What can we do? When can we do it?

WHAT CAN WE DO? WHEN CAN WE DO IT?

WHAP, ZAP, HISS, BUZZ? The sounds of the video arcade. The sounds of Star Wars. And to most Americans, unfortunately, the image of the Strategic Defense Initiative (SDI). As long as constellations of "death stars" hovering over the planet remain our view of a missile defense, effective protection for the United States is further away than a cure for the common cold. The technical truth of missile defense, however, is very different. We are much closer to a defense that works than most of us realize, or than we had dared hope in 1983. Effective strategic defenses in the 1990s can use 1980s military technologies--precision guided missiles, sophisticated micro-computers, and heat-seeking sensors.

It is impossible for a professional, let alone a layman, to keep all of the types of ballistic-missile-defense systems straight. It is no wonder that some members of Congress wouldn't buy a "late-midcourse exoatmospheric kinetic-energy interceptor system."

The missile-defense terminology of the 1960s adds to this confusion. At that time, we categorized defensive systems by the portion of an ICBM's flight the defense operated on. ICBMs of the 1960s had three distinct phases. During the first--or "boost"--phase, the ICBM's rocket engines burned for a few minutes to thrust the warhead up into space and on its way toward its target. Once in space, the rocket engines fell away and the warhead proceeded on a "ballistic" arc hundreds of miles up and thousands of miles long. This "midcourse" phase lasted about thirty minutes. In the finall minute of flight the warhead or Re-entry Vehicle (RV) re-entered the atmosphere over the target in what is called the "terminal" phase.

Offensive nuclear missiles have evolved since the 1960s, rendering this defensive terminology obsolete. Partly as a way of overwhelming the old-style defenses, both the United States and the Soviet Union developed and deployed Multiple Independently Targetable Re-entry Vehicles (MIRVs). On top of a MIRVed ICBM sits a device carrying three or more warheads. This device is called a "post-boost vehicle," or sometimes just a "bus." The bus has a small rocket engine, which fires repeatedly once it gets into space to send each warhead toward a different target. The monster Soviet SS-18 can carry ten or more warheads, each capable of reaching a different target.

Another possible development would be a "depressed-trajectory" missile. If the missile and warhead can stay low enough, there is a better chance of slipping under the defender's radar coverage and surprising his defense. To do this the ICBM or submarine-launched ballistic missile is fired at a lower angle so that it stays inside the atmosphere longer and spends less time in space, where the radar can see it.

A third approach is to give the warhead the ability to maneuver after it has entered the atmosphere in the terminal phase. As anyone who has played Missile Command in a video arcade knows, a Maneuvering RV (MaRV) can outfox some kinds of defensive systems.

For these reasons I prefer to separate defensive systems into three types, not according to the phases of the offensive missile's flight, but according to how the defensive system works and where it is based. Three distinct types of defensive systems are: point, sovereign area, and global. These systems may overlap in coverage, and may operate during more than one phase of an attacking missile's flight.

To understand how each type of missile defense works, consider U.S. naval defenses in the nineteenth century. We had point defenses based onshore near our vital harbors in the form of shore batteries. These point defenses defended only a single target each and were based near that target. We had a sovereign-area defense in our coast guard, consisting of short-range ships, capable of defending a large part, if not all, of one seaboard. These sovereign-area defenses were based in our own territory, but could defend many targets. Finally, our blue-water navy provided us a global defense, able to defend the United States or its allies at any place on the globe from an attack originating at any other place, provided we had enough notice to concentrate our ships for warding off the attack.

In missile defense, point defenses generally work only in the terminal phase. Sovereign-area missile defenses can also operate in the terminal phase, but they can, in addition, work during much of the midcourse phase, depending on how soon the sensor system picks up and tracks attacking warheads. Global missile defenses have some, if not all, of their elements based in space. From space they can defend against missiles both in their boost phase and in midcourse.

The actual hardware in a missile-defense system has three parts: sensors, weapons, and battle management. The sensors are the eyes of the system, detecting, identifying, and tracking the missile or warhead targets. The battle-management part is the brain, using computers to decide which weapon to assign to which target and to keep abreast of the overall engagement. And, of crouse, the weapons do the job of stopping the attacking missiles and warheads.

During the 1960s the United States and the Soviet Union both developed missile-defense systems, which could have been deployed in the early 1970s. But in 1972, the ABM Treaty was signed, limiting each side to one hundred missile interceptors. The United States chose to deploy its allowed missiles in the form of the Safeguard system, based near Grand Forks, North Dakota. Safeguard included two different missile-defense systems: Sprint and Spartan. Sprint was a short-range point defense. It used a small, fast missile as a last-ditch defense, just before the attacking warhead would hit its target. Spartan used a large missile that would fly deep into space to intercept attacking warheads hundreds of miles from their targets. Both missiles carried a nuclear warhead and both relied on radars and sophisticated battle-management computers.

We had barely completed the Grand Forks Safeguard system in the mid 1970s when we decided to dismantle it. Today we cannot stop even one missile from falling on the United States. However, the Soviet Union went ahead with its Galosh system around Moscow in the 1970s, and is now in the process of upgrading it.

It is also interesting to note that our Soviet friends are deploying a nationwide "air" defense weapon, the SA-12, which has the capability of performing tactical-ballistic-missile intercepts. Many experts believe that "tactical" could extend all the way up to our submarine-launched and some land-based strategic missiles. One way to make this system effective against strategic missiles is to provide it data from long-range early-warning radars. The framers of the ABM Treaty knew this and placed limitations on where and how such long-range radars could be constructed. The Soviet Union violated this provision by constructing a big, long-range radar at Krasnoyarsk, in Siberia. The Soviets have thus used the Krasnoyarsk and other treaty violations to give themselves what many believe to be a formidable point-defense system.

Thus, in assessing the feasibility of strategic defense, we must consider achievements, prospects, and problems in developing sensors, battle-management systems, and weapons for point, sovereign-area, and global defenses. Since President Reagan launched SDI in 1983 we have made impressive progress. We can now discuss, with some accuracy, what types of defense we could have in the 1990s and how good they might be. Costs are a little trickier, but I can make some general ball-park estimates. These estimates are based on today's military systems of similar complexity and technical sophistication.

I. Point Defenses

SENSORS FOR point defense exist now. The radars we built for the Sprint anti-ballistic-missile syste in the 1970s were more than adequate for point defense. Perhaps more significantly, the radars used in our air-defense systems can also detect ballistic-missile warheads. Technology similar to that used in these air-defense radars could do the job for ballistic-missile point defenses. But they would need to be tipped off that a warhead was on its way, so they would know what to look for. The tip-off can come from big early-warning radars--which is why the Soviets' Krasnoyarsk radar violates the ABM Treaty.

We built a good battle-management system for the Sprint. Of crouse, we could do a lot better today with our new, fast mini-computers. In fact, point-defense battle-management systems could be programmed on an Apple computer.

We have made major advances in point-defense weapons since the 1960s and 1970s. The old interceptor missiles used a nuclear warhead because we couldn't maneuver them close enough to their targets for other, non-nuclear means to work. Nuclear bombs are big enough fly-swatters that you don't have to worry too much if the fly isn't exactly where he's supposed to be. Today's sophisticated on-board guidance sensors and control make it possible to get close enough to kill the target without using a nuclear weapon. Once the battle-management system and ground radars guide the interceptor missile close to its target, its on-board sensors pick up the target, and it homes in for the kill--destroying the target either by crashing into it or with a conventional explosion. SDI reached a significant milestone on June 27, 1986, when the Army's Flexible Lightweight Agile Guided Experiment (FLAGE) missile hit a simulated warhead using its built-in radar to home in and hit the target. When the FLAGE program finishes next year, we will have proven out the guidance technology for an effective non-nuclear point defense.

The most important use of point defense would be to protect our own ICBM bases. Protecting our missiles enhances their deterrent value, making a Soviet attack much less likely. However, point-defense systems might also be very useful for defending critical NATO military facilities in Europe. There are about a thousand such facilities. Not surprisingly, the Soviets have several thousand tactical-ballistic-missile warheads aimed against these. A new anti-tactical defense system would be comparable to an anti-aircraft defense system in size, complexity, and cost. Each anti-aircraft missile costs about a quarter of a million dollars. Thus, we might be able to provide point defenses for the NATO facilities for a few billion dollars. We are already developing the ability to improve existing anti-aircraft missiles such as the Patriot so it could be used to stop tactical-ballistic-missile warheads. In addition, a non-nuclear system based on the FLAGE technology might be available and in service in the early 1990s.

II. Sovereign-Area Defenses

THE 1970S SPARTAN SYSTEM was a sovereign-area system relying on the large phased-array-radar sensor at Grand Forks, North Dakota. This radar is still operating as part of the U.S. early-warning system. As the radar stands, it could support a limited area defense. Other parts of our early-warning system are modern space- and ground-based early-warning sensors. Thesensors could help a big, central missile-defense radar do its job better.

In the longer run it appears that sensors carried on airplanes could do the sovereign-area-defense job better. The airplane sensors would detect and track missile warheads just as today's AWACS planes direct an air-defense battle. The AWACS planes carry radar, but missile-defense planes would carry infrared sensors that would detect the heat of the warheads. Infrared "television" would pick out the incoming warheads against the cold background of deep space, and small lasers would measure the distance to and speed of the targets. These Airborne Optical Systems (AOS) would send this information to the interceptor missiles, guiding them toward their individual targets. SDI will test some of these technologies in 1988 in the Army's Airborne Optical Adjunct experiment. The results could let us build an AOS five to seven years from now.

The Spartan system, along with the point-defense Sprint, had a battle-management system consisting of several million lines of computer code or software. Some of the code can be resurrected, and all of it could be quickly rewritten. The Army operates a computer center in Huntsville, Alabama, where new battle-management methods have been constantly developed and upgraded since the 1970s.

On June 10, 1984, a non-nuclear U.S. Army interceptor missile, the Homing Overlay Experiment (HOE), lifted off from the Kwajalein missile test range in the Pacific and sped out into space. A few minutes later, approximately one hundred miles up and several hundred miles down range, the interceptor opened its infrared eye and saw the heat of a mock warhead that had been fired several tens of minutes before from Vandenberg Air Force Base in California--thousands of miles away. On-board computers directed the interceptor to fire its rocket and maneuver into the path of the oncoming warhead, where it made a direct hit, pulverizing both itself and the warhead. SDI has a follow-up program called Exoatmospheric Re-entry Vehicle Interceptor Subsystem (ERIS). An early version of ERIS could be ready for deployment in the early 1990s. With the right kind of sensors a single ERIS site could provide a limited defense of the entire United States or Western Europe.

Another area-defense interceptor, the High Endoatmospheric Defense Interceptor (HEDI), would work inside the atmosphere while ERIS worked outside. HEDI could also be ready for deployment in the early 1990s. HEDI can catch warheads that slip through ERIS or fly under it--for example, a depressed-trajectory warhead. HEDI cannot defend as big an area from a single site as ERIS can, but a single HEDI site could defend all of New York State, for instance.

The contractors working on ERIS say that a preliminary version could be installed in conjunction with the existing radars at the old North Dakota Safeguard site. The ABM Treaty allows such a deployment, with up to one hundred missiles. The contractors believe that this deployment would cost several billion dollars and could be done in five years. It is true that a limited deployment of this type might not cover the entire United States, and it would do little against a massive attack, but it could be useful against an accidental launch of several missiles.

On a slightly longer time scale--about seven to ten years--an advanced ERIS could be ready for deployment. The advanced ERIS missiles are estimated by some contractors to weigh a few thousand pounds and cost about $1 million each. These missiles would be small enough to be carried on mobile launchers--so the Soviets could not target them with their long-range missiles. An ERIS-based defense system, supported by Airborne Optical System sensors, could be deployed as part of NATO or U.S. defenses. With several thousand interceptors and a few dozen AOS aircraft, and including an underlying point defense such as I described earlier, a system might cost about $10 billion. Since this system would negate many times that amount's worth of offensive missiles, it would be a real bargain for the United States and its allies. However, it could not be deployed in accordance with the ABM Treaty.

III. Global Defenses

THE UNITED STATES first considered global ballistic-missile defenses in 1958 and designed a Ballistic Missile Boost Intercepts system (BAMBI). The idea was good, but we didn't have the technology. But the technology has advanced in the last thirty years. We now have in hand the means to make an effective global defense feasible at the same time as sovereign-area and point defenses--namely, within a decade. In a spectacular experiment on September 5, 1986, years ahead of the schedule set in 1983 at the start of SDI, we launched two small rocket-powered satellites into space atop a Delta rocket. Each satellite had its own infrared sensor that collected data as the other maneuvered with its rocket engine. From these data we can design the best sensor system for a space-based missile defense. In a final success, one satellite homed in and collided with the other satellite, which was maneuvering at the time.

We could use an improved version of our current early-warning satellites deep in space as the initial global-defense sensors. Such satellites see the intense heat of an ICBM booster as it lifts into space and can track it and send the information to the battle-management system. In the mid 1990s we could have ready an improved type of infrared sensor, which could see not only the hot booster flame but also the much cooler warheads, even after they had been released from their bus. If we can identify, track, and intercept both warheads and boosters, the same number of defenders in orbit could stop a much larger number of attacking warheads.

SDI critics like to point to the complex computer programs needed for battle management as the Achilles' heel of a global-defense system. Yet the experiment carried out so successfully on September 5 used a battle-management computer system with more than one million lines of new code--and the program worked perfectly the first time. SDI is currently building a "national test bed" in Colorado Springs, Colorado, to perfect battle-management methods for global defenses. This facility will open in 1989. Global defenses in the 1990s do not require exotic new computer methods, but only the proper application of existing knowledge.

The September experiment also proved that there are no show-stopping technical obstacles in the way of an effective space-baed intercept system. The experiment showed us how to build a defender missile that is small and therefore affordable. We also learned how to pick out the booster target itself from the larger background of the booster exhaust. (If we could not distinguish the booster from its rocket exhaust, then a defending missile could very well miss the booster target.) Each "booster buster" would weigh as little as a few hundred pounds and would reside, along with three or four of its fellows, on a small satellite carrier. Upon receiving notice of a hostile launch, the little defending missiles would shoot out of their carrier with instructions on where to look for the targets. They would then fire small rocket motors to home in on and crash into the enemy boosters.

Although we have less experience with global than with other types of defenses, experts are becoming convinced that a space-based system could be in orbit with several thousand defending missiles by 1995. We cannot calculate system costs very accurately right now, but we have a rule of thumb for fairly good estimates. We know that the cost of a complicated space-based system goes up pretty much in proportion to what it weighs. Every pound of electronic circuits, sensors, and gyroscopes costs about as much as any other pound. All the way from little satellites to the Space Shuttle Orbiter, it costs about $10,000 per pound to manufacture and launch objects into space. If each defender missile weighed a thousand pounds--including the satellite that carried it and associated sensors--then two thousand of these defending missiles could be put in orbit, ready to protect the United States and its allies, for roughly $20 billion.

Space-baed global defenses are the key to a successful defense. Since these defenses are global, they would confront an aggressor's missiles worldwide from the moment they were launched, regardless of whether the missiles rose from the sea or from the land, from the Soviet Union or from its allies. With sovereign defenses, point or area, the aggressor can concentrate his forces to break through at the most vulnerable point. But global defenses would be everywhere and "self-healing," in the sense that a hole in the defense quickly fills as other defensive satellites orbit into place; hence, the aggressor cannot easily plan, time, and focus his offensive-missile launches to overwhelm the defense. For this reason global defenses have been publicenemy number one to the Soviets. Global defenses, even in small numbers, complicate an aggressor's plans to the point where he must abandon any hope of delivering a knockout blow to his intended victim.

IV. Countermeasures 

THE KIND OF DEFENSE we could have in the mid 1990s, with a few thousand space-based missiles and a few thousand more ground-based sovereign-area-deense interceptors, would so disrupt a Soviet first strike as to make it impossible to defeat our armed forces. In the years following an initial deployment, the United States could bring the multi-layered system to full strength--capable of stopping almost all Soviet missiles and making obsolete the Soviets' trillion-dollar offenive investment. The price tag cannot be known with certainty at this time, but it is very likely less, perhaps much less, than $100 billion--comparable to a typical new offensive strategic system at today's prices. The trillion-dollar costs that the Union of Concerned Scientists and various Soviet "scientists" like to mention are total fabrications.

The reader will note the absence of WHAP, ZAP, HISS. One should not therefore conclude that the money spent on laers and particle beams is wasted. On the contrary, these are the technologies that will close the ring around offensive missiles. We must worry over the longer run--the the early decades of the twenty-first century--that the Soviets might try to confuse our first round of defending weapons with cheap decoys (i.e., lightweight imitation warheads), or develop new booster rockets (that burn out while still inside the atmosphere, thereby confounding space-based defending missiles that don't work well inside the atmosphere), or find ways to shoot down or damage the space-based defenses.

The lasers and particle beams are needed to counter that twenty-first-century Soviet threat. They won't be ready in the 1990s, but we can demonstrate that they are feasible. For example, we are building a "neutral-particle beam"--a device that shoots atoms at a target in space. If the beam is powerful enough, these atoms can fry a warhead or missile's electronic brain or melt its guts. Less powerful beams can probe an object and tell us whether it is a decoy or a real warhead. We plan to launch a neutral-particle-beam experiment into space in 1991. to prove that we can tell decoys from real warheads. That should deter the Soviets from wasting lots of money on decoys. At the same time we've planned to build a mammoth free-electron laser in the New Mexico desert. Caroming it off a mirror in space, this laser could send a beam halfway around the world in a cosmic two-bank pool shot to reach into the upper parts of the atmosphere and stop even a fast-burn booster. Once the Soviets see that experiment demonstrated, they won't even try the costly countermeasure of the fast-burn booster.

Other vital SDI research will show us how to keep the Soviets from blasting us out of space as a prelude to a first strike. SDI critics are fond of saying that we can't make space systems survivable, but they are wrong. SDI research has already pointed the way. Just as we make military airplanes survivable today--armor, maneuver, self-defense (shoot back at the bad guys), and decoys (yes, the defense is allowed to use them too)--we can make our satellites sufficiently survivable to do the job tomorrow. For example, if the Soviets can't pick out the real defending missile satellite because it is surrounded by decoys, their ability to destroy our defense is much less. If our satellites can further dodge an attacker and even survive some attacks, it is even less possible for the Soviets to negate our defense. Finally, if the Soviets know that our satellites are defended by the very missiles we would use to stop their ballistic missiles, they would also know it is futile to attack the defense even if they could figure out which objects are real targets.

One final point about countermeasures. None of them is cheap. For example, experts estimate that a fast-burn booster might cost more than $100 million per warhead. If the Soviets chose to replace their entire arsenal with fast-burn boosters, it would take them twenty years and cost a trillion dollars. after all was said and done, an upgraded defense using lasers could still shoot them down, and at a cost of much less than a trillion dollars. These hard fiscal and technical realities would prevent the Soviets from even attempting the countermeasures critics put forward.

THERE YOU have it. All the signals indicate that we can build in the 1990s a defensive system that will do what the President asked our scientists and engineers to do. We can begin sustained testing of components of this system--thereby demonstrating their feasibility to Congress and the public--over the next two years. Our ability to do this has come despite intense domestic and Soviet opposition. Moreover, our advanced research is proving not only that an initial system will work, but that we will have the technology to counter any new tricks the Soviets may try to spring. Once this is clear, the wisdom of the President's proposal to reduce to an absolute minimum the number of ballistic missiles in both superpowers' arsenals should be perceived by the Soviets.

Our scientists and engineers are giving us the ability to return our national survival to our own hands. The decision to proceed from here is dictated by politics, not technology. The Soviet Union knows this. We must never forget it either.