YOU ARE HERE: LAT HomeCollections


The GOP's Love Affair With Ballistic Missile Defense Chimera

July 08, 1996|GARY CHAPMAN

Bob Dole is clearly having a frustrating time distinguishing himself from President Clinton, who displays an extraordinary knack for absorbing Republican ideas. But on at least one issue they are at real odds: the need for a national defense against ballistic missiles. Dole has been pushing this idea in Southern California lately, hoping to appeal to nervous aerospace workers.

The Republican fetish for strategic defense, once popularly known as "Star Wars," is difficult to fathom. The U.S. has spent a staggering $100 billion on ballistic missile defense research so far, without noticeable progress. But such defenses are still at the heart of the Republicans' plans for the military budget in a Dole presidency, despite widespread scientific and technical skepticism that ballistic missile defense is either possible or necessary.

What this may reveal, ominously, is Dole's willful ignorance about the limits of technology, a grave blind spot in someone who could be leading us into the next century.

The Strategic Defense Initiative was announced by President Ronald Reagan on March 23, 1983--a surprise segment in a televised speech about Grenada. Reagan portrayed the program as a "peace shield" in space, and he blithely said that the U.S. would share the system with the Soviet Union in order to facilitate the elimination of nuclear weapons, a proposal never endorsed or even taken seriously by the Pentagon.

Despite Reagan's hope that his unexpected announcement would lead to quick deployment of a complicated space-based defense, the whole plan was in fact submitted to several panels of experts for review. One panel was charged with looking at the computer requirements for a system that would necessarily be controlled remotely and at speeds beyond human capacities for reaction.

A member of this panel was David L. Parnas, a world-renowned software engineer and a professor of computer science in Canada (although Parnas was a U.S. citizen and a longtime Pentagon consultant). Parnas spent two days listening to Air Force briefings, then in June 1985 he resigned from the advisory panel, concluding that the fundamental computer requirements for strategic defense could never be satisfied.

Parnas eventually published his objections, and they became the basis for several years of debate in the computer science field about not only "Star Wars" but also about computer science's fiscal dependence on increasingly dubious military projects.


His basic points are simple and unalterable: By its nature, strategic ballistic missile defense cannot be tested in its conditions of use--we can't fire a missile at Los Angeles to see if our defense works. And no computer system of even modest complexity has ever been considered reliable without extensive testing in actual conditions of use. Indeed, these days customers are testers--we use software with errors, or "bugs," and then report those errors to the developing company, hoping that the bugs will be fixed in subsequent software releases.

National ballistic missile defense, of course, would require computer software of both unimaginable complexity and infallible trustworthiness--and it would have to work correctly the first time it was ever used.

While this isn't mathematical "proof" that such software wouldn't work, the probability that it would work correctly is close to zero; the goal of a trustworthy, reliable software-dependent defense against intercontinental ballistic missiles will always be a fantasy.

"Star Wars" thus displayed little or nothing to recommend it even during the height of the Cold War.


Now that the Cold War is over and the Soviet Union is history, strategic defense advocates have to scour the world to find threats that justify tens of billions of dollars in new defense spending. North Korea, Iraq and Libya are the "rogue states" most frequently mentioned as ballistic missile threats of the future.

Proponents of missile defense argue, moreover, that it will be far easier technically to deal with these types of threats then it would be to shoot down thousands of Soviet missiles and decoys, as the "Star Wars" plan unconvincingly promised.

But a similar ground-based missile defense pursued in the 1960s by the Johnson administration ran into public opposition because few Americans wanted to live in cities ringed with missile launchers. Nor are people likely to be comfortable living amid missiles linked to an early-warning system that has had, over the years, numerous computer-related and nerve-rattling false alarms.

Furthermore, the long lead time and elaborate facilities required to build an intercontinental missile mean that the U.S. and its allies would be able to deal with such a threat from a rogue state in others ways--via a preemptive strike, for example.

Los Angeles Times Articles