Exploring the High-Tech Frontier

By George J. Church.

Advocates conjure up visions of death rays flashing across thousands of miles of space to zap Soviet missiles as they rise. Critics counter with derisive pictures of the most supersophisticated Star Wars weaponry foiled by something as simple as grains of beach sand scattered in orbit. Back and forth go the millions of words of argument that have been resounding since Ronald Reagan unveiled his Star Wars plan in 1983. But the essential question raised by all the debate can be put into just three words: Can it work?

The answer depends heavily on what is meant by "work." If the meaning is Reagan's original vision of a defense that would banish the terror of nuclear war forever, by making the U.S. invulnerable to assault, then Star Wars almost certainly cannot work. Former Secretary of Defense James Schlesinger characterizes the headier versions of a Star Wars plan as "half Buck Rogers, half P.T. Barnum," and even the most ardent proponents generally con- cede that no technology now known or foreseeable could be guaranteed to destroy every warhead the Soviets could launch. Some percentage would always get through, causing death and devastation beyond the mind of man.

But supporters of the Strategic Defense Initiative (S.D.I.) increasingly argue that a defense need not be perfect or even near perfect to be worth building. Their vision is of a system that would wipe out a high enough proportion of attacking warheads to shift the odds dramatically against a Soviet first strike's succeeding. According to this view, deterrence would not be transcended, as Reagan dreams, but it would be vastly strengthened.

Can a Star Wars system work in this more limited sense? It will take a long time to find out: though a fairly crude defense could be erected by the early 1990s, some of the more advanced warhead-killing technologies, like lasers and particle beams, seem to be at least 15 to 20 years away from effective use. The obstacles are difficult even to conceive, let alone overcome. One example: tracking enemy missiles, aiming and firing at them, and then assessing almost instantaneously which ones have been hit would require a computer program so complex that it is beyond the ability of human beings to write it unaided. They would have to write instructions that special computer programs could translate into detailed Star Wars software.

To build an effective ballistic missile defense, the U.S. might have to repeal Murphy's Law ("If anything can go wrong, it will"). All parts of the S.D.I. would have to mesh smoothly without ever being tested under battle conditions. Worst of all, perhaps, critics can suggest a host of relatively simple countermeasures that might outfox the most sophisticated defense. Given all that, though, an imperfect but effective Star Wars defense just might be possible. Barely. Eventually.

Which does not necessarily mean that it ought to be built. Even if the S.D.I. proves technologically feasible, serious questions would arise as to whether it would really strengthen deterrence, and at what cost. Estimates of the money required start at $60 billion for a rudimentary system that would rely on interceptor rockets. Calculations of the cost of a fully developed laser- or particle-beam system run all the way from $100 billion to a staggering $1 trillion. Such a broad range means that all the figures are fairly wild guesses. Indeed, Cory Coll, leader of an S.D.I. research group at the University of California's Lawrence Livermore National Laboratory, predicts it will be five years before any halfway realistic estimate can be made of the cost of developing X-ray lasers as a warhead-killing weapon.

A go or no-go decision is still years away. Reagan sometimes talks as if he is determined to start on a defensive system no matter what, and some S.D.I. advocates hope to see component tests or even "demonstration shots" of prototype weapons during his second term. That is a dangerous idea, since it would violate a 1972 treaty with the Soviets and could get the U.S. irreversibly dedicated to a Star Wars defense of doubtful feasibility. Nonetheless, the Administration so far has not committed the U.S. officially to anything beyond a research program now expected to cost $26 billion in fiscal years 1985 through 1990. The requested appropriation for fiscal 1986, which starts Oct. 1: $3.7 billion.

Though the outcome of the research program cannot be predicted, the problems being faced and some of the ways in which they might be overcome are reasonably clear. Every proposal for a missile defense system begins with a profile of an enemy nuclear attack. In its roughly 30-minute flight from a silo in Siberia to detonation on top of a Minuteman silo in North Dakota--or above the White House--a Soviet warhead would go through four well-marked stages: 1) Boost. The rocket engines of, say, an SS-18 missile push it up through the atmosphere and into space. 2) Post-boost. On reaching the edge of space five minutes or so after launch, a device known as a bus detaches itself and maneuvers for another five minutes or so, releasing up to ten MIRV (for Multiple Independently Targetable Re-entry Vehicle) warheads at different targets in the U.S. The bus may also release up to 100 decoys, many of them aluminum-foil balloons. 3) Mid-course. The warheads and decoys speed through the emptiness of space for more than ten minutes. 4) Re-entry. The vehicles plunge back into the atmosphere and toward their targets.

The basic idea of Star Wars is a "layered" defense that would attack the warheads throughout their flight, destroying more at each stage. No more than, say, 70% might be wiped out during any one phase, but in the end perhaps only 10% or less of the warheads launched would explode on or over their targets. That would cause mass loss of life, but could leave most U.S. retaliatory capacity intact. Boost-phase interception, however, is generally considered to be the key to an even partially effective defense. That is when the enemy projectiles are easiest to find: the intense heat of a missile's rocket thrusters, concedes the anti-S.D.I. Union of Concerned Scientists, makes it stand out "like a firefly in a darkened room." That is also when a missile defense is most efficient: a single hit, by a laser beam, for instance, can destroy ten warheads at once. In post-boost and mid-course phases, the separated warheads are vastly more difficult to find and distinguish from decoys. On re-entry, the decoys burn up, and only the warheads continue to plunge through the atmosphere. But if there are, say, 5,000 left out of an original launch of 10,000, they could easily overwhelm any conceivable "terminal" defense. Besides, by then it might be too late to prevent terrible damage. Warheads can be set to detonate if struck by interceptor rockets or projectiles, and though hardened missile silos are almost impervious to anything except a direct hit, nuclear explosions even in the upper atmosphere can rain ghastly destruction on cities.

Weapons that could attack the warheads differ somewhat by stage of flight, but--and this is a faintly cheering thought to Star Wars researchers--most are $ adaptable to more than one phase. The systems that could zap missiles in boost generally could also hit warheads in post-boost, mid-course and perhaps even re-entry phases. A rundown of the potential missile and warhead killers that are getting the most attention from scientists:

LASERS. They are devices that generate high-powered, concentrated beams of light, almost perfectly parallel and of a single wavelength. The light from a lamp, in contrast, is a fuzzy discharge wiggling at different wavelengths and scattering in every direction. Laser beams travel at the speed of light (not surprising: they are light, though not always visible) and can be focused over thousands of miles of space to burn a hole in the skin of a Soviet missile, destroying its guidance mechanism and deactivating its warheads.

Chemical lasers, utilizing the reaction of gases such as hydrogen and fluorine, are the most powerful lasers now in use. But a missile-killing laser beam might have to be 10 million times as powerful as the one that the Air Force is now using in antisatellite weapons tests. Also, because its long wavelength somewhat spreads out its focus, a chemical laser beam might have to be held on precisely the same spot on a missile's skin for as long as seven seconds; during that time the missile might rise 20 miles. Because a ground- based laser could not send a beam around the curve of the earth, the generating apparatus would have to be carried aboard a fleet of satellites in low orbits. How many satellites would have to be sent aloft in order to keep some in range of Soviet missile-launching sites at all times is a subject of fierce debate among scientists; estimates have ranged from 100 to 1,600, which again means nobody really knows. In any case, the satellites would be monsters, each weighing 100 tons or more.

Excimer lasers, which use a different kind of chemical reaction, produce beams of short wavelengths that could destroy a missile by focusing on it for only a second or so. But the generating apparatus is so bulky that it could not be lifted into orbit; the laser stations would have to be placed on mountaintops to put them above the densest layers of the atmosphere. Even the thin upper layers would cause the beams to shimmer, however, owing to the same phenomenon that makes the light from stars appear to twinkle. The excimer laser beams would have to be bounced off mirrors in very high geosynchronous orbits over the equator, meaning that the mirrors would always hover over one spot on the earth's surface, to give the mountaintop stations a constant target to aim at. The geosynchronous mirrors would detwinkle the beam and reflect it to "battle" mirrors in low earth orbit. The battle mirrors would aim the laser beam at missiles or warheads. The mirrors would have to be gigantic, as much as 90 feet in diameter for the geosynchronous variety, and of almost unimaginable perfection; the slightest pitting or warping could cause a laser beam to scatter. Chemical lasers would need aiming mirrors (diameter: 30 ft.) atop their satellites too, and those mirrors would also have to be just about perfect. Star Wars Supporter Edward Teller considers fleets of laser satellites and orbiting mirrors too expensive to make chemical or excimer lasers practicable for missile defense.

That leaves X-ray lasers, which are being investigated under tight secrecy at the Livermore Laboratory. Grumbles Teller: "Methods that have real expectations of success are classified, and methods that have little possibility of success are advertised."

This much is known: a part of the enormous energy released by a nuclear explosion can be converted to powerful X rays by rods projecting from an atomic device in the microsecond or so before the rods themselves are vaporized. The beams are so powerful that they need no "dwell time" at all; they could knock out a missile or warhead instantaneously. Less precision is necessary in aiming them; an X-ray laser "beam" as wide as two football fields would have great destructive power.

The ideal weapon? Not quite. In theory, X-ray lasers could be based in space, but that might mean keeping something like 1,400 atomic bombs in low orbits constantly crisscrossing the Soviet Union. Says Coll, rather delicately: "I don't think it's going to be politically acceptable to put bombs in orbit." In practice, the X-ray lasers would have to be launched from earth at the first warning of attack in a "pop-up" defense (they are in fact the only laser devices compact enough for such a defense). To get high enough fast enough, they would probably have to be shot from submarines stationed just off Soviet coastlines.

Now the real problems. Flashing instructions from Washington to submarines in the first moments of nuclear war would be difficult, even assuming the submarines, held in fixed locations, had not been found and sunk by the Soviets in advance of a nuclear assault. If the subs survived and launched their laser-generating bombs, a greater difficulty would arise. All laser beams have trouble cutting through the atmosphere to destroy missiles at the start of their flight, but X-ray lasers are among the least penetrating. They could hit missiles only at the top of the boost phase, and probably would be best used for post-boost or mid-course interception. But that is when the warheads (no longer missiles) are hardest to find because they are hidden amid swarms of decoys.

PARTICLE BEAMS. They are streams of atoms or subatomic particles. In laboratories they can be accelerated to more than 99% of the speed of light by massive devices that can be two miles in length or four miles in circumference. A device that could accelerate the particles to perhaps half the speed of light, which would be poky by laser standards, but adequate for missile defense, might still weigh 500 tons, and hundreds if not thousands of the contraptions would have to be lifted into orbit. Particle beams have even more trouble penetrating the atmosphere than X rays, so they would be more useful for post-boost and mid-course interception than for boost-phase kills.

Even in space, particle beams that carried an electrical charge would be bent off course by the earth's magnetic field. To be effective as missile or warhead zappers, the beams would have to be made neutral, which involves a process of accelerating, aiming and focusing charged particles by electromagnets, then stripping off the charge just before the beams are shot out the end of an orbiting device. Why bother with them then? Primarily because the beams, which work by frying the innards of a missile or warhead with radiation, in principle are more lethal and yield a surer kill than lasers.

One type of charged-particle beam, the electron beam, can operate in the atmosphere, though currently only over very short ranges. Livermore Laboratory has been working on and off since 1958 to develop an electron beam for terminal-phase interception. The current idea is to station a sort of gun on the ground near a group of missile silos or a city and fire electron beams at incoming "physics packages" (a remarkably polite euphemism for atomic warheads) as they re-enter the atmosphere. The beams, however, are hard to aim and control. Not to mention the price tag: Researcher Bill Barletta figures this one small part of a defensive system might cost $20 billion. Says he: "If we can't do it for that price, we shouldn't bother proposing it."

KINETIC-ENERGY WEAPONS. They are simply objects like rockets, homing vehicles or even pellets fired at a missile, bus or warhead to destroy it by sheer impact. They are potentially effective at any stage from boost to re-entry, and can be fired either from the ground or from space. Their technology is well enough developed to make them available by the 1990s, much earlier than any of the beam weapons. Indeed, a terminal defense of sorts could be put into place right now. Main drawbacks: the range of kinetic-energy weapons is measured in hundreds rather than thousands of miles, and the top speed researchers are trying to reach for any projectile is about 25 miles per second. By laser or even particle-beam standards, that is slow, slow, slow.

Boost-phase interception would be carried out by low-orbit satellites firing rockets out of pods. A rocket would accelerate to the vicinity of a rising missile, then release a homing vehicle that would be guided by sensors and thrusters to a head-on collision with the missile. But as many as 20,000 rockets orbiting aboard many hundreds of satellites might be required to keep enough within range of Soviet launch sites at all times to fend off a full- scale missile onslaught.

For mid-course interception, batteries of ground-based rockets might be fired into the upper atmosphere. Each rocket would release a swarm of so- called smart rocks--vehicles powered by little thrusters and guided by tiny sensors--to hit warheads and decoys in space. An alternative is to fire the smart rocks out of devices called rail guns placed in orbit. The rail guns use a burst of electric current to accelerate the smart rocks along a rail. One problem is sheer numbers: immense swarms of smart rocks would be needed to hit warheads and decoys indiscriminately. The other option, picking out the warheads from the decoys, would require rocks that were not just smart but intellectual giants. They would have to be guided by tiny sensors sophisticated beyond any yet developed.

Terminal defense is easiest, technologically. Warheads, heated and slowed by friction with the atmosphere on re-entry to speeds of about two miles per second, could be tracked by airborne or even ground-based radar. They could be hit by interceptor rockets or pellets discharged by fragmentation bombs. But enough missiles would have to be destroyed in boost, and enough warheads in post-boost phase or mid-course, to keep the terminal defense from being overwhelmed. And then there is the problem of hitting the warheads high enough to minimize the effect of blast, fire and radiation on the ground.

The different technologies eventually would probably be used in combination. But, of course, the systems will develop at different speeds, assuming all or any prove feasible. Air Force Lieut. General James Abrahamson, who heads the Pentagon Strategic Defense Initiative Office, foresees a three-stage development: initial deployment of a "robust" defense, presumably relying mainly on kinetic-energy devices, to be started only if and when R. and D. indicates that a second and then a third generation of more sophisticated weapons will follow in not too many years.

Meanwhile, the Pentagon plans to spend nearly as much of its 1986 S.D.I. budget request on sensors and battle-management systems ($1.6 billion) as on weapons development ($1.8 billion). Sensors are crucial to any system: they have to find the targets first and aim at them over thousands of miles of space. Keith Taggart, a researcher at the Los Alamos National Laboratory, likens the job to shooting out a specific window in New York City's World Trade Center by firing a rifle bullet from the top of the John Hancock Building in Chicago. The sensors also would have to flash back instantaneous assessments of what targets had been hit, so that a battle station would not waste vital seconds aiming a laser or particle beam at a missile or warhead already destroyed.

The computer requirements are straight out of science fiction. Computers would have to keep track of tens of thousands of objects (warheads, decoys, smart rocks) moving at high speeds, analyze instantly billions of bits of information from sensors and weapons platforms, determine which weapons to fire, when to fire them and at what targets. Not only could no human write such a program unassisted, no human could check it for errors. That would have to be done by computer too. James Fletcher, who headed the Administration's original S.D.I. study, estimates the program might have to be put through 50 million debugging runs before it would be battle-ready.

Other requirements, according to Fletcher: "The computers must be able to operate in a nuclear environment and must be hardened to survive radiation and shock. To keep crucial command, control and communications capabilities out of the fray, some of the computers would be placed in high orbit halfway to the moon." Humans would make the key strategic decisions in advance, determining under what conditions the missile defense would start firing, and devise a computer system that could translate those decisions into a program. In the end the defensive response would be out of human hands: it would be activated by computer before U.S. commanders even knew that a battle had begun. Fletcher insists that the "hardware requirements" are almost "within the state of the art" now, though 20 years or more might be needed to develop and deploy the battle program.

The real question is what the Soviets would be doing during that time. And the answer is all too obvious: taking countermeasures. Many are available, and they generally require less exotic technology than the defense they would be designed to defeat.

A simple though expensive method would be to multiply the size of the attacking force in the hope of overwhelming the defense. Some experts think the U.S.S.R. could double or even triple the number of warheads it could launch before the U.S could get a defense in place. A defense could be bypassed by bombers and cruise missiles, which, because they fly relatively low, could not be zapped from the heavens. After the Soviets shifted their main emphasis to ballistic missiles, the U.S. let its once extensive air- defense system deteriorate. Today it could not fight off a bomber attack, or even detect a cruise-missile assault. (It has trouble enough with drug- running airplanes.) A Star Wars system would have to be accompanied by a strengthening of air defenses that James Schlesinger estimates would add $50 billion a year to whatever might be spent on Star Wars.

Missile defenses could also be foiled. Boost times could be shortened, perhaps to as little as 50 sec., by equipping attack missiles with more powerful rocket thrusters and toughening their skins so that they could withstand a faster trip through the atmosphere. Missiles could also be made to spin like rifle bullets, so that laser or particle beams could not dwell on one spot, and be given reflective coatings to deflect or diffuse the beams. To be sure, the Soviets would pay a price: such measures would reduce the numbers of warheads and decoys that a missile could carry, and that would make post- boost or mid-course interception somewhat easier for the U.S. But there are clever ways to get around that too. Smart rocks, for example, might be fooled by equally smart decoys that sent out signals like those of warheads.

Finally, the Soviets could attack a Star Wars system directly. Orbiting satellites are vastly easier than missiles or warheads to track and draw a bead on. Just two possibilities: the Soviets could orbit a "space mine" that would blow up near an American satellite and destroy it, or a countersatellite that would discharge a cloud of pellets, capable at orbital speeds of piercing steel, or even beach sand, which could pit and disable laser mirrors. American satellites might be defended against such attacks. But once that kind of cycle begins, says William Shuler, coordinator of S.D.I. research at Livermore, "we are going to be in the counter-countermea sure game forever."

The final decision to build or not to build a Star Wars system cannot be guided by technological prospects alone. The overriding issue is whether a functioning missile defense would enhance or upset nuclear stability. But the research program at least ought to determine what is and is not feasible. Three possible conclusions can be foreseen.

One is that the U.S. could indeed develop a system that would be highly effective against both missile attack and countermeasures at something resembling a reasonable cost. That seems the least likely outcome. No one should say flatly that it is impossible, however. Anyone who does is speedily reminded by S.D.I. advocates that eminent scientists once doubted the feasibility of building nuclear bombs and ballistic missiles and of undertaking flights to the moon.

Research must continue, but it should be conducted by open minds willing to accept a second possible outcome: a conclusion that Star Wars just will not work, and that the large sums likely to be spent on it in the next five years or so might have to be written off. S.D.I. must not become one of these projects to which Government leaders develop such a strategic, political and emotional commitment that they keep pouring money into it regardless of what the research shows. That mind-set could be disastrous. It might, for example, lead to premature tests that could provoke the already worried Soviets into an accelerated buildup of offensive and defensive arms--even if the tests failed.

The third conceivable outcome: the U.S. might conclude that it could, at enormous expense, build an imperfect but effective Star Wars system that the Soviets could counter only by extraordinary, costly and dangerous countermeasures. That would probably bring both nations, and the world, to a truly fateful decision: an arms race surpassing anything thus far imagined, or an arms-control deal reducing all categories of weapons. There would be precious little in between.

With reporting by Jay Branegan and Bruce van Voorst/Washington and Dick Thompson/San Francisco, with other bureaus