condition for recommending deployment:

A strategic defense system developed as a consequence of research, development, test, and evaluation conducted in the Strategic Defense Initiative program may not be deployed in whole or in part unless—

Senate and House conferees agreed that the terms of the law would permit deployment of particular systems that might form parts of a future strategic defense, provided that their primary purpose was not strategic defense at the time of initial deployment. Thus, sensor satellites could be deployed for conventional surveillance purposes without assurance that the Nitze criteria were being observed.

As the SDIO sought to press ahead with plans for early deployment, however, its officials sought to broaden the meaning of the criteria and to treat them as guidelines rather than restrictions. In May 1986 Abrahamson suggested that the criteria ought not to be construed so narrowly as to defeat the larger goal of national policy:

The reason that many people, including us, were worried about cost-effective at the margin is for the fundamental principle: we are trying to get the Russians to modify their behavior. … There may be many reasons for wanting to make a positive judgment that we can or cannot go forward. And it shouldn't be our criteria; it should be our best judgment about what will produce the best results for our nation both in terms of Soviet behavior and our capability. … [The Nitze standard] shouldn't be defined so narrowly as to preclude a deployment that would be sound for other reasons.^[64]   

The 1986 SDIO annual report ignored the Nitze criteria altogether and instead laid down four different criteria, much easier to meet, that would justify deployment:

Potential role in U.S. strategy

Deterrent to surprise attack and enemy escalation

Contribution to our arms control objectives

Technical feasibility

In lieu of cost-effectiveness at the margin, the report asserted that "affordability" ought to be the economic standard. By this measure, the trade-off between the comparative cost of U.S. defenses and Soviet countermeasures could simply be ignored. When SDIO was criticized in Congress for failing to maintain the second of the Nitze criteria, its rebuttal came in a May 1987 report. Here the agency restated its adherence to the criterion but sought to diminish its significance as a basis for a deployment decision by observing that it is "much more than an economic concept." As the Senate staff report noted, this formulation was a rather obvious effort on the part of the organization to gain some maneuvering room. Nevertheless, it could not escape the requirement laid down by Congress that in order to request any deployment of strategic defenses, the president must certify that both of the Nitze criteria have been met.^[65]   

The Senate report reaffirmed an earlier finding that SDIO had altered the program's original mandate: "The goal of completing the research by the early 1990s to determine if comprehensive ballistic missile defenses were feasible has evolved into making a decision by the early 1990s as to whether a 'thin' ballistic missile system embodied in Phase I is feasible."^[66]    The report charged that contrary to the agency's claim that programs earmarked for Phase I deployment would receive only 14 percent of the FY1989 budget request, these programs would actually be consuming 50 percent of the budget, if relevant program elements carrying other designations were counted. The report contended that the agency was sacrificing long-term research projects to promote near-term candidates for deployment. Furthermore, although SDIO officials had advised Congress that the costs of the contemplated Phase I deployment would be between $40 to $100 billion or more, the SDIO's own figures indicated that life-cycle costs for Phase I would be as high as $171 billion—an estimate virtually identical to the one arrived at in the

Marshall Institute study, whose authors claimed to be relying on SDIO calculations.^[67]   

The Senate report found that the proposed deployment could not be justified either on technological grounds or on grounds of survivability. The decision to press for early deployment—or, as the SDIO preferred, "phased deployment"—was not made because of technological breakthroughs: "The brightness proposed for a prototype chemical laser [under consideration for Phase I deployment] is exactly that described in the Fletcher report as 'not worthy of early deployment.'" As to survivability, the report noted that even proponents of the initiative worry that the large satellites contemplated as weapons carriers would be extremely vulnerable—"the aircraft carriers of space."^[68]   

Both the early-deployment issue and the interpretation of the ABM Treaty came in for a great deal of attention in the winter and spring of 1987. On March 11 Senate Minority Leader Robert Dole (R., Kan.), along with senators Quayle, Wallop, and Wilson, wrote to Weinberger advising him that he must submit a report on near-term applications of SDI or find funding "much more difficult." This request was formally mandated in the 1987 Defense Authorization bill. One Democrat, Sen. Ernest Hollings (D., S.C.), joined the Republicans in criticizing his fellow Democrats who called for a compromise on the ABM Treaty to permit a strategic-arms-control agreement with the Soviets. To compromise on the treaty, he argued, is to compromise and end SDI.^[69]   

The major development in Congress, however, was the decision of Sen. Sam Nunn, chairman of the Senate Armed Services Committee, to challenge the new interpretation of the ABM treaty. Nunn had become the Democrats' bellwether on defense, along with Les Aspin (D., Wis.) in the House. A political centrist, he had been known for supporting most of the Reagan strategic-modernization efforts, concentrating on examining the particulars and the implementation of the program. In March of 1987 he took to the floor of the Senate to attack the administration's effort to reinterpret the ABM Treaty, arguing that when the Senate ratified the treaty, it did so on the understanding that contrary to the administration's reinterpretation, testing in space was prohibited. For the administration now to reinterpret the treaty would therefore provoke a "constitutional confrontation" with Congress. The administration, said Nunn, was "wrong in its analysis of the Senate ratification debate, wrong in its analysis of subsequent practice … and wrong in its analysis of the negotiating record itself." Nunn castigated the State Department's legal advisor, Judge Sofaer, for not conducting a rigorous

study of the ratification process, and Sofaer subsequently conceded that Nunn was correct in this aspect of his complaint, but blamed the failure on inexperienced staff members. After examining the record of the negotiations, Nunn concluded that although there were ambiguities in the negotiating record, these were not as important with respect to the interpretive issues as either the treaty itself or the interpretation of the treaty by both U.S. and Soviet officials and that both the United States and the Soviet Union had, in fact, adopted the view that the treaty banned the development of space-based systems. Nunn also pointed out that the space-based kinetic kill vehicles contemplated for early deployment could not be deployed within the terms of the treaty because they would include a mobile system incorporating technology not in use when the treaty was signed.^[70]   

A new phalanx of opposition to early deployment and the broad interpretation coalesced around Nunn, including Johnston, Proxmire, Carl Levin (D., Mich.), and Paul Simon (D., Ill.). Budget resolutions were introduced in both the House and Senate stipulating that funding for SDI be made conditional on adherence to the traditional interpretation of the ABM Treaty. In the Senate, thirty-four Republicans signed a letter urging the president to veto any such provision—enough to sustain a veto. The Democrats rejoined that in the event Reagan decided to veto and the veto was sustained, they would move to cut funding in retaliation. Eventually a tacit understanding was reached and a confrontation was avoided.

These stirrings on Capitol Hill were not the result only of concern over SDI. They also reflected a desire on the part of many lawmakers to play a more active role in defense policy. This interest had been building for some time. Until the Vietnam War, Congress generally was content to accept a more passive role in defense and foreign policy. From World War II to Vietnam, as Edward Weisband observes, presidents were "formulating bold and forward-looking policies for the nation while an unimaginative Congress appeared to be hindering those efforts and defending special interests."^[71]    Many came to believe, even on Capitol Hill, that Congress was too decentralized and that individual members were too independent and too parochial for the legislature to play more of a role in formulating foreign policy—a role that was in any case assigned by the Constitution primarily to the executive. In 1973, however, Congress passed the War Powers Act sharply limiting the power of the executive to commit troops to battle. In 1974–75 it conducted a review of intelligence activities, which led to legislation affecting covert actions.

From 1974 to 1978 it prohibited military sales to Turkey over the objections of the Carter administration. In 1979 congressional opposition contributed to President Carter's decision not to submit the SALT II Treaty for ratification. In the Reagan administration, congressional opposition to arms sales to Saudi Arabia compelled the president to undertake an extraordinary lobbying effort.

Well before the advent of the Reagan administration, Congress's increasing activism in foreign policy led a number of observers to predict that Congress would be a force to be reckoned with in this area, and their predictions have been amply borne out. Thus, in 1976 Graham Allison and Peter Szanton observed that "congressional involvement will make the largest single difference between foreign policy making in the last quarter century, and that of preceding decades."^[72]    In terms of the scrutiny of the defense budget alone, the change has been dramatic. Congressional staffs have grown larger, and the number of hours devoted to hearings has also increased substantially. The use of staffs to investigate and requests for OTA to prepare reports on SDI gave the Senate the information it needed to review the program in detail. The decision to legislate a ban on testing ASATs against objects in space reflected the confidence gained by Congress and represented an extraordinary departure from previous congressional procedure. This precedent was followed by attempts to require executive adherence to the ABM Treaty.

How far Congress will go in asserting its role in foreign policy will depend on how successful presidents are in asserting leadership. If presidents can achieve arms-control agreements, they will certainly steal Congress's thunder. Even a president who does not favor arms control is in a position to rally national sentiment behind programs like SDI by arguing that the Soviets have prevented agreement, and that to inhibit any of these programs is, as Reagan put it in a State of the Union address, to enact Gorbachev's wishes into law.

Drawing The Political Lesson

If there is a single lesson to be drawn from the political debates over SDI, it is that although a popular president can gain support for a new venture in military technology by "going public" or by using the mass media to appeal directly to the electorate,^[73]    the representative system allows opponents of the policy to keep it under control by restraining expenditures, casting doubt on its technical prospects, and warning of

its potentially destabilizing effects on international relations. In view of the widespread alarm over the failures of technology in the Chernobyl and Challenger disasters, there is ample opportunity to play on public fears, counterbalancing efforts to play on the public's hopes. But even without trying to contest the president in the arena of public opinion, opponents of SDI have discovered that they can have considerable effect by aiming their efforts at Congress. The defense budget is the largest single part of the federal budget subject to congressional review, and although it is composed for the most part of fixed expenses, special programs like SDI are tempting targets for budget cutters.

Some of the efforts of the opposition would therefore have been effective despite public approval of SDI, but the weakening of Reagan's presidency, owing to difficulties in the conduct of other dimensions of foreign policy, made the rest of his program more vulnerable. The SDI probably suffered from the general loss of public confidence in Reagan's leadership following his much criticized "arms for hostages" deal with Iran. By refusing to compromise on SDI in the arms-control negotiations with the Soviets, moreover, the president triggered anxieties on that front as well, both domestically and among allies. These anxieties, coupled with criticisms generated by the decision of pro-SDI forces to push for early deployment, played into the hands of the opponents.

Political considerations will continue to influence SDI in the Bush administration. President George Bush will probably want to make less of an issue of SDI, but if the START negotiations falter, SDI's supporters will undoubtedly press him to give it renewed emphasis. The most likely prospect is that the new administration will be no more able than its predecessor to impose a reinterpretation of the treaty and an early-deployment decision over the opposition in Congress, and that SDI in general, because it has become so much identified with Reagan, will be much less of a political issue than it was during his term in office. Progress in arms-control negotiations with the Soviets would counteract pressures for early deployment. Even without such progress, however, concern over the economic implications of moving to development and deployment—the subject of the next chapter—could dampen enthusiasm, especially in view of mounting concern for reducing the U.S. budget deficit.