The Atlantic | April 1988 | The Rush to Deploy SDI

More on politics from The Atlantic Monthly.

The Rush to Deploy SDI - Page 3
1 | 2 | 3

his leaves the midcourse and terminal phases as potential times to stop Soviet ICBMs. In these phases the United States will have to shoot down Soviet nuclear warheads after they have separated from the rest of the ICBM. Yet these warheads are only a few feet long, travel four miles per second, have no easily identifiable fiery plume, and are protected by a heat shield capable of withstanding re-entry into the earth's atmosphere. They are not easy targets.

In addition, the Soviets could deploy hundreds of thousands of decoys along with the tens of thousands of warheads. The decoys would probably consist of cheap aluminized Mylar balloons. The warheads themselves could be encased in the same balloons, so that warheads and decoys would look exactly the same. The decoys can be carried in much larger numbers on ICBMs, because they are much lighter than actual nuclear warheads. In the vacuum of outer space, heavy warheads and light decoys would travel at the same speed.

If the United States is to have a chance of destroying Soviet warheads as they travel through outer space, the real warheads must be distinguishable from the decoys—there is neither time nor weaponry enough to shoot them all down. Yet recent studies by Pentagon scientists have yielded very pessimistic results. Radar cannot distinguish warheads from even the simplest decoys, and, as General Abrahamson testified to Congress, "passive discrimination [infrared or optical sensing] is not quite sufficient" to pick out the warheads from a cloud of decoys.

Following this major setback, the SDI Organization pursued a new concept—"interactive discrimination," using neutral-particle beams. This is why the SDI Organization gave McDonnell- Douglas the $481 million contract to build a prototype beam. Unfortunately, even this new concept is unlikely to be very successful in quickly separating the warheads from a cloud of decoys, according to testimony by Robert Cooper and a recent study from the Massachusetts Institute of Technology. (And the SDI Organization recently canceled the contract.)

With the prospects for boost-phase and midcourse-phase intercept only marginal—using existing or near-term technology—we are left with the prospect of stopping more than 30,000 warheads when they are above the United States, only minutes from their targets. However, the Soviets may have taken steps—as we have—to thwart terminal defenses.

In March of 1987 Lawrence Woodruff, the deputy undersecretary of defense for strategic and theater nuclear forces, described the contest between the offense and the defense to the House Armed Services Committee this way:

The Soviets have been developing their Moscow [ABM] defenses for over ten years at a cost of billions of dollars. For much less expense we believe we can still penetrate these defenses with a small number of Minuteman missiles equipped with highly effective chaff and decoys. And if the Soviets should deploy more advanced or proliferated defenses, we have new penetration aids as counters under development.... We are developing a new maneuvering re-entry vehicle that could evade interceptor missiles....

Woodruff is correct, and the implications of his statement for the President's astrodome vision are, to say the least, sobering: if the Soviets can't even defend one city, think how hard it would be to defend a nation.

The most effective way for the United States to guard against a sudden treaty-violating Soviet deployment of anti-ballistic-missile systems would be to increase our research efforts on decoys and improved penetrating warheads. These can defeat any known or anticipated Soviet land-based ABM system. Change the word "Soviet" in that sentence to "American," and you begin to see how relatively easily the offense can overwhelm the defense in nuclear warfare.

As if the hardware problems were not serious enough, most experts believe that the computer software to manage the entire battle would be even more difficult to develop. (See Jonathan Jacky, "The 'Star Wars' Defense Won't Compute," June, 1985, Atlantic.) It would be much more complicated than the software that runs the space shuttle, and yet it could never be tested in a realistic nuclear-attack scenario; it would have to work the first time it was used. A recent report by the General Accounting Office has severely criticized the SDI Organization's "inadequate direction and planning" of the crucial battle-management program. After a year-long study, the GAO said, "The problems, if not corrected, could reduce [the SDI Organization's] ability to provide the needed information for an informed systems development decision planned for the early 1990s. " Yet one of the directives that Weinberger dashed off before leaving the Department of Defense actually calls for reduced oversight of the SDI Organization by the Secretary of Defense.

The bottom line is not very encouraging. Louis Marquet, SDI's former deputy director for technology, admits that a near-term deployment of interceptors is "not going to have a lot of capability against a determined first strike.... ultimately, given a dedicated Soviet effort at countermeasures, it will run out of steam." The answer, he says, would be to improve the system by adding more and better orbiting battle stations and other weapons at additional expense.

Finally, Phase I or any later phase of SDI will have little potential for stopping any nuclear weapons other than ICBMs, with their high ballistic trajectory. As a nation that has very limited air-defense capabilities, we should recognize the relative ease with which our defenses can be penetrated by low-trajectory submarine-launched ballistic missiles, Soviet bombers, and air-launched and sea-launched cruise missiles. According to the Library of Congress, by the end of the century the Soviet Union could triple its inventory of submarine-launched-ballistic-missile warheads from 3,400 to 13,000 and vastly increase its bomber-warhead inventory, from 800 to 7,500. The Soviets are already reported to be building a new type of submarine-launched cruise missile, the SS-NX-24, which could defeat SDI by penetrating U.S. airspace at a low altitude.

The conclusion is inescapable: There is no convincing evidence that the result of premature deployment would likely be anything other than a defense-offense-defense cycle weighted in the offense's favor. Because of technical limitations inherent to the proposed SDS, the result of deploying it would likely be to send the arms race spiraling in time and space.

Dr. Robert Sproull, a member of the SDI Advisory Committee, warned Congress that the SDI Organization "tries to keep the discussion on technical details." He said, "They go straight from the President's vision to specifics of engineering. They avoid the questions of survivability and cost."

There is some evidence for this charge. The Defense Authorization Acts for fiscal years 1986 and 1987 required the Defense Department to submit estimates of the cost for research, development, and procurement of anticipated SDI systems to Congress along with the 1989 budget request. In March of last year General Abrahamson explained, "We did not submit a report with the FY 1988-89 budget request because the information necessary to prepare an accurate, comprehensive report is simply not available."

Yet the previous January, SDI's deputy director for technology, Louis Marquet, had said that it would cost "on the order of $100 billion" merely to reach the starting line for production of an early-deployment system. In June, General Abrahamson submitted detailed cost estimates to the Defense Acquisition Board. And several independent cost estimates have been prepared, including a study by the Senate which suggests that it could cost as much as $1 trillion simply to launch an SDI system into space, not including the cost of the satellites themselves. If scientists and senators can calculate possible costs with considerably fewer resources than the SDI office has, and top Department of Defense officials can come before Congress and testify with dead certainty that comprehensive strategic defenses are feasible and can be deployed within a given time frame, then these officials should be forthcoming with detailed estimates of how much the deployment will cost the taxpayer.

Crash Program

n light of the substantial technical, military, and economic problems confronting SDI, to say nothing of the nearly universal skepticism from the nation's scientific community, it may be asked, How can President Reagan support so dubious a program? Further, how can it have advanced so far, consuming $13 billion over the past five years, and now be rushing toward deployment?

The President is a man of tremendous communication abilities, and his efforts are aided by the simplicity of his goal and the exquisite perfection he anticipates. His closest advisers on SDI have been figures like Edward Teller. A famous scientist but a man who has obviously spoken inaccurately about the problems confronting SDI, Teller is highly motivated by his desire to assure the realization of his dream. In addition to the personal, political, and ideological forces driving SDI, there has sprung up a group of defense contractors whose officers and stockholders could greatly benefit from the majestic proportions of the envisioned project. There is also a bizarre combination of idealists seeking an easy way to rid the world of war and hawks who never saw a weapon they didn't like or a war they wouldn't want to fight. Finally, there is the growing SDI bureaucracy, which now consists of 225 full-time and 300 part-time employees, in addition to thousands of scientists and engineers working on more than 3,000 SDI contracts across the country. Together these people and these forces have advanced a juggernaut of such technical complexity and with such strong presidential support as to be capable of intimidating any sensible majority. It is time to step back from the brink and restore reason to this program.

There is no pressing national-security interest that compels a crash program to deploy SDI. The Scowcroft Commission officially closed the so-called window of vulnerability in 1983, when it determined that the chance of a successful Soviet first strike was very remote. If the purpose of SDI is now to "strengthen deterrence by denying the Soviets any confidence in a successful first strike," as Weinberger said last August, we are already spending over $35 billion annually on other strategic-forces programs to do just that.

The two thirds of our ballistic-missile submarine fleet normally kept at sea, the one third of our bomber force normally kept on strip-alert, and the tremendous uncertainties involved in attacking our ICBM forces supply credible deterrence that will be sufficient for the rest of the century. Improvements that are under way on all legs of the U.S. strategic triad and on the U.S. command-and-control system should increase our confidence: new motors, warheads, and penetration aids for the Minuteman; fifty MX missiles; possibly Midgetman mobile missiles; Ohio-class nuclear submarines; Trident II sea-launched ballistic missiles; ground-, sea-, and air-launched cruise missiles, advanced "stealth" cruise missiles; and the deployment of 100 B-1 bombers, with Stealth bombers to follow. The U.S. troops and their families stationed in Europe serve as a credible link between the United States and the defense of NATO. There is no firm evidence that a limited ballistic-missile defense to "complicate Soviet attack plans" is anything other than unnecessary.

Further, recent experience demonstrates that there are grave dangers in crash programs. The House Armed Services Committee last year investigated the B-1 bomber and the MX missile—two key strategic programs that were pushed through by this Administration without adequate thought or testing.

As it turns out, we will have 100 B-1s that cannot penetrate Soviet airspace because, among other defects, when their advanced electronic systems are switched on, the defensive systems interfere with the offensive systems, and both act as a giant beacon guiding enemy missiles to the planes. It will cost at least $6 billion to fix the planes, over and above the $28 billion already spent. Meanwhile, the MX is still not survivable and, worse, its high-technology guidance units do not work as promised. At least twelve of the thirty deployed MXs cannot be fired for lack of these units, and the rest can't be counted on to hit their targets, because of shoddy workmanship. A Northrop Corporation engineer who worked on the guidance equipment testified that it is "most assuredly defective." As a result, he said, "these missiles stand as much chance of hitting their targets in the Soviet Union as landing right here in Washington." Basic tests are still being conducted on the B-1 and the MX, years after the first weapons were delivered.

These programs were pushed through Congress by a skillful distribution of defense contracts. Rockwell Corporation, for example, farmed out work on the B-1 to some 5,200 subcontractors, until there was a contract in almost every congressional district. The same was true for the MX. If SDI can reach that level of spending, it may be impossible to stop. We do not need another crash program fueled by contractors' hunger for profits and bureaucrats' concern about their careers. In January of last year Allan Mense, then the SDI program's acting chief scientist, said, "If we don't come up with something specific, people are not going to let us play in the sandbox for ten years." But there is strong support in Congress and among the American people for a sound basic-research program.

What should we do? First, we should commit to a careful, long-term research program on ballistic-missile defense—not the program as it is now structured but one that would allow us vigorously to explore advanced technologies to see if someday we can achieve the President's vision of an impenetrable shield, as unlikely as that may be. Since the Soviets have an SDI Program themselves, the better part of wisdom is for us to maintain our lead in strategic-defense technologies and avoid the possibility of their developing something really effective while we are left lagging behind. Expert testimony to Congress, including that of Sproull, Cooper, and the physicist Peter Zimmerman, indicates that a robust research program can be carried out for about $3 billion a year. (The Administration wants to spend more than $5 billion in 1989.)

Second, we should explore the possibility of deploying a defense similar to the one the Soviet Union now has around Moscow, which might protect us against an accidental ICBM launch. The ERIS program, which is part of SDS, could possibly serve this function. Lockheed Corporation is lobbying hard to accelerate the ERIS schedule for a 1990s deployment regardless of the fate of the rest of SDI. Lockheed estimates that the total cost would be about $5 billion for a system of 100 ground-launched interceptors and radars, which might be in compliance with the existing ABM treaty. If feasible, ERIS would be an expensive insurance policy, but we should investigate this option. It may be a reasonable hedge against an accidental Soviet launch of a single ICBM headed for New York or Washington, and it could give us hands-on experience with ballistic-missile defense. The widely mentioned notion that such a system could defend the United States against terrorist attack seems dubious, given how unlikely it is that terrorists would use a ballistic missile; they are much more likely to deliver their bombs in a suitcase or on a fishing trawler. These are hardly the only conceivable alternative delivery systems. A witness before the Senate Armed Services Committee recently noted that "the Soviets could just put nuclear weapons inside bales of marijuana, since they know we can't prevent that from entering the country. "

Third, we should use the SDI program as leverage to negotiate a treaty with the Soviets under which both countries would eschew space weapons and further agree to substantial and fair reductions in offensive weapons. This may prove to be the most valuable contribution of the SDI program. Such a treaty would be mutually beneficial, verifiable, and enforceable. The benefits are obvious, and we have the means to verify an agreement through existing spy satellites and the new willingness of the Soviets to allow on-site inspections. We should test the Soviets on their stated offer to arrange inspections of all satellites before launch. Such inspections could ensure that space weapons were not placed in orbit surreptitiously. We could enforce such a treaty by including a provision that allowed the inspection of satellites suspected of violating the ban. The inspection could perhaps be accomplished with the space shuttle.

Independent experts, retired and active-duty military spokesmen, members of Congress, and congressional staff have raised questions indicating that serious technical, legal, and military issues pertaining to SDI remain unresolved. As the Army Chief of Staff, General John A. Wickham, Jr., told Congress early last year, "We are all in agreement that we don't know enough to make a deployment decision."

Defending our population from incoming missiles is an admirable goal. But in trying to achieve that goal we must not let the political-industrial complex produce weapons that not only are ineffective but also increase the chance for ever more lethal offensive escalations.

1 | 2 | 3

What do you think? Discuss this article in Post & Riposte.