The Anti-Ballistic Missile Defense System after September 11

President George W. Bush�s proposal that we build an ABM system may become one of the casualties of the terrorist attacks of September 11�and perhaps, in the long run, the most serious casualty. It was under attack by all the bien pensant elements of the Western world even before the destruction of the World Trade Center and part of the Pentagon, but the fact that the terrorists attacked us this time with our own aircraft is, in the minds of many, the final proof that the proposal is utterly misdirected. They are wrong; a charitable verdict would be that they are simply confused. Their confusions are both technical and political, with the former the more easily cleared up, and the latter the more important and delicate. First, the technical issues:


Technical Issue I: �It won�t work! In particular, the software can�t be debugged!�


One of the principal claims made by opponents of the ABM proposal is that the system can�t be made to work. They cite in support of that claim the record compiled by the ABM tests conducted so far�a decidedly mediocre though improving record. To this they add the claim that an enemy technically advanced enough to mount a missile attack on us would also be advanced enough to incorporate many kinds of dummy warheads and decoys into that attack, forcing us to try to shoot down an impossible number of objects in an attempt to destroy all the real warheads. And even though the tests run so far have shown, at least three times now, that an ABM can intercept a target missile following a trajectory of the kind that would be taken by an actual intercontinental missile, we are still told that the ABM system is attempting the supposedly quixotic feat of �hitting a bullet with a bullet.�


To this claim there are several counter-arguments:


  1. The ABM tests to date have been conducted under such artificial constraints, and with such technical handicaps imposed by treaties and other political considerations, that the wonder is that they have been as successful as they have. The kind of ABM system that the United States could deploy, freed of such constraints and allowed to use all its technological muscle, would be many times better than the crippled systems displayed so far. That even the rudimentary system so far tested has been able several times to destroy its target is a most promising indication.

  2. It is by no means obvious that a nation or group possessing ballistic missiles would also have dummies and decoys. These deceptive devices are far from simple to design, construct, test, and deploy; they are in some ways more sophisticated than the actual warheads that they will attempt to mimic�lies are always trickier than the truth. They are also expensive. Insofar as the proposed ABM system is simply a defense against �rogue states� such as North Korea, Iraq, or Libya, it may well be able to do its job without worrying about advanced forms of deception.

  3. �Hitting a bullet with a bullet� is good sloganeering, but nothing more. Its effectiveness lies in its ability to implant a false picture in our minds: it almost irresistibly suggests a human marksman attempting an utterly impossible feat. But the feat of hitting a bullet with a bullet�already successfully done several times, with far from optimum systems�is in fact not particularly difficult for a computer-based system. A bullet or other purely ballistic missile is one that follows a predetermined course, a course that can be predicted with great precision from just a few measurements. And the bullet that is fired at that bullet�the ABM�is another such purely ballistic missile until its final moments, when it has freedom to make any last-minute adjustments in its course that may be necessary if it is to hit its target. This is exactly the kind of work that computers are good at; predicting the course that an enemy missile will take, and determining the course that an ABM must follow in order to intercept it, is the computer�s meat and drink. In fact, every time we launch an artificial satellite into its prescribed orbit, or send astronauts up to the ISS, or put a capsule on the surface of Mars or any other celestial body, we are trying to �hit a bullet with a bullet��and we succeed so routinely that our newspapers put the stories back on page 7. << mention Bradley Graham, Hit to Kill: The New Battle Over Shielding America From Missile Attack . NY: Public Affairs, 2001 . He deals with the capability of the planned ABM system to protect against Russian and Chinese attacks (�accidental or unauthorized launch�=AUL ) on pages 211-214 and 315 >>

    The kind of computation ABM computers will have to do is of a very well understood kind. They will be called on to process inputs from sensors such as radar and infra-red detectors, deriving from that data the positions and momenta of objects either in powered trajectories or in free fall. This processing will predict where those objects will be at later times, and where other such objects, or beams of energy or particles, must be directed in order to intercept them. This is a kind of processing in which we have a great deal of experience, and which requires no scientific breakthroughs or even much new software. There are indeed some stringent requirements that ABM computers must meet, but they do not involve the generation of much new code; the big problems are, first, speed�these computations will have to be performed faster, perhaps orders of magnitude faster, than the relatively leisurely computations required by a mission to the Moon, for example�and second, radiation resistance. Insofar as the computers themselves are located outside the earth�s atmosphere, they will be subject to ionizing radiation that would be destructive to devices based on ordinary, personal-computer grade components. These two requirements are imposed on the hardware, however�a technology where we are making rapid progress, and may have made enough progress already�and not on the software, where masses of fresh code inevitably means a swarm of new bugs.

  4. The bullet-hitting-a-bullet issue is beside the point in another way, too; much ABM research and development is being devoted not to hit-to-kill weapons that would have to intercept an enemy missile or warhead directly, and in full stride, but to ABM weapons that would not have to meet even those requirements. These alternative ABM weapons would attack enemy missiles either when they were much more vulnerable�during their lift-off phase, when they would be moving relatively slowly, and still over enemy territory�or with a laser beam, which flies straight and at the speed of light�or with weapons that did not trouble to hit enemy objects one at a time, nor to distinguish between real warhead and decoys, but that would simply explode with such force as to destroy any targets, real or decoy, within a large radius.

It should be observed that the technical problems posed by an ABM system are of the kind that the United States, in particular, has always been good at. Large-scale, highly ambitious engineering projects are our specialty: think of the Panama Canal, the Manhattan Project, the Polaris Submarine-Launched Missile project, the Apollo Lunar Landing, the International Space Station, the Hubble Telescope�why should it be doubted that the country that succeeded in all of these should fail with the development of an ABM system?

The Technical Issues, II: �It can�t be tested!�


The second great argument against the proposed system is that it cannot ever be tested realistically. Perfectly true, and completely misleading; it is neither possible nor necessary to test that system in a full, realistic way; it may not even be desirable.


It is not possible because it is simply not in the nature of deterrent systems to be fully testable, in particular when they threaten dire retaliation in the event of such extreme acts as ballistic-missile attacks; nothing short of provoking a potential enemy to mount a missile attack on us would give us the opportunity to test the efficacy of our defensive measures. (For some reason, we never hear the objection of non-testability made against other war-prevention schemes; against disarmament, for example, although of all such schemes it is by far the least testable, as well as the one most in need of being tested.)


It is not necessary because if a deterrent is powerful enough to repel even just a significant part of an attack, and has even a small probability of working, it will be effective against rational adversaries (irrational ones will be discussed later). If a would-be aggressor had to face the possibility that only some fraction of his attack would be successful, and that there was some real chance that we would survive to counterattack, he would be faced with a degree of uncertainty that would be too daunting to overcome. (The system the ABM proposal would replace, Mutual Assured Destruction, cannot be tested either, but the opponents of ABM ignore this. Their excuse is that MAD may be considered to have proved itself in view of the fact that there has been no nuclear attack on the United States since the inception of that system; a remarkably starry-eyed inference. This line of argument is dealt with below, as one of the political objections.)


It may not, however paradoxical it sounds, even be desirable that an ABM system be thoroughly testable. If it could be so tested, its limits would thereby become known, to us and to potential enemies alike. And when those limits became known, they could be the subject of rational analysis by those potential enemies�which means that those potential enemies could consider ways to defeat our system, and at least be able to estimate their chances of success accurately. But if the system�s exact efficacy is unknown, such rational analysis becomes much harder or impossible; with so many unknowns in the equation, the question of whether to attack the United States turns into a largely emotional one, and in the war councils of the potential enemy, the hawks and the doves will be screaming at one another. And while they scream, we are safe. It is not a bad idea for a fastball pitcher to be just a little wild sometimes, and occasionally put a pitch somewhere near a batter�s head.


And an untested ABM yields another benefit, one particularly reassuring to liberals and progressives. People of this description used to worry, and perhaps still do, about the possibility that some faction of superhawks in the Pentagon, feeling secure behind the protection of an ABM system, might launch an unauthorized missile attack on some country that they thought posed a threat to us. If our own missile shield were of unknown quality, such superhawks would not feel as safe in taking such action as they would if it were known to be perfect (or even just good enough to deal with a counter-attack from the target country). The ideal situation would be one in which we had an ABM shield that was perfect, but not known by anyone to be so.


Political issue 1: We can be attacked in so many ways, why worry about missiles?


Another major objection to the proposed ABM system: the United States is vulnerable to many different kinds of attack, of which a missile attack is only one, and probably not the most likely�so why spend so much money building a system that will leave us still vulnerable? This is an argument that is so obviously fallacious that its opponents are sometimes disarmed by an inability to believe that it can be seriously maintained.


Granted that a missile attack is only one kind of attack that might be made on us, why should we not defend ourselves against it if we can? Do we neglect to lock our front door, on the grounds that intruders might enter through the windows? Do we refuse to develop treatments for cancer, on the grounds that we�re subject to other fatal diseases as well? When we have locked our front door, we can concentrate on barring the windows; when we have cured cancer, we can turn our attention to cardiovascular problems; if we can successfully defend ourselves against missile attack, that will free us to deal with other potential attacks. The notion that measures that do not solve all our problems are not worth taking is simply ludicrous; a search for a panacea would be the worst possible obstacle to medical progress.


But this ridiculous argument is heard much more frequently now that the terrorists of September 11 have struck in the way they did; we are now asked to believe, in effect, that terrorists are constrained to attack us in only one way, and since they have now indicated that their choice is the use of hijacked commercial aircraft as kamikaze vehicles, it is foolish for us to protect ourselves against ballistic missile attack. The truth is the opposite of this: the attacks of September 11 have alerted us to attacks based on the infiltration of terrorists into our country, and on the use of our own aircraft as weapons, with the result that we are taking extensive precautions against that class of attacks. So it will be a great deal harder in future for terrorists to attack us in those ways, and correspondingly more attractive for them to attack us in other ways�including via missiles. The attacks of September 11 make it more likely, not less, that someday we will be attacked by missiles.


Political issue 2: It would nullify treaties that have kept us safe for the last thirty years!


A principal political objection to the proposed ABM system is that it would cause us to violate, or to withdraw entirely from, some treaties that govern the military use of space and the deployment of ABM systems. As with so many of the anti-ABM arguments, these bare facts are correct; but the argument itself does not spring from those facts; it springs instead from a view of the world that an overwhelming majority of us would reject if it were explicitly presented to us. It is true that during the time that the treaties in question have been in force, we have not suffered a missile attack, but those who attribute our safety to those treaties are falling victim to the classic fallacy of post hoc ergo propter hoc . To see the emptiness of their argument one need only translate the high-minded views that it rests on into concrete, humanly understandable terms.


Those who would forgo an ABM system in order to preserve the treaties in question (principally the 1967 Outer Space Treaty and the 1972 Anti-Ballistic Missile Treaty ) are asking us to believe that a potential aggressor, ready in all other respects to attack us with missiles carrying nuclear or other Weapons-of-Mass-Destruction (WMD) warheads, would refrain from carrying out that attack because he, or some predecessor of his, had signed a piece of paper. Reliance on the forbearance and respect for treaties of an Iraqi or a North Korean leader, we are asked to believe, is more realistic than reliance on a physical shield that makes no assumptions about the likely actions of despots who hate us. If this is how ABM opponents conceive of the mind of aggressors who have no compunctions about killing tens of millions of people�including not just Americans but very likely millions of their own people as well�then they are not living in the world that most of us inhabit.


Political issue 3: It would set off an arms race!


The next great objection: our abandonment of the various ABM-limiting treaties to which we are more or less party would cause��force� is the word commonly used�various nations, particularly Russia and China, to start an arms race in an attempt to regain their ability to threaten us with a missile attack or something equivalent. (This point is often made in a tone of voice that implies that other countries somehow have a right to be able to threaten us, and that we are infringing on that right by rendering their missile forces impotent.) It is even sometimes suggested�how seriously it�s hard to tell�that an ABM shield is actually an offensive measure, since it would enable us to attack another country with impunity. In the same way, presumably, buying a fire extinguisher is to be thought of as an incendiary step, since it may encourage you to be more careless with fire.


But there is no reason why we should want Russia, China, or any other country to be in a position to threaten us with weapons of mass destruction, whether missiles or any other. If the political leaders of those countries cannot maintain their positions unless they can show their generals that they have military parity with the United States, that is their problem, not ours. It�s unfortunate that some of our own political leaders are so afraid of �instability� that they feel they have to support all existing foreign regimes, even if that involves letting those regimes threaten this country with destruction. It is a constant complaint of liberals and progressives that our government is always supporting dictators and �strong men� for fear of the turmoil that might ensue if they were to fall, but many of the same people are perfectly willing to let Russia and China, for example, have the ability to destroy us if that�s what it takes to keep their present leaders�always supposedly reasonable men engaged in struggles with unreasonable rivals�in power.


And even if those nations should feel aggrieved at being denied the ability to destroy us, what can they do about it? Can Russia, or China, or any other nation in the world, build a system capable of defeating the kind of ABM system we can build, taking that system in combination with all our other forces? The technical and economic superiority of the United States is such that no country, or plausible combination of countries, can hope to build such a system. They would ruin themselves in trying�not only would they fail, but the attempt would impoverish their own populations to such a degree that they would probably face revolt. (Perhaps we should hope that such countries do try to defeat our ABM system�a revolt by their people may be just what we, and they, need).


Political issue 4: Irrational Opponents


So far we have considered only rational opponents, by which I mean those who want to survive both as individuals and as nations if at all possible, and who calculate probabilities much as we do. But of course we cannot count on our opponents to be rational, and in fact those potential enemies we see in the world today are notoriously irrational�you just can�t get high-class enemies these days.


And it is in dealing with just such opponents that an ABM system really comes into its own. If we face opponents who are irrational in the sense just defined, then clearly MAD or any other deterrent system that depends on the rationality of all parties is useless; against mad bombers, the only protection is a physical shield�precisely what an ABM system offers. And an attack mounted by the irrational is likely to be of just the kind against which such a system would be successful: it would be a relatively small-size attack, and one not involving very sophisticated spoofing devices.


Political issue 5: The Real Targets of an ABM System


The chief reason why so much of the commentary on the ABM proposal, pro and con both, misses the mark is that it rests on a falsehood�a falsehood about the real enemies toward whom it is addressed: Russia and China. These countries are our (passive) enemies; they are not prepared to go to war with us, but if they could destroy us by pressing a button, they would do so gladly. But it is our official position that both are aspirants, at least, to the status of respected, peace-loving members of the world community, and that we have nothing to fear from them (except, perhaps, that �rogue elements� of the Russian military might launch missiles at us without official authorization). This diplomatic fiction may be necessary, but it also necessarily vitiates any discussion of an ABM system by anyone who has to maintain it. Forced to pretend that we are concerned only with one-shot, insane attacks from the likes of North Korea, the Administration is stuck with relatively weak arguments for its case.


It is not clear whether the domestic opponents of the ABM proposal are genuinely deceived by the pretence that it is just the rogue states against which the system is intended, or are gleefully pretending to believe it so as not to have to counter the real reasons for an ABM system. The opposition to the ABM proposal by Russia and China, on the other hand, is not just their usual opposition to anything we propose, but a rational response to a real threat: the threat they may be deprived of their ultimate bargaining chip. It is a curious sidelight on the thinking of so many American liberals and progressives that they should feel no qualms at finding themselves on the same side of this issue as these two nations, one burning with resentment at its loss of superpower status, the other burning with resentment at its failure to have secured so far full recognition as a superpower.


No doubt missile attacks by rogue states, no matter how insane such things may seem to us, are possible, and since they are possible, they must be guarded against. But if such attacks were the only sort against which the ABM system were to guard, it would indeed be arguable that the money it would cost might be better spent on other defensive measures. If our debate over the proposed system is to have any relevance to reality, therefore, it must be recognized at least by the debaters, even if it cannot be acknowledged by our government, that the enemies against whom it is really directed are the usual suspects; they are the ones who have both the means and the motive to attack us if they can, and who can hope to do so thorough a job of it that we will not be able to retaliate. They have, as I say, the means and the motive; the ABM system is designed to deny them the opportunity.


Con clusions


It is commonly urged that there are many dimensions to the ABM proposal, and many consequences that cannot now be foreseen. But these caveats apply to any possible handling of the problem that proposal is meant to deal with, including doing nothing whatever, or relying on diplomacy and treaties (conceived as an alternative to weaponry, offensive or defensive), or disarmament, or anything else so far conceived. We have to do something , and that something must be determined by the best information we have at our disposal. I submit that the best information we have today supports the present Administration�s ABM proposal. To summarize my reasons:


  1. There is every reason to expect the technology will work. The ABM project is of exactly the sort that we have succeeded with in the past. No scientific breakthroughs are required, and no large amount of new software need be written.

  2. It cannot be realistically and thoroughly tested�which is to say, it is in this respect just like all other proposals for defending ourselves against missile or any other kind of WMD attack. But such testing is also unnecessary, and in a way even undesirable.

  3. It does not guard against all possible attacks, maybe not even against the most likely attack�but it does guard against at least one very dangerous kind of attack, and its success in doing so would free our assets and attention to deal with other kinds.

  4. It would not set off an arms race, if only because none of our potential enemies, either alone or in combination, could afford to engage us in such a race. If they were to devote to such a race anything like that fraction of their GNP that it would take even to get started, they would be in danger of revolt by their own populations.

  5. We may well find ourselves dealing with enemies who are irrational, and who cannot be deterred by any threat we can mount, much less by treaties and other such �civilized� measures. Against such enemies, only a physical shield will help�and that is exactly what is promised by the ABM proposal.

  6. Finally, we had better understand that a major purpose of the ABM system is to guard against missile attack (and threat of such attack) by either or both of our principal national enemies, Russia and China. Until that is acknowledged, the debate will continue to be fruitless and even harmful.